[Diffusion] Register runtime cache for direct DLO H2D

Avoid recurrent mmap-to-pinned packing in no-AllGather DLO by registering transform-complete runtime-cache mappings under an explicit per-worker budget. Copy mapped tensors directly into the rotating device buffers while preserving bounded-staging fallback and ordered shutdown cleanup. Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
[Diffusion] Document runtime cache hardware tradeoffs
2026-08-17 11:50:32 +08:00 · 2026-08-17 00:43:06 +08:00 · 2026-08-16 23:26:25 +08:00 · 2026-08-16 23:15:48 +08:00
247 changed files with 3677 additions and 20476 deletions
@@ -159,7 +159,19 @@ steps:
          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
          - export VLLM_USE_DEEP_GEMM="0"
          - export VLLM_MOE_USE_DEEP_GEMM="0"
-          - pytest -s -v tests/e2e/ -m "full_model and L4 and tts" --run-level "full_model" --ignore=tests/e2e/accuracy
+          - pytest -s -v tests/e2e/ -m "full_model and L4 and tts" --run-level "full_model" --ignore=tests/e2e/accuracy --ignore=tests/e2e/online_serving/test_voxcpm2_tts_expansion.py
+        mirror_hardwares: l4_1
+
+      - label: ":full_moon: TTS · Accuracy Test · VoxCPM2"
+        timeout_in_minutes: 30
+        commands:
+          - |
+            timeout 30m bash -c "
+              export VLLM_LOGGING_LEVEL=DEBUG
+              export VLLM_WORKER_MULTIPROC_METHOD=spawn
+              export VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
+              pytest -s -v tests/e2e/online_serving/test_voxcpm2_tts_expansion.py -m 'full_model and cuda and tts' --run-level 'full_model'
+            "
        mirror_hardwares: l4_1

      - label: ":full_moon: TTS · Perf Test"
@@ -176,6 +188,35 @@ steps:
            exit $$EXIT
        mirror_hardwares: h100_2

+      - label: ":full_moon: TTS · VoxCPM2 · Perf Test"
+        key: nightly-tts-performance-voxcpm2
+        timeout_in_minutes: 180
+        commands:
+          - export BENCHMARK_DIR=tests/dfx/perf/results
+          - export VLLM_ALLOW_LONG_MAX_MODEL_LEN="1"
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_voxcpm2.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
+            exit $$EXIT
+        mirror_hardwares: h100_1
+
+      - label: ":full_moon: TTS · Higgs-Audio-V3 · Perf Test"
+        key: nightly-tts-performance-higgs-audio-v3
+        timeout_in_minutes: 180
+        commands:
+          - export BENCHMARK_DIR=tests/dfx/perf/results
+          - export VLLM_USE_DEEP_GEMM="0"
+          - export VLLM_MOE_USE_DEEP_GEMM="0"
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_higgs_audio_v3.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
+            exit $$EXIT
+        mirror_hardwares: h100_1
+
  - group: ":card_index_dividers: Tiny Model Tests [Multi-GPU]"
    key: nightly-tiny-model-test-group
    depends_on: upload-nightly-pipeline
@@ -205,7 +246,8 @@ steps:
      build.pull_request.labels includes "nightly-test" ||
      build.pull_request.labels includes "diffusion-x2iat-test"
    steps:
-      # Use explicit file shards so single-card cases do not reserve 2+ GPUs.
+      # Use explicit file shards so single-card cases do not reserve 2+ GPUs,
+      # and Bagel feature coverage is not duplicated via a broad tests/e2e sweep.

      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Single-GPU"
        timeout_in_minutes: 120
@@ -213,6 +255,15 @@ steps:
          - >-
            pytest -sv
            tests/e2e/online_serving/test_qwen_image_expansion.py
+            tests/e2e/online_serving/test_qwen_image_edit_expansion.py
+            tests/e2e/online_serving/test_qwen_image_layered_expansion.py
+            tests/e2e/online_serving/test_boogu_image.py
+            tests/e2e/online_serving/test_boogu_image_edit.py
+            tests/e2e/online_serving/test_boogu_image_expansion.py
+            tests/e2e/online_serving/test_boogu_image_edit_expansion.py
+            tests/e2e/online_serving/test_bagel_expansion.py
+            tests/e2e/online_serving/test_krea2_expansion.py
+            tests/e2e/offline_inference/test_krea2_expansion.py
            -m "full_model and diffusion and H100 and not distributed_cuda"
            --run-level "full_model"
        mirror_hardwares: h100_1
@@ -223,14 +274,47 @@ steps:
          - >-
            pytest -sv
            tests/e2e/online_serving/test_qwen_image_expansion.py
+            tests/e2e/online_serving/test_qwen_image_edit_expansion.py
+            tests/e2e/offline_inference/test_krea2_expansion.py
            -m "full_model and diffusion and H100 and distributed_cuda"
            --run-level "full_model"
        mirror_hardwares: h100_2

+      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Multi-GPU Layered/LongCat/FLUX"
+        timeout_in_minutes: 90
+        commands:
+          - >-
+            pytest -sv
+            tests/e2e/online_serving/test_qwen_image_layered_expansion.py
+            -m "full_model and diffusion and H100 and distributed_cuda"
+            --run-level "full_model"
+        mirror_hardwares: h100_2
+
+      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with H100 · Multi-GPU Bagel"
+        timeout_in_minutes: 120
+        commands:
+          - >-
+            pytest -sv
+            tests/e2e/online_serving/test_bagel_expansion.py
+            -m "full_model and diffusion and H100 and distributed_cuda"
+            --run-level "full_model"
+        mirror_hardwares: h100_2
+
+      - label: ":full_moon: Diffusion X2I(&A&T) · GLM-Image Function Test with H100"
+        timeout_in_minutes: 120
+        commands:
+          - >-
+            pytest -sv
+            tests/e2e/online_serving/test_glm_image_expansion.py
+            tests/e2e/offline_inference/test_glm_image_autoround_w4a16_expansion.py
+            -m "full_model and diffusion and H100"
+            --run-level "full_model"
+        mirror_hardwares: h100_2
+
      - label: ":full_moon: Diffusion X2I(&A&T) · Function Test with L4"
        timeout_in_minutes: 120
        commands:
-          - pytest -sv tests/e2e/ -k "not test_wan and not hunyuan" -m "full_model and diffusion and L4" --run-level "full_model" --ignore=tests/e2e/accuracy
+          - pytest -sv tests/e2e/ -k "not test_wan and not test_bagel_expansion and not hunyuan" -m "full_model and diffusion and L4" --run-level "full_model" --ignore=tests/e2e/accuracy
        mirror_hardwares: l4_4

      - label: ":full_moon: Diffusion X2I(&A&T) · Doc Test"
@@ -243,7 +327,7 @@ steps:
        timeout_in_minutes: 180
        commands:
          - export VLLM_HTTP_TIMEOUT_KEEP_ALIVE=120
-          - pytest -s -v tests/e2e/accuracy/test_qwen_image.py --run-level full_model
+          - pytest -s -v tests/e2e/accuracy/test_qwen_image*.py --run-level full_model
          - pytest -s -v tests/e2e/accuracy/test_diffusers_backend_similarity.py -k '2i_matches_diffusers' -m full_model --run-level full_model
        mirror_hardwares: h100_1

@@ -271,6 +355,70 @@ steps:
            exit $$EXIT
        mirror_hardwares: h100_4

+      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit"
+        key: nightly-diffusion-x2iat-performance-qwen-image-edit
+        timeout_in_minutes: 180
+        commands:
+          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
+          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
+          - export CACHE_DIT_VERSION=1.5.0
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_vllm_omni.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
+            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
+            exit $$EXIT
+        mirror_hardwares: h100_4
+
+      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit-2511"
+        key: nightly-diffusion-x2iat-performance-qwen-image-edit-2511
+        timeout_in_minutes: 180
+        commands:
+          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
+          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
+          - export CACHE_DIT_VERSION=1.5.0
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_2511_vllm_omni.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
+            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
+            exit $$EXIT
+        mirror_hardwares: h100_4
+
+      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Layered"
+        key: nightly-diffusion-x2iat-performance-qwen-image-layered
+        timeout_in_minutes: 120
+        commands:
+          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
+          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
+          - export CACHE_DIT_VERSION=1.5.0
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_qwen_image_layered_vllm_omni.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
+            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
+            exit $$EXIT
+        mirror_hardwares: h100_1
+
+      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · BAGEL"
+        key: nightly-diffusion-x2iat-performance-bagel
+        timeout_in_minutes: 180
+        commands:
+          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
+          - export DIFFUSION_ATTENTION_BACKEND=FLASH_ATTN
+          - export CACHE_DIT_VERSION=1.5.0
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py --test-config-file tests/dfx/perf/tests/test_bagel_vllm_omni.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
+            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
+            exit $$EXIT
+        mirror_hardwares: h100_1
+
  - group: ":card_index_dividers: Diffusion X2V Model Test"
    key: nightly-diffusion-x2v-group
    depends_on: upload-nightly-pipeline
@@ -292,6 +440,12 @@ steps:
          - pytest -s -v tests/e2e/online_serving/test_wan22_expansion.py -k "i2v" -m "full_model and cuda" --run-level "full_model"
        mirror_hardwares: h100_2

+      - label: ":full_moon: Diffusion X2V · LingBot Video MoE Function Test"
+        timeout_in_minutes: 90
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_lingbot_video_moe.py -k "video_generation" -m "full_model and cuda" --run-level "full_model"
+        mirror_hardwares: h100_1
+
      - label: ":full_moon: Diffusion X2V · Doc Test"
        timeout_in_minutes: 60
        commands:
@@ -302,8 +456,15 @@ steps:
        timeout_in_minutes: 90
        commands:
          - pytest -s -v tests/e2e/offline_inference/test_wan22_autoround_w4a16_expansion.py -m "full_model and cuda" --run-level "full_model"
+          - pytest -s -v tests/e2e/online_serving/test_wan_2_1_vace_expansion.py tests/e2e/online_serving/test_ltx.py -m "full_model and cuda" --run-level "full_model"
        mirror_hardwares: h100_2

+      - label: ":full_moon: Diffusion X2V · LTX-2.5 Online Test"
+        timeout_in_minutes: 60
+        commands:
+          - pytest -s -v tests/e2e/online_serving/test_ltx25.py -m "full_model and cuda" --run-level "full_model"
+        mirror_hardwares: h100_1
+
      - label: ":full_moon: Diffusion X2V · Wan2.2 Accuracy Test"
        timeout_in_minutes: 180
        commands:
@@ -322,8 +483,23 @@ steps:
        commands:
          - pytest -s -v tests/e2e/accuracy/hunyuanvideo15_t2v/test_hunyuanvideo15_t2v_video_similarity.py -m full_model --run-level full_model
          - pytest -s -v tests/e2e/accuracy/hunyuanvideo15_i2v/test_hunyuanvideo15_i2v_video_similarity.py -m full_model --run-level full_model
+          - pytest -s -v tests/e2e/accuracy/test_video_streaming_output_similarity.py -m full_model --run-level full_model
        mirror_hardwares: h100_2

+      - label: ":full_moon: Diffusion X2V · LTX Accuracy Test"
+        timeout_in_minutes: 180
+        commands:
+          - pytest -s -v 'tests/e2e/accuracy/ltx/test_ltx_official_similarity.py::test_ltx_matches_official[ltx2_3_i2v]' -m full_model --run-level full_model
+          - pytest -s -v 'tests/e2e/accuracy/ltx/test_ltx_official_similarity.py::test_ltx_matches_official[ltx23_two_stage_layer_fused_i2v]' -m full_model --run-level full_model
+        mirror_hardwares: h100_1
+
+      - label: ":full_moon: Diffusion X2V · LTX-2.5 Official Parity Test"
+        timeout_in_minutes: 180
+        commands:
+          - pytest -s -v tests/e2e/accuracy/ltx/test_ltx25_official_similarity.py::test_ltx25_distilled_two_stage_matches_official -m full_model --run-level full_model
+          - pytest -s -v tests/e2e/accuracy/ltx/test_ltx25_official_similarity.py::test_ltx25_full_matches_official -m full_model --run-level full_model
+        mirror_hardwares: h100_1
+
      - label: ":full_moon: Diffusion X2V · Perf Test"
        key: nightly-diffusion-x2v-performance
        timeout_in_minutes: 180
@@ -399,6 +575,16 @@ steps:
          - pytest -sv tests/diffusion/quantization -m 'full_model and cuda and L4' --run-level "full_model"
        mirror_hardwares: l4_1

+  - label: ":bar_chart: Testcase Statistics"
+    key: nightly-testcase-statistics
+    timeout_in_minutes: 120
+    depends_on: upload-nightly-pipeline
+    if: build.env("NIGHTLY") == "1"
+    commands:
+      - python tools/nightly/buildkite_testcase_statistics.py -o tests/dfx/perf/results/buildkite_testcase_statistics.html
+      - buildkite-agent artifact upload "tests/dfx/perf/results/*.html"
+    mirror_hardwares: h100_1
+
  # NOTE: ":email: Nightly Collection & Email" is deprecated and will be removed soon.
  # Do not add new depends_on entries or extend this step.
  - label: ":email: Nightly Collection & Email"
@@ -409,20 +595,32 @@ steps:
      - nightly-omni-performance-vllm-text
      - nightly-omni-performance-multi-replicas
      - nightly-tts-performance
+      - nightly-tts-performance-higgs-audio-v3
      - nightly-diffusion-x2iat-performance-qwen-image
+      - nightly-diffusion-x2iat-performance-qwen-image-edit
+      - nightly-diffusion-x2iat-performance-qwen-image-edit-2511
+      - nightly-diffusion-x2iat-performance-qwen-image-layered
+      - nightly-diffusion-x2iat-performance-bagel
      - nightly-diffusion-x2v-performance
+      - nightly-testcase-statistics
    if: build.env("NIGHTLY") == "1" && build.env("EMAIL_DISTRIBUTION") == "1"
    commands:
      - pip install openpyxl
      - export DEFAULT_INPUT_DIR=tests/dfx/perf/results
      - export DEFAULT_OUTPUT_DIR=tests/dfx/perf/results
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-tts-performance
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-tts-performance-higgs-audio-v3
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-no-async-chunk
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-async-chunk
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-vllm-text
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-omni-performance-multi-replicas
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-edit
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-edit-2511
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-qwen-image-layered
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2iat-performance-bagel
      - buildkite-agent artifact download "tests/dfx/perf/results/*.json" . --step nightly-diffusion-x2v-performance
+      - buildkite-agent artifact download "tests/dfx/perf/results/*.html" . --step nightly-testcase-statistics
      - python tools/nightly/generate_nightly_perf_excel.py
      - python tools/nightly/generate_nightly_perf_html.py
      - python tools/nightly/send_nightly_email.py --report-file "tests/dfx/perf/results/*.xlsx, tests/dfx/perf/results/*.html"
@@ -30,6 +30,14 @@ steps:
      - pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m "slow"
    mirror_hardwares: h100_4

+  - label: "Reliability Test - voxcpm2"
+    timeout_in_minutes: 180
+    depends_on: upload-weekly-pipeline
+    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
+    commands:
+      - pytest -s -v tests/dfx/reliability/test_reliability_voxcpm2.py -m "slow"
+    mirror_hardwares: l4_1
+
  - group: "Reliability Test - Invalid parameters Test"
    depends_on: upload-weekly-pipeline
    if: build.env("WEEKLY") == "1" || build.pull_request.labels includes "weekly-test"
@@ -141,4 +149,4 @@ steps:
            pytest -sv
            tests/e2e/ -m "slow and diffusion and H100 and distributed_cuda"
            --run-level "full_model"
-        mirror_hardwares: h100_4
+        mirror_hardwares: h100_2
@@ -69,12 +69,6 @@ steps:
          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
          HF_TOKEN: "${HF_TOKEN}"
        commands:
-          # Pin the Seed-TTS scorer to the second card: _get_eval_device() otherwise
-          # auto-picks npu:0, which the omni server under test already fills with its
-          # KV cache, so Whisper OOMs on `.to(device)` and every clip reports
-          # "ASR / WER failed".
-          - export SEED_TTS_WER_EVAL=1
-          - export SEED_TTS_EVAL_DEVICE=npu:1
          - |
            set +e
            pytest -s -v tests/e2e/accuracy/minicpmo_4_5/test_minicpmo_4_5.py -m "full_model and npu and A3" --run-level full_model
@@ -102,6 +96,33 @@ steps:
            EXIT=$$?
            buildkite-agent artifact upload "tests/dfx/perf/results/*.json"
            exit $$EXIT
+  - group: ":card_index_dividers: Diffusion X2I(&A&T) Model Test"
+    key: nightly-diffusion-x2iat-group
+    depends_on: upload-nightly-pipeline
+    if: >-
+      build.env("NIGHTLY") == "1" ||
+      build.pull_request.labels includes "nightly-test" ||
+      build.pull_request.labels includes "diffusion-x2iat-test"
+    steps:
+      - label: ":full_moon: Diffusion X2I(&A&T) · Perf Test · Qwen-Image-Edit-2511"
+        key: nightly-diffusion-x2iat-performance-qwen-image-edit-2511
+        timeout_in_minutes: 180
+        mirror_hardwares: a3_npu_8
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
+          HF_TOKEN: "${HF_TOKEN}"
+          DIFFUSION_ATTENTION_BACKEND: TORCH_SDPA
+        commands:
+          - export DIFFUSION_BENCHMARK_DIR=tests/dfx/perf/results
+          - export CACHE_DIT_VERSION=1.5.0
+          - |
+            set +e
+            pytest -s -v tests/dfx/perf/scripts/run_diffusion_benchmark.py -m npu --test-config-file tests/dfx/perf/tests/test_qwen_image_edit_2511_vllm_omni.json
+            EXIT=$$?
+            buildkite-agent artifact upload "tests/dfx/perf/results/diffusion_result_*.json"
+            buildkite-agent artifact upload "tests/dfx/perf/results/logs/*.log"
+            exit $$EXIT
  - group: ":card_index_dividers: Diffusion I2I Model Test"
    key: nightly-HunyuanImage3-x2i-test-group
    depends_on: upload-nightly-pipeline
@@ -1,15 +1,4 @@
 steps:
-  - label: "Platforms Test"
-    key: ready-platforms-test
-    depends_on: upload-ready-pipeline
-    mirror_hardwares: a3_npu_2
-    commands:
-      - >-
-        pytest -s -v
-        tests/platforms/npu/test_minicpmo_code2wav_npugraph.py
-        -m 'core_model and npu and A3 and omni'
-        --run-level 'core_model'
-
  - group: ":card_index_dividers: Omni Model Test"
    key: ready-omni-group
    depends_on: upload-ready-pipeline
@@ -66,7 +66,7 @@ repos:
      entry: python tools/pre_commit/check_tts_adapter.py
      language: python
      pass_filenames: false
-      files: ^(vllm_omni/entrypoints/openai/serving_speech\.py|tools/pre_commit/check_tts_adapter\.py)$
+      files: ^(vllm_omni/entrypoints/openai/(serving_speech\.py|tts_adapters/__init__\.py)|tools/pre_commit/check_tts_adapter\.py)$
      # The budget-is-now-slack reminder is printed on a PASSING run, and
      # pre-commit hides output from passing hooks unless the hook is verbose.
      verbose: true
@@ -43,7 +43,6 @@ nav:
    - Overview: features/README.md
    - Runtime and Stage Execution:
      - Execution Modes and Streaming: user_guide/diffusion/execution_modes.md
-      - Prefill-Decode Disaggregation (experimental): features/pd_disaggregation.md
      - Sleep Mode: features/sleep_mode.md
    - Quantization:
      - Overview: user_guide/quantization/overview.md
@@ -14,6 +14,10 @@ For introduction, please check [Introduction for stage config](./stage_configs.m

 - **[GPU Memory Calculation and Configuration](./gpu_memory_utilization.md)** - Guide on how to calculate memory requirements and set up `gpu_memory_utilization` for optimal performance

+## Multi-Stage Recipes
+
+- **[Prefill-Decode Disaggregation](./pd_disaggregation.md)** - How to derive a PD-aware Qwen3-Omni stage config from the default config without introducing another bundled YAML
+
 ## Optimization Features

 - **[Diffusion Features Overview](../user_guide/diffusion_features.md)** - Complete overview of all diffusion model features and supported models
@@ -0,0 +1,171 @@
+# Prefill-Decode (PD) Disaggregation
+
+PD disaggregation splits the Qwen3-Omni thinker into separate prefill and decode
+stages so prompt processing and token generation can run on different workers.
+
+This is documented as a stage-config recipe instead of a bundled YAML because the
+deployment-specific values usually change per environment:
+
+- GPU placement
+- `tensor_parallel_size`
+- connector backend and connector ports
+- connector IPs or bootstrap addresses
+
+Start from the [default Qwen3-Omni stage config](gh-file:vllm_omni/deploy/qwen3_omni_moe.yaml)
+and copy it to your own file, for example `qwen3_omni_pd.yaml`. Then apply the
+changes below.
+
+## Requirements
+
+- 3+ GPUs for a basic layout: prefill, decode, and talker+code2wav
+- A KV connector supported by vLLM, such as `MooncakeConnector`
+- Matching `tensor_parallel_size` on the prefill and decode thinker stages
+
+## 1. Split the thinker into prefill and decode stages
+
+Replace the original thinker stage with two stages:
+
+```yaml
+stage_args:
+  - stage_id: 0
+    stage_type: llm
+    is_prefill_only: true
+    runtime:
+      devices: "0"
+    engine_args:
+      max_num_seqs: 16
+      model_stage: thinker
+      model_arch: Qwen3OmniMoeForConditionalGeneration
+      worker_type: ar
+      scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
+      gpu_memory_utilization: 0.9
+      enforce_eager: true
+      trust_remote_code: true
+      engine_output_type: latent
+      distributed_executor_backend: "mp"
+      enable_prefix_caching: false
+      max_num_batched_tokens: 32768
+      hf_config_name: thinker_config
+      tensor_parallel_size: 1
+      kv_transfer_config:
+        kv_connector: "MooncakeConnector"
+        kv_role: "kv_producer"
+        kv_rank: 0
+        kv_parallel_size: 2
+        kv_connector_extra_config:
+          mooncake_bootstrap_port: 25201
+    final_output: false
+    is_comprehension: true
+    default_sampling_params:
+      temperature: 0.4
+      top_p: 0.9
+      top_k: 1
+      max_tokens: 2048
+      seed: 42
+      detokenize: True
+      repetition_penalty: 1.05
+
+  - stage_id: 1
+    stage_type: llm
+    is_decode_only: true
+    runtime:
+      devices: "1"
+    engine_args:
+      max_num_seqs: 64
+      model_stage: thinker
+      model_arch: Qwen3OmniMoeForConditionalGeneration
+      worker_type: ar
+      scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
+      gpu_memory_utilization: 0.9
+      enforce_eager: true
+      trust_remote_code: true
+      engine_output_type: latent
+      distributed_executor_backend: "mp"
+      enable_prefix_caching: false
+      max_num_batched_tokens: 32768
+      hf_config_name: thinker_config
+      tensor_parallel_size: 1
+      kv_transfer_config:
+        kv_connector: "MooncakeConnector"
+        kv_role: "kv_consumer"
+        kv_rank: 1
+        kv_parallel_size: 2
+        kv_connector_extra_config:
+          mooncake_bootstrap_port: 25202
+    engine_input_source: [0]
+    final_output: true
+    final_output_type: text
+    is_comprehension: true
+    default_sampling_params:
+      temperature: 0.4
+      top_p: 0.9
+      top_k: 1
+      max_tokens: 2048
+      seed: 42
+      detokenize: True
+      repetition_penalty: 1.05
+```
+
+Notes:
+
+- `is_prefill_only: true` marks the thinker stage that only saves KV.
+- `is_decode_only: true` marks the thinker stage that resumes from remote KV.
+- `kv_transfer_config` is required on both stages.
+- The orchestrator forces the prefill stage to run with `max_tokens=1`, so the
+  prefill side only processes the prompt and exports KV.
+
+## 2. Shift the downstream stages by one index
+
+After inserting the extra thinker stage, renumber the remaining stages:
+
+```yaml
+  - stage_id: 2
+    runtime:
+      devices: "2"
+    engine_input_source: [1]
+    sync_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.thinker2talker_token_only
+
+  - stage_id: 3
+    runtime:
+      devices: "2"
+    engine_args:
+      max_num_seqs: 1
+    engine_input_source: [2]
+    sync_process_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_omni.talker2code2wav_token_only
+```
+
+Compared with the default Qwen3-Omni config:
+
+- the talker becomes stage `2` instead of stage `1`
+- the code2wav stage becomes stage `3` instead of stage `2`
+- the talker now reads from decode stage `1`
+
+## 3. Add runtime edges for the four-stage pipeline
+
+```yaml
+runtime:
+  enabled: true
+  edges:
+    - from: 0
+      to: 1
+    - from: 1
+      to: 2
+    - from: 2
+      to: 3
+```
+
+## 4. Launch with your custom config
+
+```bash
+vllm serve Qwen/Qwen3-Omni-30B-A3B-Instruct --omni --port 8091 \
+    --deploy-config /path/to/qwen3_omni_pd.yaml
+```
+
+## Operational Notes
+
+- `MooncakeConnector` does not support heterogeneous TP sizes across the PD
+  pair. Keep prefill and decode at the same `tensor_parallel_size`.
+- If the thinker requires TP=2, both thinker stages must use TP=2 and be given
+  separate GPU sets, for example `"0,1"` for prefill and `"2,3"` for decode.
+- Choose connector ports and addresses that match your deployment. The values
+  shown above are examples only.
@@ -38,12 +38,11 @@ Each entry under `stages:` accepts any `StageDeployConfig` field directly (no ne
 |-------|------|----------|---------|-------------|
 | `stage_id` | int | required | — | Stage identity; matched against `PipelineConfig.stages[*].stage_id`. |
 | `max_num_seqs` | int | optional | `64` | Max concurrent sequences per stage. |
-| `gpu_memory_utilization` | float | optional | `0.9` | Per-stage total memory target; used for automatic KV-cache sizing. |
-| `kv_cache_memory_bytes` | int \| null | optional | `null` | Explicit per-rank KV-cache budget in bytes; overrides automatic sizing. |
+| `gpu_memory_utilization` | float | optional | `0.9` | Per-stage memory budget. |
 | `tensor_parallel_size` | int | optional | `1` | TP degree for this stage. |
 | `enforce_eager` | bool | optional | `false` | Disable CUDA graphs. |
-| `max_num_batched_tokens` | int | optional | `32768` | Per-stage prefill/token budget; also contributes to the native maximum in-flight token limit. |
-| `max_model_len` | int \| null | optional | `null` | Per-sequence context or KV length; `-1` enables native cache-capacity auto-fitting, while values above the HF default auto-set `VLLM_ALLOW_LONG_MAX_MODEL_LEN=1`. |
+| `max_num_batched_tokens` | int | optional | `32768` | Prefill budget. |
+| `max_model_len` | int \| null | optional | `null` | Per-stage context length (auto-sets `VLLM_ALLOW_LONG_MAX_MODEL_LEN=1` when larger than HF default). |
 | `async_scheduling` | bool \| null | optional | `null` | Per-stage async scheduling toggle. |
 | `devices` | str | optional | `"0"` | `CUDA_VISIBLE_DEVICES`-style device list. |
 | `output_connectors` | dict \| null | optional | `null` | Keyed by `to_stage_<n>`; values are names registered under top-level `connectors:`. |
@@ -184,6 +184,9 @@ Failed jobs: 1/2
    # Reliability: HunyuanImage DiT (H100 × 4)
    pytest -s -v dfx/reliability/test_reliability_hunyuan_image.py -m slow

+    # Reliability: VoxCPM2 (L4)
+    pytest -s -v dfx/reliability/test_reliability_voxcpm2.py -m slow
+
    ```

    The latest L5 CI jobs (reliability + invalid-parameter weekly steps) are in [test-weekly.yml](https://github.com/vllm-project/vllm-omni/blob/main/.buildkite/cuda/test-weekly.yml).
@@ -562,6 +562,7 @@ L5 level testing focuses on the performance of model services under ***long-runn
    -   `test_reliability_qwen3_omni.py` — Qwen3-Omni chat / multimodal reliability (GPU OOM, process kill, recovery, error contract under `--async-chunk` vs default).
    -   `test_reliability_wan22.py` — Wan2.2 I2V video API reliability (`/v1/videos` under OOM and process kill, recovery).
    -   `test_reliability_hunyuan_image.py` — HunyuanImage-3.0-Instruct DiT-only reliability (`/v1/images/generations` under OOM and process kill; deploy `hunyuan_image3_dit.yaml`, H100 × 4).
+    -   `test_reliability_voxcpm2.py` — VoxCPM2 TTS reliability (`/v1/audio/speech` under OOM and process kill; deploy `voxcpm2.yaml`, L4 × 1).
    -   `helpers.py` — Shared primitives used by current suites: raw HTTP probes for `/v1/chat/completions` and `/health`, OpenAI-style error parsing, GPU OOM sidecar (`inject_gpu_oom` / `stop_gpu_oom_hogs`), and `pgrep`-based process-kill injector construction (`make_process_kill_fault_injector`).
    -   `conftest.py` — `fault_injector` and `omni_server_after_fault` / `omni_server_after_fault_function` fixtures to run a callable **after** the server is ready.
    -   `README.md` — Short local run commands for this directory.
@@ -661,12 +662,13 @@ Reliability tests are **short fault-injection** integration runs (L5 **(b)** in
 | `test_reliability_qwen3_omni.py` | Qwen3-Omni: OOM vs **text vs text+audio** error contract, large multimodal chat under OOM, concurrent pressure, **SIGKILL** on `VLLM::Worker`, `/health` → 503 + fast-fail + concurrent chat; optional OOM recovery scenario (may be skipped while tracked in issues). |
 | `test_reliability_wan22.py` | Wan2.2 I2V: large `/v1/videos` under OOM, **SIGKILL** on `multiprocessing.spawn` chain, health / fast-fail / concurrent video requests; optional recovery test (may be skipped). |
 | `test_reliability_hunyuan_image.py` | HunyuanImage DiT-only: large `/v1/images/generations` under OOM, **SIGKILL** on `vLLM-Omni::` workers and serve/tree targets, health / fast-fail / concurrent image requests; some OOM/recovery cases may be skipped while tracked in issues. |
+| `test_reliability_voxcpm2.py` | VoxCPM2: `/v1/audio/speech` under OOM (error contract), **SIGKILL** on `VLLM::` workers and serve/tree targets, health / fast-fail / concurrent speech requests; some OOM cases may be skipped while tracked in issues. |
 | `README.md` | Minimal run / collect examples. |

 ##### Parametrization and markers

 - Each test module defines a **`RELIABILITY_SCENARIOS`** list (`test_name`, `server_params`: model, `stage_config_name` or diffusion `server_args`, etc.). **`create_reliability_omni_server_params()`** in `tests/dfx/conftest.py` resolves stage paths (including XPU substitutions where applicable) and builds **`OmniServerParams`** lists consumed by **`@pytest.mark.parametrize(..., indirect=True)`** on `omni_server` or `omni_server_function`.
- Cases are tagged **`@pytest.mark.slow`** for weekly / selective CI. GPU-heavy suites use **`@hardware_test(res={"cuda": "H100"}, num_cards=...)`** (Qwen3-Omni **2**× H100; Wan2.2 **1**× H100; HunyuanImage DiT **4**× H100).
+- Cases are tagged **`@pytest.mark.slow`** for weekly / selective CI. GPU-heavy suites use **`@hardware_test(res={"cuda": "H100"}, num_cards=...)`** or **`@hardware_test(res={"cuda": "L4"}, num_cards=1)`** (Qwen3-Omni **2**× H100; Wan2.2 **1**× H100; HunyuanImage DiT **4**× H100; VoxCPM2 **1**× L4).
 - **Process-kill** tests use **`@pytest.mark.skipif(os.name == "nt", ...)`** because injection uses POSIX **`pgrep` / `kill`**.

 ##### CI trigger
@@ -678,11 +680,13 @@ Weekly Buildkite (`.buildkite/cuda/test-weekly.yml`) runs one step per model sui
 | Reliability Test - qwen3-omni | `test_reliability_qwen3_omni.py` | H100 × 2 (`mithril-h100-pool`) |
 | Reliability Test - wan22 | `test_reliability_wan22.py` | H100 × 2 (`mithril-h100-pool`) |
 | Reliability Test - hunyuan-image | `test_reliability_hunyuan_image.py` | H100 × 4 (`mithril-h100-pool`) |
+| Reliability Test - voxcpm2 | `test_reliability_voxcpm2.py` | L4 × 1 (`gpu_1_queue`) |

 ```bash
 pytest -s -v tests/dfx/reliability/test_reliability_qwen3_omni.py -m "slow"
 pytest -s -v tests/dfx/reliability/test_reliability_wan22.py -m "slow"
 pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m "slow"
+pytest -s -v tests/dfx/reliability/test_reliability_voxcpm2.py -m "slow"
 ```

 ##### Local commands
@@ -692,6 +696,7 @@ pytest --collect-only tests/dfx/reliability
 pytest -s -v tests/dfx/reliability/test_reliability_qwen3_omni.py -m slow
 pytest -s -v tests/dfx/reliability/test_reliability_wan22.py -m slow
 pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m slow
+pytest -s -v tests/dfx/reliability/test_reliability_voxcpm2.py -m slow
 ```

 ##### Adding a new model suite
@@ -704,4 +709,4 @@ pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m slow
 </details>

 -   -   ***Stability***: `pytest -s -v tests/dfx/stability/scripts/test_stability_qwen3_omni.py` or `pytest -s -v tests/dfx/stability/scripts/test_stability_wan22.py` (or add `test_stability_<model>.py` alongside a matching JSON config)
-    -   ***Reliability***: `pytest -s -v tests/dfx/reliability/test_reliability_<model>.py -m slow` (current suites: `qwen3_omni`, `wan22`, `hunyuan_image`; add `test_reliability_<suite>.py` and a matching step in `.buildkite/cuda/test-weekly.yml` for new models)
+    -   ***Reliability***: `pytest -s -v tests/dfx/reliability/test_reliability_<model>.py -m slow` (current suites: `qwen3_omni`, `wan22`, `hunyuan_image`, `voxcpm2`; add `test_reliability_<suite>.py` and a matching step in `.buildkite/cuda/test-weekly.yml` for new models)
@@ -10,30 +10,6 @@ For user-facing commands, see the
 [distributed layerwise offloading guide](../../../user_guide/diffusion/offloader/distributed_layerwise_offload.md)
 and the [Cosmos3 recipe](https://github.com/vllm-project/vllm-omni/blob/main/recipes/cosmos3/Cosmos3-DistOffload.md).

-## Feature compatibility
-
-Host-storage optimization and runtime compatibility are separate decisions.
-When direct checkpoint mmap is unavailable, DLO can still use tensors produced
-by the ordinary loader. "Compatibility path" below means that fallback is
-implemented but has less end-to-end coverage than the primary path.
-
-Legend: ✅ supported, ⚠️ compatibility path or limited validation, ❌ unsupported.
-
-| Feature | DLO + AllGather | DLO without AllGather |
-|---|---|---|
-| **DP** | ✅ Primary path; host weights are sharded across the DP group. | ✅ Each DP rank streams complete rank-local blocks. |
-| **SP** | ✅ When DP=1, DLO uses the SP group for weight sharding. | ✅ SP remains active without a DLO weight collective. |
-| **TP > 1** | ⚠️ Ordinary TP-aware loader only; no direct checkpoint mmap. | ⚠️ Ordinary TP-aware loader only; no direct checkpoint mmap. |
-| **HSDP** | ❌ Rejected to avoid double-sharding parameters. | ⚠️ Limited end-to-end coverage. |
-| **Per-tensor online FP8 linears** | ✅ Ordinary loader finalizes weights and scales before DLO sharding. | ✅ Ordinary loader retains complete rank-local tensors. |
-| **Other online quantization methods** | ❌ Rejected until runtime packing and scale layouts are validated. | ⚠️ Allowed through the ordinary loader; validation is method-specific. |
-| **Model-level or standard layerwise CPU offload** | ❌ Disabled because DLO takes priority. | ❌ Disabled because DLO takes priority. |
-| **Resident leading layers** | ❌ Rejected. | ✅ Requires eligible resident paths in the model's `OffloadPlan`. |
-
-See [Parallelism compatibility](#parallelism-compatibility) and
-[Request and loading constraints](#request-and-loading-constraints) for the
-detailed contracts and validation boundaries.
-
 ## Status

 DLO is implemented for multi-device diffusion execution. The default
@@ -43,291 +19,23 @@ DLO weight collective.

 Host storage is selected separately from the transfer protocol. The loader can
 produce a direct-checkpoint mmap plan for a proven-compatible runtime layout;
-otherwise it uses the ordinary loader. In no-AllGather mode, the Phase B host
-weight cache can explicitly normalize supported final ordinary-loader DiT
-tensors into an immutable node-local mmap entry. Ordinary pinned weights remain
-the default after checkpoint mmap fallback when the cache is not selected; cache
-publication and validation add startup and disk lifecycle costs. Replicas can
-share either proven-compatible checkpoint pages or, when selected, equivalent
-final runtime layouts. Under the existing
-pinned-memory policy, the backend attempts to make the complete final mapping
-directly H2D-copyable when the platform provides registration, avoiding the
-recurrent host packing otherwise required by mmap. An optional positive limit
-adds a per-worker safety ceiling. CUDA is the first implementation.
+otherwise it uses the ordinary loader. In no-AllGather mode, an opt-in Phase B
+runtime cache can normalize those final ordinary-loader DiT tensors into an
+immutable node-local mmap entry. Consequently, replicas can share either
+proven-compatible checkpoint pages or equivalent final runtime layouts. An
+opt-in CUDA registration budget can make the complete final mapping directly
+H2D-copyable, avoiding the recurrent host packing otherwise required by mmap.

 The Phase A shared-mmap support boundary is TP1. TP greater than one is an
 ordinary-loader compatibility path: DLO can consume the resulting TP-local
 tensors. Direct checkpoint mmap still does not cover TP, but the Phase B cache
 can share each final TP coordinate among equivalent processes.

-The compatibility matrix below defines the current implementation boundary.
-Hardware and performance qualification is tracked separately from this
-long-lived design contract.
+The compatibility matrix below describes the current implementation. The
+unit-level guards are covered, but not every parallelism combination has a
+full model-and-hardware end-to-end test.

-## Architecture
-
-DLO is divided into four planes: configuration, loader-owned host storage,
-backend setup, and the per-block inference hot path. Keeping these decisions
-separate is intentional. In particular, an mmap cache is a host-storage choice;
-it does not imply a DLO AllGather, and disabling DLO AllGather does not decide
-whether the host source is private memory, checkpoint mmap, or host weight cache
-mmap. The shared control plane is described once below; the AllGather and
-no-AllGather startup and hot paths are then shown separately.
-
-### Component ownership
-
-The map includes optional mode-specific components, but the ownership rules
-and loader-to-backend boundary are shared.
-
-```mermaid
-flowchart TB
-    CLI["CLI / EngineArgs"] --> Config["OmniDiffusionConfig<br/>validation and operator policy"]
-    Config --> Runner["DiffusionModelRunner<br/>startup coordinator"]
-
-    subgraph LoaderPlane["Loader-owned host-storage plane"]
-        Loader["DiffusersPipelineLoader"]
-        Checkpoint["checkpoint mmap preflight"]
-        Ordinary["ordinary load + callbacks + calibration"]
-        HostWeightCache["host_weight_cache<br/>final-layout publication / join"]
-        Contract["HostWeightPlan<br/>backing + bindings + transforms"]
-        Loader --> Checkpoint
-        Loader -->|"registered final layout requested"| Ordinary
-        Checkpoint -->|"compatible"| Contract
-        Checkpoint -->|"fallback"| Ordinary
-        Ordinary -->|"host weight cache enabled"| HostWeightCache
-        HostWeightCache -->|"enabled and compatible"| Contract
-    end
-
-    Runner -->|"load_model"| Loader
-    Loader -->|"pipeline + optional HostWeightPlan"| Runner
-
-    Config --> Factory["get_offload_backend<br/>OffloadConfig normalization"]
-    Runner -->|"device + optional plan"| Factory
-    Factory --> Backend["DistributedLayerwiseOffloadBackend<br/>resource and transfer owner"]
-    Topology["existing DP / SP process groups"] --> Backend
-    Discovery["ModuleDiscovery + optional OffloadPlan"] --> Backend
-    Contract -.->|"passed once by the runner"| Backend
-
-    Backend -.->|"host weight cache + pinned-memory policy"| Registration["host_registration<br/>platform-neutral lifecycle"]
-    Registration --> CUDA["CUDA registration backend"]
-    Backend --> Hooks["DistributedLayerwiseOffloadHook<br/>one circular hook per streamed block"]
-    Forward["pipeline.forward"] --> Hooks
-    Hooks --> DataPath["two device slots<br/>H2D only or H2D + AllGather"]
-
-    Shutdown["DiffusionWorker.shutdown"] -->|"disable before process-group teardown"| Backend
-```
-
-| Component | Owns | Must not own |
-| --- | --- | --- |
-| CLI, engine arguments, and `OmniDiffusionConfig` | User policy, defaults, validation, and propagation to every worker | Checkpoint inspection, cache publication, or buffer allocation |
-| `OffloadConfig.from_od_config()` | Offload-strategy selection, effective DLO group size, AllGather policy, resident-layer count, pinned-memory policy, and optional registration ceiling | Host-weight compatibility decisions |
-| `DiffusersPipelineLoader` | Ordinary loading lifecycle and the decision to return an optional `HostWeightPlan` | DLO process groups or block-stream scheduling |
-| Checkpoint mmap planner | Proving that checkpoint tensors can represent the required runtime layout before ordinary DiT materialization is skipped | Host weight cache identity or transfer policy |
-| `host_weight_cache` | Normalizing final ordinary-loader tensors into an immutable node-local entry and returning a validated plan | Model mutation, host registration, H2D, or collectives |
-| `HostWeightPlan` | The one-way loader-to-offloader contract: backing kind, tensor bindings, source coverage, and any bounded transforms | Cache construction or runtime scheduling |
-| `ModuleDiscovery` and model `OffloadPlan` | Discovering DiT/components and declaring block, resident, and on-demand paths | Loader or cache behavior embedded in model pipelines |
-| `DistributedLayerwiseOffloadBackend` | Consuming the exact plan, selecting the existing group, mapping/repointing storage, allocating shared buffers and streams, optional host registration, installing hooks, and cleanup | Re-running loader compatibility checks or orchestrating replicas |
-| `DistributedLayerwiseOffloadHook` | The block hot path: prefetch, optional DLO AllGather, readiness events, parameter repointing, and replacement with offload placeholders | Cache keys, manifest validation, or request routing |
-| `host_registration` and platform backend | All-or-nothing registration lifecycle for existing immutable mappings | Choosing which weights are cached or changing the two-slot device-buffer contract |
-| `DiffusionWorker` | Calling backend teardown before destroying the distributed environment | Managing backend-owned mappings or registrations directly |
-
-The configuration fields are routed to the component that can enforce their
-contract. The cache root and lock timeout are consumed by the loader.
-`dlo_use_host_weight_cache` explicitly selects final-layout cache publication
-instead of direct-checkpoint mmap. It defaults to false so the ordinary
-pinned-loader path remains available without requiring an opt-out.
-With `pin_cpu_memory=True`, selecting the host weight cache deliberately chooses
-the transform-complete layout even when direct checkpoint mmap is available,
-because only that final layout can be registered without deferred transforms.
-The backend consumes `dlo_host_weight_cache_pin_limit_gib` as an optional
-registration ceiling; zero applies no additional ceiling and does not select
-the cache by itself.
-`dlo_use_allgather` controls loader plan routing because the host weight cache
-is no-AllGather-only, and it selects the backend transfer protocol. This is
-coordination through typed configuration and `HostWeightPlan`, not through
-model-specific global state.
-
-### Runtime host-layout identity (v1)
-
-Host weight cache v1 identifies the loader-produced host representation, not
-the topology used later to move weights to devices. These are separate
-contracts:
-
- `RuntimeHostLayout` describes tensor ownership and representation after the
-  ordinary loader has applied its parallel-aware transforms. This is the
-  layout stored in the host weight cache.
- `DLOTransferLayout` describes whether each worker copies its local weights
-  directly or participates in the existing DLO AllGather protocol. It is not
-  part of the host weight cache in v1; AllGather continues to use its existing
-  process-owned host shards.
-
-The v1 physical-sharing invariant is one cache entry per TP coordinate,
-shared by equivalent DP and SP consumers on the same node. The identity
-therefore includes TP world size and TP rank, because different TP coordinates
-own different tensors. It excludes DP world size/rank and SP rank, because
-those dimensions do not change the final tensor ownership supported by v1.
-SP world size and implementation details remain conservative compatibility
-guards until equivalent layouts across those configurations are proven.
-
-PP, HSDP/DTensor, expert parallelism, and CFG parallelism are rejected rather
-than assigned speculative cache identities. Supporting one of these layouts
-requires first defining its loader-produced ownership, representation, and
-sharing domain; it does not follow automatically from supporting its runtime
-collectives. Consequently, this phase implements a TP-coordinate runtime host
-layout, not a general parallel-layout cache.
-
-### Shared startup and lifecycle
-
-The shared control path selects host backing and transfers its ownership to
-the backend before either transfer mode installs hooks:
-
-```mermaid
-sequenceDiagram
-    participant C as OmniDiffusionConfig
-    participant R as DiffusionModelRunner
-    participant L as DiffusersPipelineLoader
-    participant B as DLO backend
-    participant W as DiffusionWorker
-
-    C->>R: validated offload policy
-    R->>L: load_model(config)
-    L->>L: select and validate host backing
-    L-->>R: pipeline + optional HostWeightPlan
-    R->>B: enable(config, device, optional plan)
-    B->>B: discover modules and consume host backing
-    B-->>R: mode-specific backend is ready
-    W->>B: shutdown calls disable()
-    B->>B: synchronize, unregister, remove hooks, close mappings
-```
-
-The loader returns a plan only after its contract is valid: checkpoint plans
-have preflight-proven source coverage and metadata, while host weight cache
-plans also have full manifest, file, and mapped-content validation. The runner
-then transfers ownership exactly once with
-`model_loader.take_host_weight_plan()`. If no plan is returned, the backend
-requires materialized ordinary-loader tensors. If a plan caused ordinary DiT
-materialization to be skipped, failure to create or enable its consumer is a
-hard startup error rather than a fallback to uninitialized parameters.
-
-At inference time, the model pipeline remains the caller: normal block forward
-calls activate the installed hooks. Both modes use one copy stream and two
-shared device slots sized for the largest streamed block. The hooks do not
-call the loader or cache.
-
-### AllGather startup and hot path
-
-```mermaid
-sequenceDiagram
-    participant B as DLO backend
-    participant H as block hooks
-    participant G as existing DP or SP group
-
-    B->>B: choose existing group and prepare persistent local shards
-    B->>B: allocate two device slots and a communication stream
-    B->>H: install hooks and prime the first block
-    loop each streamed block
-        H->>H: copy the next local shard to the alternate slot
-        H->>G: all_gather_into_tensor(next block)
-        G-->>H: complete next block is ready
-        H->>H: execute current block and rotate slots
-    end
-```
-
-AllGather uses the existing DP group when DP is greater than one, otherwise
-the existing SP group when SP is greater than one. TP is never selected as
-DLO's weight-sharding group. Every participating rank must request blocks in
-the same order; fast P2P reduces communication cost but does not remove that
-lockstep contract. An effective group size of one issues no collective and
-uses rank-local transfer.
-
-### No-AllGather startup and hot path
-
-```mermaid
-sequenceDiagram
-    participant L as DiffusersPipelineLoader
-    participant C as host_weight_cache
-    participant B as DLO backend
-    participant P as platform registration
-    participant H as block hooks
-
-    alt host weight cache explicitly selected
-        L->>L: load, mutate, validate, and calibrate final tensors
-        L->>C: publish or join final-layout entry
-        C-->>L: validated host weight cache plan or fail-closed fallback
-        L-->>B: cache plan or materialized fallback via runner
-    else automatic checkpoint selection
-        L->>L: checkpoint mmap preflight
-        alt loader selects a compatible checkpoint plan
-            L-->>B: checkpoint plan via runner
-        else ordinary loading
-            L->>L: load, mutate, validate, and calibrate final tensors
-            L-->>B: materialized tensors via runner
-        end
-    end
-    B->>B: consume complete rank-local block sources
-    opt host weight cache plan and pinned-memory policy
-        B->>P: register the complete immutable mapping
-        P-->>B: direct-H2D source or bounded-staging fallback
-    end
-    B->>H: install hooks and prime the first block
-    loop each streamed block
-        alt registered host weight cache mapping
-            H->>H: copy complete next block directly to device
-        else pageable mmap mapping
-            H->>H: pack into a bounded staging slot, then copy to device
-        else ordinary runtime tensors
-            H->>H: copy complete next block to device
-        end
-        H->>H: execute current block and rotate slots
-    end
-```
-
-No-AllGather creates no DLO weight group and issues no DLO collective.
-Workers may therefore schedule requests independently. Ordinary TP and SP
-collectives remain inside model execution; host registration changes only
-the source of the complete-block H2D copy.
-
-### Host backing and transfer composition
-
-| Loader-selected host backing | Backend transfer mode | Setup result | Per-block hot path |
-| --- | --- | --- | --- |
-| Ordinary runtime tensors | AllGather group larger than one | Each rank retains a pinned local weight shard | shard H2D → DLO AllGather → device slot |
-| Checkpoint mmap | AllGather group larger than one | Each rank prepares its persistent shard, then releases the source mapping | shard H2D → DLO AllGather → device slot |
-| Ordinary runtime tensors | Rank-local / no-AllGather | Each rank retains the complete loader-produced local layout | complete-block H2D → device slot |
-| Checkpoint mmap, or host weight cache with pinned memory disabled/registration unavailable | Rank-local / no-AllGather | Immutable mapping plus two bounded host-staging slots | mmap views → staging slot → device slot |
-| Host weight cache mmap with pinned memory enabled and registration available | Rank-local / no-AllGather | Complete mapping is registered by the platform backend | registered mmap views → device slot |
-
-The three axes therefore remain independent:
-
-1. **Offload policy:** whether DLO is enabled and which layers are resident.
-2. **Host backing:** ordinary tensors, checkpoint mmap, or final host weight cache
-   mmap, selected by the loader.
-3. **Transfer protocol:** rank-local H2D or H2D plus DLO AllGather, selected by
-   the backend from validated configuration and existing topology.
-
-### Failure and lifecycle boundaries
-
- Invalid option combinations fail during configuration validation.
- Checkpoint mmap preflight failure falls back before ordinary loading is
-  skipped.
- Host weight cache incompatibility, timeout, or publication failure retains
-  the already-valid ordinary tensors.
- Host-registration failure rolls back and retains bounded staging. A rollback
-  failure aborts startup because unmapping a range still owned by the platform
-  would be unsafe.
- During shutdown, DLO synchronizes outstanding device work, unregisters host
-  mappings, removes hooks, and closes mmap handles only after safe unregister;
-  the worker destroys its process groups afterward.
-
-DLO deliberately does not own DP replica placement, independent-engine request
-routing, cross-engine admission control, or cross-node cache distribution.
-Those belong to an external router/orchestrator. It also does not replace TP or
-SP model collectives; it only adds an optional weight collective over an
-already-existing DP or SP group.
-
-## Shared design
-
-The following topology and loader contracts apply to both transfer modes.
+## Design

 ### DLO consumes the existing parallel topology

@@ -335,10 +43,15 @@ DLO does not create a new DP, TP, or SP topology. It reads the configured
 `DiffusionParallelConfig` and attaches offload hooks to the DiT blocks after the
 standard distributed groups have been initialized.

-AllGather mode selects one existing group for DLO weight sharding; no-AllGather
-mode creates no DLO group. Neither mode changes ordinary TP or SP model
-collectives. The mode-specific group and scheduling contracts are described
-below.
+The DLO weight-sharding group is selected as follows:
+
+1. Use the existing DP group when `data_parallel_size > 1`.
+2. When DP is one and SP is greater than one, use the SP group.
+3. Otherwise, run rank-locally without a DLO process group.
+
+TP is deliberately not used as DLO's AllGather group. HSDP has its own
+parameter-sharding lifecycle and is not allowed to be sharded a second time by
+DLO's AllGather path.

 ### The loader owns host-weight planning

@@ -370,24 +83,91 @@ flags such as `_supports_mmap_loading` or parameter attributes for mmap-only
 transforms. Model-specific direct-layout knowledge, when required, lives in a
 checkpoint adapter beside the ordinary loader.

-## AllGather design
+### Post-load runtime cache

-### Group selection
+The Phase B cache preserves the same ownership boundary. DLO neither derives
+cache keys nor interprets model loaders. When enabled with no-AllGather, every
+rank first completes the ordinary loader, custom callbacks, post-load
+processing, strict validation, and calibration. The loader then:

-The DLO weight-sharding group is selected in this order:
+1. discovers final CPU DiT parameters and persistent buffers;
+2. rejects unsupported non-contiguous, aliased, quantized, distributed, or
+   ambiguous layouts before changing the model;
+3. hashes tensor names, metadata, and final bytes;
+4. builds or joins one content-addressed entry under a strict POSIX lock; and
+5. returns a `HostWeightPlan(backing_kind="runtime_cache")` only after file,
+   manifest, and remapped-tensor content validation succeeds.

-1. Use the existing DP group when `data_parallel_size > 1`.
-2. When DP is one and SP is greater than one, use the existing SP group.
-3. Otherwise, use an effective group size of one and issue no collective.
+The identity excludes DP size/rank and SP rank. It includes TP world size/rank
+and conservative SP implementation/world-size guards. DP replicas and
+equivalent SP ranks can therefore share; different TP coordinates cannot.
+Final tensor content is authoritative, so an unmodeled loader/config change
+that changes bytes creates a different entry instead of risking silent
+sharing.

-TP is deliberately not used as DLO's AllGather group. HSDP has its own
-parameter-sharding lifecycle and cannot be sharded a second time by DLO.
+Publication writes sharded safetensors and a JSON manifest into a same-
+filesystem temporary directory, flushes them, and atomically renames the
+directory. The manifest includes per-file SHA-256 digests and the final tensor
+content digest. Kernel `flock` ownership distinguishes a live writer from a
+stale lock file: process death releases the lock automatically. Stale temporary
+entries are removed only by the next lock owner. A bounded wait or any cache
+error retains the already-valid ordinary tensors.

-### Sharded transfer and double buffering
+DLO maps this plan without moving the DiT to `meta` and without rerunning any
+post-load hook. It changes tensor storage while preserving `Parameter` objects,
+then runs the existing `gc.collect()`/`malloc_trim()` cleanup before installing
+streaming hooks. The source mappings remain open through DLO disable. With the
+default zero registration budget they feed the existing bounded two-slot host
+staging path. Every streamed block is first packed from pageable mmap storage
+into one of those pinned slots and is then copied to the device. This removes
+model-sized private pinned storage, but adds a recurrent host-to-host copy and
+can amplify TP synchronization waits when several ranks compete for host
+bandwidth.

-The loader may provide ordinary runtime tensors or a compatible checkpoint
-plan. During backend setup, either source is normalized into persistent local
-shards. The final-layout host weight cache is not selected in AllGather mode.
+This version is deliberately opt-in and steady-state-oriented. It requires
+roughly one local-disk model copy per runtime identity, has no eviction, and
+still performs ordinary loading plus full content-hash passes in every rank.
+Skip-load-on-hit is a separate follow-up after loader side effects can be
+modeled safely.
+
+### Registered runtime-cache transfer
+
+`dlo_runtime_cache_pin_limit_gib > 0` opts a CUDA worker into complete
+runtime-cache registration. It also changes loader selection: no-AllGather
+uses the ordinary loader and final runtime cache even when a TP1 direct-
+checkpoint plan is available. A checkpoint binding can carry a deferred
+adapter such as grouped-QKV reordering. Registering that raw source would not
+remove the adapter's recurrent CPU allocation/copy, whereas the runtime cache
+contains the transform-complete final tensors.
+
+After mapping and validating every final tensor, DLO groups source storages by
+backing file, page-aligns their address ranges, and coalesces overlapping or
+adjacent ranges within that mapping. The complete byte count is checked
+against the per-worker budget before any CUDA call. It then registers every
+range with the read-only CUDA host-registration flag and verifies that PyTorch
+recognizes every mapped tensor as pinned. Failure rolls back any ranges already
+registered and preserves the existing two-slot staging path. Partial direct
+transfer is deliberately unsupported. A rollback failure aborts worker
+initialization so a still-registered range is retained until process/context
+teardown rather than being unsafely unmapped.
+
+On success, the hooks copy each immutable tensor view directly into its offset
+in the existing flattened rotating device buffer on the copy stream. Parameter
+repointing, two-block HBM residency, and H2D/compute overlap are unchanged. The
+implementation currently issues more H2D copy operations than block packing,
+but avoids the much larger recurrent CPU copy. A packed cache schema should be
+considered only if launch-count measurements justify its added format and
+publication complexity.
+
+Registration is per process and CUDA context; physical file-cache pages remain
+shared and are not copied into one private host allocation per worker. The
+operator must nevertheless budget page-locked memory and satisfy OS/CUDA
+registration limits for every worker. During shutdown, DLO first synchronizes
+outstanding device work, unregisters all ranges, and then closes the safetensors
+mappings. Worker teardown invokes this cleanup before destroying the
+distributed environment and CUDA context.
+
+### AllGather path

 With the default `dlo_use_allgather=True`, each rank stores approximately
 `1 / group_size` of each streamable block in pinned host memory. The next
@@ -395,15 +175,6 @@ block's shard is copied to a device buffer and reconstructed with
 `all_gather_into_tensor` on a communication stream while the current block is
 executing.

-![AllGather DLO sharded host weights and lockstep reconstruction](../../figures/dlo/dlo_allgather_sharded_host.svg)
-
-The figure uses DP2 with TP2 to expose both parallel dimensions. DLO creates
-one AllGather group across the DP replicas for each TP coordinate; TP0 workers
-reconstruct TP0 blocks together and TP1 workers reconstruct TP1 blocks
-together. Ordinary TP collectives remain inside each DP replica. Unlike the
-no-AllGather host weight cache path, these persistent pinned host shards are
-process-owned, and every participating rank must follow the same block order.
-
 ```text
 Compute:    [Block N]             [Block N+1]          [Block N+2]
 H2D:                      [shard N+1]           [shard N+2]
@@ -434,25 +205,13 @@ independently scheduled replicas—even on one NVSwitch baseboard—no-AllGather
 plus shared mmap storage avoids coupling their request schedules and failure
 domains.

-## No-AllGather design
-
-### Rank-local transfer
+### Rank-local path without DLO AllGather

 With `--dlo-no-use-allgather`, DLO forces its internal offload shard size to
 one and streams complete blocks using H2D copies only. The host backing may be
-a loader-approved checkpoint mapping, a normalized host weight cache mapping, or
+a loader-approved checkpoint mapping, a normalized runtime-cache mapping, or
 ordinary runtime tensors.

-![No-AllGather DLO shared host weights and independent runtime orchestration](../../figures/dlo/dlo_no_allgather_shared_host.svg)
-
-The figure uses TP2 to make per-coordinate physical sharing explicit. Each
-engine contains two rank-local workers and devices. The cache keeps one
-immutable entry per TP coordinate, so equivalent TP0 workers across engines
-share the TP0 entry and equivalent TP1 workers share the separate TP1 entry.
-An external router or orchestrator chooses each engine's requests; DLO
-coordinates only the worker-local mapping, hooks, transfer streams, events,
-and device slots.
-
 For direct mmap, each process retains immutable safetensors views and uses two
 bounded pinned host staging slots. Processes on the same node that map the same
 files share physical checkpoint pages through the OS page cache. This removes
@@ -460,17 +219,14 @@ the persistent private full-model copy per pure-DP process, but each process
 still packs and transfers every complete block. Sharing is node-local; each
 node has its own page cache.

-For a host weight cache mapping, the default pinned-memory policy attempts to
-replace those host slots with direct copies from the shared final tensor views.
-Disabling pinned memory or a failed registration retains staging behavior.
+For a runtime-cache mapping, a positive registration budget replaces those
+host slots with direct copies from the shared final tensor views. A zero budget
+or failed registration retains the bounded staging behavior.

-When direct mmap preflight fails or the host weight cache is selected, the
-regular model loader remains responsible for preparing each rank's weights,
-including TP-local tensors or HSDP-managed parameters. Ordinary pinned tensors
-remain the default without cache selection. When the host weight cache is
-explicitly selected,
-supported CPU layouts enter the cache; unsupported layouts keep one private
-runtime copy per process.
+When direct mmap preflight fails, the regular model loader remains responsible
+for preparing each rank's weights, including TP-local tensors or HSDP-managed
+parameters. Supported CPU layouts may then enter the opt-in runtime cache.
+Unsupported layouts keep one private runtime copy per process.

 This mode means:

@@ -481,109 +237,18 @@ This mode means:
 - TP/HSDP/SP collectives, if configured, are not disabled by this flag; only
  DLO's additional weight AllGather is disabled.
 - Pure DP deployments share one checkpoint-backed copy per node when direct
-  mmap is selected. The host weight cache can share equivalent ordinary-loader
+  mmap is selected. The runtime cache can share equivalent ordinary-loader
  outputs; its fallback keeps one private runtime copy per rank.
 - The scheduler does not require a synchronized DP request wave for DLO.

-### Post-load host weight cache
-
-The Phase B cache preserves the same ownership boundary. DLO neither derives
-cache keys nor interprets model loaders. No-AllGather DLO first attempts a
-compatible direct-checkpoint mmap plan. When that plan is unavailable, every
-rank completes the ordinary loader, custom callbacks, post-load processing,
-strict validation, and calibration. When cache use or registration is
-explicitly selected, the loader:
-
-1. discovers final CPU DiT parameters and persistent buffers;
-2. rejects unsupported non-contiguous, aliased, quantized, distributed, or
-   ambiguous layouts before changing the model;
-3. hashes tensor names, metadata, and final bytes;
-4. builds or joins one content-addressed entry under a strict POSIX lock; and
-5. returns a `HostWeightPlan(backing_kind="host_weight_cache")` only after file,
-   manifest, and remapped-tensor content validation succeeds.
-
-The identity excludes DP size/rank and SP rank. It includes TP world size/rank
-and conservative SP implementation/world-size guards. DP replicas and
-equivalent SP ranks can therefore share; different TP coordinates cannot.
-Final tensor content is authoritative, so an unmodeled loader/config change
-that changes bytes creates a different entry instead of risking silent
-sharing.
-
-Publication writes sharded safetensors and a JSON manifest into a same-
-filesystem temporary directory, flushes them, and atomically renames the
-directory. The manifest includes per-file SHA-256 digests and the final tensor
-content digest. Kernel `flock` ownership distinguishes a live writer from a
-stale lock file: process death releases the lock automatically. Stale temporary
-entries are removed only by the next lock owner. A bounded wait or any cache
-error retains the already-valid ordinary tensors.
-
-DLO maps this plan without moving the DiT to `meta` and without rerunning any
-post-load hook. It changes tensor storage while preserving `Parameter` objects,
-then runs the existing `gc.collect()`/`malloc_trim()` cleanup before installing
-streaming hooks. The source mappings remain open through DLO disable. With
-pinned memory disabled, or when registration is unavailable, they feed the
-existing bounded two-slot host-staging path. Every streamed block is first
-packed from pageable mmap storage into a staging slot and is then copied to the
-device. This removes model-sized private pinned storage, but adds a recurrent
-host-to-host copy and can amplify TP synchronization waits when several ranks
-compete for host bandwidth.
-
-This version is deliberately steady-state-oriented. It requires roughly one
-local-disk model copy per runtime identity, has no eviction, and still performs
-ordinary loading plus full content-hash passes in every rank. Cache publication
-or validation failure falls back to the already-valid ordinary host tensors.
-Skip-load-on-hit is a separate follow-up after loader side effects can be
-modeled safely.
-
-### Registered host weight cache transfer
-
-Selecting the host weight cache with `pin_cpu_memory=True` opts a worker into
-registration of the complete mapping when its platform supports it. It also
-changes loader selection: no-AllGather uses the ordinary loader and final-layout
-host weight cache even when a TP1 direct-checkpoint plan is available. A
-checkpoint binding can carry a deferred adapter such as
-grouped-QKV reordering. Registering that raw source would not remove the
-adapter's recurrent CPU allocation/copy, whereas the host weight cache contains
-the transform-complete final tensors.
-
-After mapping and validating every final tensor, DLO groups source storages by
-backing file, page-aligns their address ranges, and coalesces overlapping or
-adjacent ranges within that mapping. When a positive per-worker ceiling is
-configured, the complete byte count is checked against it before any platform
-mutation. The CUDA backend
-first requires `cudaDevAttrHostRegisterReadOnlySupported`, then registers every
-range with the read-only host-registration flag and verifies that PyTorch
-recognizes every mapped tensor as pinned. A device without that capability
-retains bounded staging; vLLM-Omni does not replace immutable mappings with a
-writable or copy-on-write CUDA-specific layout. Other registration failures
-roll back any ranges already registered and preserve the existing two-slot
-staging path. Partial direct transfer is deliberately unsupported. A rollback
-failure aborts worker initialization so a still-registered range is retained
-until process/context teardown rather than being unsafely unmapped.
-
-On success, the hooks copy each immutable tensor view directly into its offset
-in the existing flattened rotating device buffer on the copy stream. Parameter
-repointing, two-block HBM residency, and H2D/compute overlap are unchanged. The
-direct path trades recurrent host packing for finer-grained H2D operations. A
-packed cache schema should be considered only if that additional format and
-publication complexity becomes necessary.
-
-Registration is per process and platform context; physical file-cache pages
-remain shared and are not copied into one private host allocation per worker.
-The operator must nevertheless budget page-locked memory and satisfy the
-platform and OS registration limits for every worker. During shutdown, DLO
-first synchronizes outstanding device work, unregisters all ranges, and then
-closes the safetensors mappings. Worker teardown invokes this cleanup before
-destroying the distributed environment and platform context.
-
 ## Parallelism compatibility

 | Parallelism | DLO + AllGather | DLO without AllGather |
 |---|---|---|
-| **DP** | Supported primary path. DLO shards host weights across the DP group and can run DP multi-concurrency. | Supported rank-local path. Compatible TP1 replicas can share checkpoint pages; the optional host weight cache shares equivalent final layouts while excluding DP rank from its identity. CUDA workers may register the complete final mapping for direct H2D. |
-| **SP** | Supported in the implementation. With DP=1, DLO uses the SP group for host-weight sharding; SP still shards sequence/activation work. | SP remains active without a DLO weight collective. Host weight cache v1 excludes SP rank and includes conservative SP implementation/world-size guards. Registered transfer does not remove ordinary SP activation/attention collectives. |
-| **TP > 1** | Outside the Phase A direct-mmap scope. The loader preserves TP-local layouts and DLO may apply DP/SP host sharding to ordinary runtime tensors. | The ordinary TP-aware loader produces rank-local tensors. Host weight cache v1 creates one entry per TP coordinate, shareable by equivalent DP/SP processes. Registering those final mappings avoids the recurrent staging copy but does not remove ordinary TP collectives. |
-| **HSDP** | Rejected. HSDP has already sharded parameters, so DLO AllGather would double-shard them. | Accepted only with ordinary-loader rank-local tensors. HSDP owns parameter sharding and its own gathers; host weight cache v1 rejects HSDP/DTensor layouts. |
+| **DP** | Supported primary path. DLO shards host weights across the DP group and can run DP multi-concurrency. | Supported rank-local path. Compatible TP1 replicas can share checkpoint pages; the optional runtime cache shares equivalent final layouts while excluding DP rank from its identity. CUDA workers may register the complete final mapping for direct H2D. |
+| **SP** | Supported in the implementation. With DP=1, DLO uses the SP group for host-weight sharding; SP still shards sequence/activation work. | SP remains active without a DLO weight collective. Runtime-cache v1 excludes SP rank and includes conservative SP implementation/world-size guards. Registered transfer does not remove ordinary SP activation/attention collectives. |
+| **TP > 1** | Outside the Phase A direct-mmap scope. The loader preserves TP-local layouts and DLO may apply DP/SP host sharding to ordinary runtime tensors. | The ordinary TP-aware loader produces rank-local tensors. Runtime-cache v1 creates one entry per TP coordinate, shareable by equivalent DP/SP processes. Registering those final mappings removes staging-driven TP skew but does not remove ordinary TP collectives. |
+| **HSDP** | Rejected. HSDP has already sharded parameters, so DLO AllGather would double-shard them. | Accepted by configuration. HSDP owns parameter sharding and its own gathers; DLO only stages rank-local parameters. End-to-end coverage is limited. |

 ### Combined dimensions

@@ -599,9 +264,7 @@ destroying the distributed environment and platform context.
 - **HSDP + DP or TP:** rejected independently by the diffusion parallel
  configuration.

-## Mode constraints
-
-### AllGather constraints
+## Request and loading constraints

 AllGather DP multi-concurrency requires:

@@ -609,54 +272,201 @@ AllGather DP multi-concurrency requires:
 - the same `num_inference_steps` for all requests in a wave; and
 - identical request arguments that affect the collective execution path.

-Per-tensor online FP8 linears can use DLO AllGather after the ordinary loader
-finalizes their runtime weights and scales. Other online methods must use
-`--dlo-no-use-allgather` or disable online quantization until their runtime
-layouts are validated.
-
-### No-AllGather constraints
-
 The no-AllGather path does not impose these DLO-specific synchronized-wave
-requirements. Host weight cache v1 is no-AllGather only and rejects all
-quantization configurations, HSDP/DTensor, expert parallelism, CFG
-parallelism, PP, non-contiguous tensors, and tied/shared storage.
-
-### Shared loading constraint
+requirements.

 Direct checkpoint mmap can back either transfer path. It is currently limited
 to proven TP1, non-HSDP, non-online-quantized layouts. Other layouts use the
-ordinary loader.
+ordinary loader. Online quantization remains incompatible with DLO AllGather;
+use `--dlo-no-use-allgather` or disable online quantization. Runtime-cache v1
+is no-AllGather only and rejects all quantization configurations, HSDP/DTensor,
+expert parallelism, CFG parallelism, PP, non-contiguous tensors, and
+tied/shared storage.

-## Validation status
+## Validation coverage

-Historical test counts and hardware measurements are intentionally not kept
-in this design document because they can drift as the implementation changes.
-Revalidation must run against the final commit and publish commit-pinned
-artifacts before making hardware or performance claims. Track that work in
-[issue #6231](https://github.com/vllm-project/vllm-omni/issues/6231).
+Current source-level validation includes:
+
+- HSDP + DLO + AllGather rejection;
+- HSDP + DLO without AllGather acceptance at configuration level;
+- loader preflight fallback for TP, HSDP, online quantization, unknown custom
+  loaders, missing keys, and shape/dtype mismatches;
+- exact loader-to-backend plan transfer and ordinary-loader fallback;
+- rank-local mmap source retention, bounded two-slot staging, and adapter
+  transforms without parameter-side flags;
+- runtime-cache content identity, sharded atomic publication, concurrent
+  writer election, corruption rebuild, lock timeout, and fail-closed layout
+  rejection;
+- loader ordering through final post-load mutation and DLO remapping without
+  rerunning post-load hooks;
+- all-or-nothing page-aligned registration, budget rejection, partial-failure
+  rollback, direct-copy staging bypass, and explicit shutdown cleanup;
+- resident-layer requests requiring no-AllGather;
+- DP request-wave validation for denoising-step compatibility;
+- sharding, double-buffer, AllGather-size, and heterogeneous-block regression
+  tests.
+
+### B300 parallel-topology smoke matrix
+
+A four-GPU B300 smoke test covered MiniMax-H3 FL2VA with the same prompt, seed,
+CUDNN attention backend, 256x256 output, two denoising steps, and
+`dlo_resident_layers=0`. The TP2 rows used DiT DP2xTP2 with the text encoder and
+VAEs at TP1. They validate the ordinary-loader fallback only, not direct mmap
+or shared-mmap host-memory savings.
+
+| Configuration | Result | Warm E2E | Peak device memory | Host PSS |
+|---|---:|---:|---:|---:|
+| DP4xTP1 AllGather | Passed, 4 concurrent requests | 2.87 s / 4 requests | 13.84 GiB | 211.99 GiB |
+| DP4xTP1 no-AllGather | Passed, 1 request | 15.02 s | 13.23 GiB | 187.77 GiB |
+| DP2xTP2 AllGather | Passed, 2 concurrent requests | 4.16 s / 2 requests | 12.50 GiB | 211.97 GiB |
+| DP2xTP2 no-AllGather | Passed, 1 request | 3.51 s | 11.88 GiB | 314.01 GiB |
+
+Within each topology, the AllGather and no-AllGather video and audio outputs
+were byte-identical. All four runs completed without an `ERROR` or traceback
+and released their device allocations. For DP4xTP1, no-AllGather direct mmap
+reduced total PSS by 24.22 GiB (11.4%) and `Private_Dirty` from 211.33 to
+125.32 GiB (40.7%) relative to AllGather. For DP2xTP2, preflight selected the
+ordinary loader as designed; no-AllGather PSS was 314.01 GiB, about 48% above
+AllGather, because DP replicas did not share checkpoint-backed runtime
+weights. This is a functional and memory smoke test, not a production-quality
+performance or output-quality benchmark.
+
+### Host-memory measurement
+
+A two-worker MiniMax-H3 FL2VA measurement on one L20X node compared the
+ordinary-loader fallback with direct mmap. Both runs used
+DP=2, TP=1, no DLO AllGather, BF16 weights, two denoising steps, and a
+256x256 four-second request. The ordinary-loader workers were sampled after
+initialization. The mmap workers were sampled after one completed request, so
+the checkpoint working set had been faulted into the page cache; this is the
+more conservative point for mmap.
+
+The values below come from `/proc/<worker>/smaps_rollup` and include the whole
+worker, not only the DiT. The stable rank-to-rank difference comes from other
+pipeline components, so each worker should be compared with the same worker in
+the other storage mode.
+
+| Worker | Ordinary RSS | mmap RSS | Ordinary PSS | mmap PSS | PSS reduction |
+|---|---:|---:|---:|---:|---:|
+| DP worker 0 | 168.27 GiB | 132.76 GiB | 167.84 GiB | 101.43 GiB | 66.40 GiB |
+| DP worker 1 | 116.19 GiB | 79.97 GiB | 115.73 GiB | 48.64 GiB | 67.09 GiB |
+| **Two-worker total** | — | — | **283.56 GiB** | **150.08 GiB** | **133.48 GiB (47.1%)** |
+
+The direct-mmap workers each reported 62.45 GiB `Shared_Clean` but only
+31.20 GiB `Pss_File`, which is the proportional charge expected when the same
+resident checkpoint pages are mapped by two workers. `Private_Dirty` also fell
+from 167.53 to 70.24 GiB for worker 0 and from 115.40 to 17.44 GiB for worker
+1, a reduction of about 97–98 GiB per worker. RSS understates this benefit
+because it counts a shared physical page in every process that maps it; summed
+PSS is the appropriate node-memory comparison.
+
+### L20X runtime-cache smoke matrix
+
+Phase B was also exercised on four L20X GPUs in one NVLink domain with
+MiniMax-H3 FL2VA, CUDNN attention, eager execution, 256x256 four-second
+outputs, two denoising steps, `dlo_resident_layers=0`, text-encoder TP1, and
+VAE TP1. This is a functional, memory, and profiler smoke test rather than a
+production-quality performance benchmark. Host values below are node-wide
+maxima during request waves from descendant-process `smaps_rollup`; they do
+not represent the private startup peak. HBM is the engine-reported request
+peak because sampled `nvidia-smi` data can miss short-lived allocations.
+
+One internal DP2xTP2 engine compared the three host-storage/transfer choices.
+Each measured wave contained two requests. NVLink TX is the physical per-link
+counter delta summed across the four GPUs; RX had the same value and is not
+added again here.
+
+| Mode | Two-request wave | Throughput | Host PSS | `Private_Dirty` | Engine peak HBM | NVLink TX / wave |
+|---|---:|---:|---:|---:|---:|---:|
+| AllGather, ordinary loader | 3.41 s | 0.587 req/s | 210.70 GiB | 209.72 GiB | 12,498 MB | 73.37 GiB |
+| no-AllGather, private loader weights | 6.96 s | 0.287 req/s | 312.56 GiB | 311.56 GiB | 11,882 MB | 19.78 GiB |
+| no-AllGather, runtime cache (cold publish) | 18.94 s | 0.106 req/s | 178.71 GiB | 116.00 GiB | 11,882 MB | 19.78 GiB |
+| no-AllGather, runtime cache (warm reuse) | 24.23 s | 0.083 req/s | 178.29 GiB | 115.58 GiB | 11,882 MB | 19.78 GiB |
+
+The no-AllGather cache preserved the no-AllGather NVLink traffic and HBM
+footprint while reducing node PSS by about 134 GiB relative to private loader
+weights. Cold and warm labels describe cache publication or reuse during
+initialization; both request phases ran after initialization. A warm hit still
+performed ordinary loading and full content hashing, so it was not a startup
+fast path. All measured no-AllGather private/cache outputs were byte-identical
+for matching seeds.
+
+The deployment shape targeted by the cache was tested separately as two
+independent DP1xTP2 engines on the same node. Both engines mapped the same two
+TP-coordinate entries, including the same file device/inode tuples; the cache
+occupied 66,281,995,903 bytes. A start barrier kept all four measured requests
+ahead of profiling.
+
+| Storage | Mean request latency, engines A / B | Combined throughput | Host PSS | `Private_Dirty` | Engine peak HBM |
+|---|---:|---:|---:|---:|---:|
+| Private loader weights | 3.50 / 3.60 s | 0.463 req/s | 366.20 GiB | 365.22 GiB | 11,882 MB |
+| Shared runtime cache | 8.96 / 12.30 s | 0.142 req/s | 232.23 GiB | 169.52 GiB | 11,882 MB |
+
+The shared cache reduced request-time PSS by 133.97 GiB (36.6%) and
+`Private_Dirty` by 195.70 GiB (53.6%), but reduced combined throughput by
+69.3% in this host-contention smoke. Across one profiled request per engine,
+aggregate GPU compute was unchanged (1.359 versus 1.357 seconds) and H2D
+payload was exactly 264.24 GiB in both modes. Aggregate CPU `aten::copy_` time
+rose from 3.75 to 32.49 seconds, while communication-kernel time rose from
+3.56 to 5.80 seconds as TP ranks waited on skewed staging. These durations are
+summed across ranks and can overlap; they identify the mechanism, not wall
+time. Matching private/cache outputs were byte-identical.
+
+The result confirms that zero-budget runtime-cache staging is a host-capacity
+option, not a latency-neutral replacement for private pinned weights. Broader
+SP, model, platform, and production-quality validation remains tracked in
+[issue #6231](https://github.com/vllm-project/vllm-omni/issues/6231). The
+historical B300 rows above predate the runtime cache and must not be treated as
+Phase B memory validation.
+
+### L20X registered-runtime-cache smoke
+
+The read-only registered path was then tested with the same MiniMax-H3
+workload. These are functional and mechanism-validation measurements, not a
+production benchmark. The TP1 pair used one request per wave. Both modes moved
+exactly 121.86 GiB H2D in the profiled request and produced byte-identical video
+and audio for both measured seeds.
+
+| TP1 no-AllGather mode | Mean wave | Throughput | Request-time host PSS | `Private_Dirty` | Engine peak HBM | CPU `aten::copy_` | GPU compute |
+|---|---:|---:|---:|---:|---:|---:|---:|
+| Checkpoint mmap + bounded staging | 33.87 s | 0.0295 req/s | 133.65 GiB | 71.16 GiB | 13,226 MB | 21.91 s | 419.28 ms |
+| Final runtime cache + registered direct H2D | 4.88 s | 0.2055 req/s | 129.41 GiB | 66.93 GiB | 13,226 MB | 1.64 s | 419.05 ms |
+
+Registration covered 61.73 GiB in 13 page-aligned ranges. Mean wave latency
+fell by 85.6% and throughput increased 6.94x. GPU compute and H2D payload were
+unchanged, identifying recurrent host transform/packing as the removed work.
+The direct path issued 2,425 H2D events versus 1,945 for packed staging, but
+that launch increase was much smaller than the eliminated CPU cost.
+
+Two independent TP2 engines also registered the same two TP-coordinate cache
+entries across four workers. Compared with the zero-budget cache, mean wave
+latency improved from 9.21/12.55 seconds to 3.80/3.86 seconds for engines A/B;
+the private-loader reference was 3.61/3.72 seconds. Registered request-time
+host PSS was 224.26 GiB versus 232.23 GiB staged and 366.20 GiB private.
+Aggregate CPU `aten::copy_` time was 3.83 seconds registered, 32.49 seconds
+staged, and 3.75 seconds private; aggregate GPU compute and the 264.24 GiB H2D
+payload were effectively unchanged. Ordinary TP communication remained, but
+the extra wait caused by staging-driven rank skew largely disappeared.
+
+All workers explicitly unregistered before releasing their mappings, the
+system's acquired-minus-released pin counter returned to its pre-run value,
+and every GPU allocation was released. Cold runtime-cache publication remained
+expensive because it still performs ordinary loading and multiple full-content
+passes; that startup problem is outside the registered transfer scope.

 ## Recommendations

-### AllGather
-
- Use **DP + DLO AllGather** when ranks already execute the same block path in
-  lockstep and belong to the same fast P2P domain. It shards persistent host
-  weights across that group.
+- Use **DP + DLO AllGather** for the supported throughput and host-memory
+  scaling path.
 - Use **SP + DLO AllGather** for long-sequence workloads when DP concurrency is
  not the goal.
-
-### No-AllGather
-
 - Use **no-AllGather** when independent replica execution is required. TP1
  direct-mmap deployments can share checkpoint pages per node. Enable the
-  host weight cache only when independent replicas have repeated final layouts
-  and benchmark the target CPU/memory topology, especially for matching
-  coordinates across multiple TP engines. With the default pinned-memory
-  policy, supported CUDA workers prefer registered direct H2D. Configure a
-  positive ceiling when an explicit page-locking bound is required, or disable
-  pinned memory when pageable staging is operationally preferable.
-
-### HSDP
-
- Do not combine HSDP with DLO AllGather or host weight cache v1. HSDP with
-  no-AllGather remains an ordinary-loader rank-local path.
+  runtime cache only when independent replicas have repeated final layouts and
+  benchmark the target CPU/memory topology, especially for matching
+  coordinates across multiple TP engines. On CUDA, provide an explicit
+  registration budget large enough for the complete final layout when
+  recurrent staging cost is unacceptable; leave it at zero when page-locking
+  that layout is not operationally acceptable.
+- Prefer **HSDP alone** for production HSDP deployments until the combined
+  HSDP + DLO no-AllGather path has broader end-to-end coverage.
@@ -190,15 +190,12 @@ configured parallelism strategy:
 - `WeightsMapper` transformations must be applied before matching ignored-layer
  lists or component prefixes.

-Distributed layerwise offload has an additional boundary: direct checkpoint
-mmap cannot prepare online-quantized tensors because it bypasses the ordinary
-loader that creates their runtime weights and scales. Per-tensor online FP8
-linears can still use DLO AllGather through the ordinary loader: each rank first
-finalizes the FP8 tensors, then DLO shards those runtime tensors for
-reconstruction during prefetch. Other online methods remain limited to
-`--dlo-no-use-allgather` until their packing and scale layouts are validated.
-See the [DLO design document](offloader/distributed_layerwise_offload.md) for
-the full compatibility matrix.
+Distributed layerwise offload has an additional boundary: its sharded mmap
+AllGather path is not compatible with online quantization because the path
+expects already materialized checkpoint weights. Use the rank-local
+`--dlo-no-use-allgather` path or a compatible pre-quantized checkpoint when
+combining these features. See the [DLO design document](offloader/distributed_layerwise_offload.md)
+for the full compatibility matrix.

 ## Adding a quantization backend

@@ -1,165 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" width="1280" height="850" viewBox="0 0 1280 850" role="img" aria-labelledby="title desc">
-  <title id="title">AllGather DLO sharded host weights and lockstep device reconstruction</title>
-  <desc id="desc">An external orchestrator submits one synchronized wave to a DP2 by TP2 diffusion engine. Each DP replica contains TP0 and TP1 workers with private pinned host shards and two rotating GPU weight slots. DLO AllGather groups connect matching TP coordinates across DP replicas, reconstructing full rank-local blocks on each device in lockstep. Ordinary TP groups remain within each DP replica.</desc>
-  <defs>
-    <marker id="request-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#7c3aed"/>
-    </marker>
-    <marker id="h2d-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#d97706"/>
-    </marker>
-    <marker id="collective-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto-start-reverse">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#7c3aed"/>
-    </marker>
-    <style>
-      .canvas { fill: #ffffff; stroke: #e2e8f0; }
-      .node { fill: #f8fafc; stroke: #cbd5e1; }
-      .engine { fill: #ffffff; stroke: #cbd5e1; }
-      .replica { fill: #f8fafc; stroke: #cbd5e1; }
-      .router { fill: #f5f3ff; stroke: #c4b5fd; }
-      .scope { fill: #f8fafc; stroke: #cbd5e1; }
-      .host { fill: #fff7ed; stroke: #fdba74; }
-      .shard { fill: #d97706; }
-      .gpu { fill: #f0fdf4; stroke: #86efac; }
-      .slot-current { fill: #16a34a; }
-      .slot-next { fill: #bbf7d0; stroke: #22c55e; }
-      .lockstep { fill: #f5f3ff; stroke: #c4b5fd; }
-      .title { font: 700 25px system-ui, -apple-system, Segoe UI, sans-serif; fill: #0f172a; }
-      .subtitle { font: 500 13px system-ui, -apple-system, Segoe UI, sans-serif; fill: #64748b; }
-      .section { font: 700 15px system-ui, -apple-system, Segoe UI, sans-serif; fill: #1e293b; }
-      .section-sub { font: 500 11.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #64748b; }
-      .eyebrow { font: 700 10px system-ui, -apple-system, Segoe UI, sans-serif; letter-spacing: .8px; fill: #64748b; }
-      .box-title { font: 700 12px system-ui, -apple-system, Segoe UI, sans-serif; fill: #1e293b; }
-      .box-text { font: 500 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #475569; }
-      .shard-title { font: 700 13px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .shard-text { font: 500 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .slot-title { font: 700 10px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .slot-text { font: 500 9px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .slot-next-title { font: 700 10px system-ui, -apple-system, Segoe UI, sans-serif; fill: #166534; }
-      .slot-next-text { font: 500 9px system-ui, -apple-system, Segoe UI, sans-serif; fill: #166534; }
-      .request-label { font: 600 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #6d28d9; }
-      .h2d-label { font: 600 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #b45309; }
-      .collective-label { font: 700 10px system-ui, -apple-system, Segoe UI, sans-serif; fill: #6d28d9; }
-      .lock-title { font: 700 12px system-ui, -apple-system, Segoe UI, sans-serif; fill: #6d28d9; }
-      .lock-text { font: 500 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #5b21b6; }
-      .request-path { fill: none; stroke: #7c3aed; stroke-width: 2; marker-end: url(#request-arrow); }
-      .h2d-path { fill: none; stroke: #d97706; stroke-width: 2; marker-end: url(#h2d-arrow); }
-      .collective-path { fill: none; stroke: #7c3aed; stroke-width: 2.5; marker-start: url(#collective-arrow); marker-end: url(#collective-arrow); }
-    </style>
-  </defs>
-
-  <rect x="12" y="12" width="1256" height="826" rx="14" class="canvas"/>
-
-  <text id="title-text" x="40" y="50" class="title">AllGather DLO: sharded host weights, lockstep reconstruction</text>
-  <text x="40" y="74" class="subtitle">DP2 &#215; TP2 shown: DLO groups span DP replicas per TP coordinate; ordinary TP stays within each replica</text>
-
-  <!-- The external control plane submits one synchronized DP wave. -->
-  <rect x="390" y="94" width="500" height="68" rx="10" class="router"/>
-  <text x="640" y="121" text-anchor="middle" class="section">External router / orchestrator</text>
-  <text x="640" y="144" text-anchor="middle" class="section-sub">submits one synchronized DP wave &#183; outside vLLM-Omni DLO</text>
-
-  <rect x="30" y="184" width="1220" height="634" rx="14" class="node"/>
-  <text x="52" y="211" class="eyebrow">ONE NODE &#183; ONE SYNCHRONIZED ENGINE, PROCESS-OWNED HOST SHARDS</text>
-
-  <path d="M 640 162 L 640 232" class="request-path"/>
-  <text x="654" y="202" class="request-label">request wave</text>
-
-  <!-- One engine contains two DP replicas that execute in lockstep. -->
-  <rect x="45" y="232" width="1190" height="262" rx="12" class="engine"/>
-  <text x="68" y="261" class="section">One synchronized diffusion engine &#183; DP2 &#215; TP2</text>
-  <text x="68" y="281" class="section-sub">two DP requests follow the same block order; DLO reuses existing DP groups at each TP coordinate</text>
-
-  <!-- DP replica 0. -->
-  <rect x="65" y="296" width="515" height="186" rx="10" class="replica"/>
-  <text x="82" y="318" class="box-title">DP replica 0 &#183; request 0</text>
-  <text x="563" y="318" text-anchor="end" class="section-sub">ordinary TP group</text>
-
-  <rect x="82" y="332" width="473" height="64" rx="8" class="gpu"/>
-  <text x="98" y="351" class="eyebrow">TP0 &#183; GPU 0</text>
-  <text x="98" y="376" class="box-text">DLO backend &#183; local shard H2D</text>
-  <rect x="250" y="340" width="132" height="44" rx="6" class="slot-current"/>
-  <text x="316" y="357" text-anchor="middle" class="slot-title">SLOT 0 &#183; CURRENT</text>
-  <text x="316" y="374" text-anchor="middle" class="slot-text">full TP-local N &#183; compute</text>
-  <rect x="397" y="340" width="142" height="44" rx="6" class="slot-next"/>
-  <text x="468" y="357" text-anchor="middle" class="slot-next-title">SLOT 1 &#183; PREFETCH</text>
-  <text x="468" y="374" text-anchor="middle" class="slot-next-text">full TP-local N+1 after AG</text>
-
-  <rect x="82" y="410" width="473" height="64" rx="8" class="gpu"/>
-  <text x="98" y="429" class="eyebrow">TP1 &#183; GPU 1</text>
-  <text x="98" y="454" class="box-text">DLO backend &#183; local shard H2D</text>
-  <rect x="250" y="418" width="132" height="44" rx="6" class="slot-current"/>
-  <text x="316" y="435" text-anchor="middle" class="slot-title">SLOT 0 &#183; CURRENT</text>
-  <text x="316" y="452" text-anchor="middle" class="slot-text">full TP-local N &#183; compute</text>
-  <rect x="397" y="418" width="142" height="44" rx="6" class="slot-next"/>
-  <text x="468" y="435" text-anchor="middle" class="slot-next-title">SLOT 1 &#183; PREFETCH</text>
-  <text x="468" y="452" text-anchor="middle" class="slot-next-text">full TP-local N+1 after AG</text>
-
-  <!-- DP replica 1. -->
-  <rect x="700" y="296" width="515" height="186" rx="10" class="replica"/>
-  <text x="717" y="318" class="box-title">DP replica 1 &#183; request 1</text>
-  <text x="1198" y="318" text-anchor="end" class="section-sub">ordinary TP group</text>
-
-  <rect x="725" y="332" width="473" height="64" rx="8" class="gpu"/>
-  <text x="741" y="351" class="eyebrow">TP0 &#183; GPU 2</text>
-  <text x="741" y="376" class="box-text">DLO backend &#183; local shard H2D</text>
-  <rect x="893" y="340" width="132" height="44" rx="6" class="slot-current"/>
-  <text x="959" y="357" text-anchor="middle" class="slot-title">SLOT 0 &#183; CURRENT</text>
-  <text x="959" y="374" text-anchor="middle" class="slot-text">full TP-local N &#183; compute</text>
-  <rect x="1040" y="340" width="142" height="44" rx="6" class="slot-next"/>
-  <text x="1111" y="357" text-anchor="middle" class="slot-next-title">SLOT 1 &#183; PREFETCH</text>
-  <text x="1111" y="374" text-anchor="middle" class="slot-next-text">full TP-local N+1 after AG</text>
-
-  <rect x="725" y="410" width="473" height="64" rx="8" class="gpu"/>
-  <text x="741" y="429" class="eyebrow">TP1 &#183; GPU 3</text>
-  <text x="741" y="454" class="box-text">DLO backend &#183; local shard H2D</text>
-  <rect x="893" y="418" width="132" height="44" rx="6" class="slot-current"/>
-  <text x="959" y="435" text-anchor="middle" class="slot-title">SLOT 0 &#183; CURRENT</text>
-  <text x="959" y="452" text-anchor="middle" class="slot-text">full TP-local N &#183; compute</text>
-  <rect x="1040" y="418" width="142" height="44" rx="6" class="slot-next"/>
-  <text x="1111" y="435" text-anchor="middle" class="slot-next-title">SLOT 1 &#183; PREFETCH</text>
-  <text x="1111" y="452" text-anchor="middle" class="slot-next-text">full TP-local N+1 after AG</text>
-
-  <!-- DLO collectives connect matching TP coordinates across DP replicas. -->
-  <text x="640" y="350" text-anchor="middle" class="collective-label">DLO AG &#183; TP0 GROUP</text>
-  <path d="M 555 364 L 725 364" class="collective-path"/>
-  <text x="640" y="428" text-anchor="middle" class="collective-label">DLO AG &#183; TP1 GROUP</text>
-  <path d="M 555 442 L 725 442" class="collective-path"/>
-
-  <!-- All participants must reach the same collective in the same order. -->
-  <rect x="365" y="516" width="550" height="48" rx="8" class="lockstep"/>
-  <circle cx="395" cy="540" r="12" fill="#7c3aed"/>
-  <text x="395" y="545" text-anchor="middle" font-family="system-ui, sans-serif" font-size="15" font-weight="700" fill="#ffffff">&#8644;</text>
-  <text x="420" y="536" class="lock-title">DLO AllGather requires lockstep block order</text>
-  <text x="420" y="553" class="lock-text">every participant enters each block collective in the same sequence</text>
-
-  <!-- Each replica copies two process-owned host shards before reconstruction. -->
-  <path d="M 455 650 C 412 625, 340 610, 340 575 L 340 540 C 340 516, 318 500, 300 482" class="h2d-path"/>
-  <text x="276" y="618" text-anchor="middle" class="h2d-label">local shard H2D &#215;2</text>
-  <path d="M 825 650 C 868 625, 940 610, 940 575 L 940 540 C 940 516, 962 500, 980 482" class="h2d-path"/>
-  <text x="1004" y="618" text-anchor="middle" class="h2d-label">local shard H2D &#215;2</text>
-
-  <!-- Persistent host shards are private to workers, not mmap-shared. -->
-  <rect x="58" y="650" width="242" height="140" rx="10" class="scope"/>
-  <text x="78" y="676" class="eyebrow">GROUP TOPOLOGY</text>
-  <text x="78" y="702" class="box-title">DP2 &#215; TP2</text>
-  <text x="78" y="725" class="box-text">DLO groups: same TP coordinate</text>
-  <text x="78" y="746" class="box-text">across DP replicas</text>
-  <text x="78" y="767" class="box-text">TP stays the model-parallel group.</text>
-
-  <rect x="330" y="650" width="620" height="140" rx="10" class="host"/>
-  <text x="350" y="676" class="eyebrow">PROCESS-OWNED PINNED HOST SHARDS &#183; NOT HOST WEIGHT CACHE MMAP</text>
-  <rect x="370" y="692" width="260" height="60" rx="8" class="shard"/>
-  <text x="500" y="716" text-anchor="middle" class="shard-title">DP replica 0 &#183; private</text>
-  <text x="500" y="738" text-anchor="middle" class="shard-text">TP0 shard 0/2 &#183; TP1 shard 0/2</text>
-  <rect x="650" y="692" width="260" height="60" rx="8" class="shard"/>
-  <text x="780" y="716" text-anchor="middle" class="shard-title">DP replica 1 &#183; private</text>
-  <text x="780" y="738" text-anchor="middle" class="shard-text">TP0 shard 1/2 &#183; TP1 shard 1/2</text>
-  <text x="640" y="773" text-anchor="middle" class="box-text">Across each DLO group, shards total approximately one rank-local runtime block copy</text>
-
-  <rect x="980" y="650" width="242" height="140" rx="10" class="scope"/>
-  <text x="1000" y="676" class="eyebrow">HOT PATH &#183; PER WORKER</text>
-  <text x="1000" y="702" class="box-title">Prefetch next block</text>
-  <text x="1000" y="725" class="box-text">local pinned shard &#8594; GPU</text>
-  <text x="1000" y="746" class="box-text">DLO AllGather &#8594; full block</text>
-  <text x="1000" y="767" class="box-text">rotate slots; overlap compute</text>
-</svg>
@@ -1,202 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" width="1280" height="850" viewBox="0 0 1280 850" role="img" aria-labelledby="title desc">
-  <title id="title">No-AllGather DLO shared host weights and independent runtime orchestration</title>
-  <desc id="desc">An external orchestrator routes requests to two independently scheduled TP2 diffusion engines on one node. Each engine contains two rank-local workers and devices, with private loaders, mmap views, DLO runtime hooks, streams, events, and rotating GPU slots. Equivalent TP0 workers across engines map one immutable TP0 entry in the host weight cache, while equivalent TP1 workers map a separate TP1 entry, without a cross-engine DLO AllGather.</desc>
-  <defs>
-    <marker id="request-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#7c3aed"/>
-    </marker>
-    <marker id="mapping-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#d97706"/>
-    </marker>
-    <marker id="runtime-arrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="7" markerHeight="7" orient="auto">
-      <path d="M 0 0 L 10 5 L 0 10 z" fill="#2563eb"/>
-    </marker>
-    <style>
-      .canvas { fill: #ffffff; stroke: #e2e8f0; }
-      .node { fill: #f8fafc; stroke: #cbd5e1; }
-      .engine { fill: #ffffff; stroke: #cbd5e1; }
-      .router { fill: #f5f3ff; stroke: #c4b5fd; }
-      .scope { fill: #f8fafc; stroke: #cbd5e1; }
-      .cache { fill: #fff7ed; stroke: #fdba74; }
-      .payload { fill: #d97706; }
-      .loader { fill: #f1f5f9; stroke: #cbd5e1; }
-      .mapping { fill: #fffbeb; stroke: #f59e0b; }
-      .hook { fill: #eff6ff; stroke: #60a5fa; }
-      .gpu { fill: #f0fdf4; stroke: #86efac; }
-      .device { fill: #ffffff; stroke: #86efac; }
-      .slot-current { fill: #16a34a; }
-      .slot-next { fill: #bbf7d0; stroke: #22c55e; }
-      .no-collective { fill: #fef2f2; stroke: #fca5a5; }
-      .title { font: 700 25px system-ui, -apple-system, Segoe UI, sans-serif; fill: #0f172a; }
-      .subtitle { font: 500 13px system-ui, -apple-system, Segoe UI, sans-serif; fill: #64748b; }
-      .section { font: 700 15px system-ui, -apple-system, Segoe UI, sans-serif; fill: #1e293b; }
-      .section-sub { font: 500 11.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #64748b; }
-      .eyebrow { font: 700 10px system-ui, -apple-system, Segoe UI, sans-serif; letter-spacing: .8px; fill: #64748b; }
-      .box-title { font: 700 12px system-ui, -apple-system, Segoe UI, sans-serif; fill: #1e293b; }
-      .box-text { font: 500 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #475569; }
-      .payload-title { font: 700 14px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .payload-text { font: 500 11px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .device-title { font: 700 9px system-ui, -apple-system, Segoe UI, sans-serif; fill: #166534; }
-      .slot-text { font: 700 8px system-ui, -apple-system, Segoe UI, sans-serif; fill: #ffffff; }
-      .slot-next-text { font: 700 8px system-ui, -apple-system, Segoe UI, sans-serif; fill: #166534; }
-      .arrow-label { font: 600 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #475569; }
-      .request-label { font: 600 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #6d28d9; }
-      .mapping-label { font: 600 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #b45309; }
-      .no-title { font: 700 12px system-ui, -apple-system, Segoe UI, sans-serif; fill: #b91c1c; }
-      .no-text { font: 500 10.5px system-ui, -apple-system, Segoe UI, sans-serif; fill: #7f1d1d; }
-      .request-path { fill: none; stroke: #7c3aed; stroke-width: 2; marker-end: url(#request-arrow); }
-      .mapping-path { fill: none; stroke: #d97706; stroke-width: 2; marker-end: url(#mapping-arrow); }
-      .runtime-path { fill: none; stroke: #2563eb; stroke-width: 2; marker-end: url(#runtime-arrow); }
-      .startup-path { fill: none; stroke: #94a3b8; stroke-width: 1.5; stroke-dasharray: 4 3; marker-end: url(#runtime-arrow); }
-    </style>
-  </defs>
-
-  <rect x="12" y="12" width="1256" height="826" rx="14" class="canvas"/>
-
-  <text id="title-text" x="40" y="50" class="title">No-AllGather DLO: TP-coordinate host sharing, independent engines</text>
-  <text x="40" y="74" class="subtitle">TP2 shown: each engine owns two devices; equivalent workers share one host payload per TP coordinate</text>
-
-  <!-- The control plane remains outside DLO. -->
-  <rect x="390" y="94" width="500" height="68" rx="10" class="router"/>
-  <text x="640" y="121" text-anchor="middle" class="section">External router / orchestrator</text>
-  <text x="640" y="144" text-anchor="middle" class="section-sub">independent request placement &#183; outside vLLM-Omni DLO</text>
-
-  <!-- One node contains independent engine runtimes over one shared physical backing. -->
-  <rect x="30" y="184" width="1220" height="634" rx="14" class="node"/>
-  <text x="52" y="211" class="eyebrow">ONE NODE &#183; PROCESS-LOCAL ENGINES OVER SHARED PHYSICAL HOST MEMORY</text>
-
-  <!-- Independent request routing from the external orchestrator. -->
-  <path d="M 500 162 C 456 181, 400 204, 332 232" class="request-path"/>
-  <text x="416" y="194" class="request-label">request A</text>
-  <path d="M 780 162 C 824 181, 880 204, 948 232" class="request-path"/>
-  <text x="864" y="194" class="request-label">request B</text>
-
-  <!-- Independent engine A. -->
-  <rect x="45" y="232" width="575" height="250" rx="12" class="engine"/>
-  <text x="68" y="261" class="section">Independent engine A</text>
-  <text x="68" y="281" class="section-sub">TP2 &#183; two rank-local workers &#183; own request schedule and DLO state</text>
-
-  <rect x="68" y="310" width="98" height="130" rx="8" class="loader"/>
-  <text x="117" y="335" text-anchor="middle" class="eyebrow">STARTUP &#215;2</text>
-  <text x="117" y="359" text-anchor="middle" class="box-title">rank-local</text>
-  <text x="117" y="377" text-anchor="middle" class="box-title">loaders</text>
-  <text x="117" y="402" text-anchor="middle" class="box-text">final transforms</text>
-  <text x="117" y="421" text-anchor="middle" class="box-text">publish / join</text>
-
-  <rect x="180" y="310" width="112" height="130" rx="8" class="mapping"/>
-  <text x="236" y="335" text-anchor="middle" class="eyebrow">HOST VIEWS &#215;2</text>
-  <text x="236" y="359" text-anchor="middle" class="box-title">TP0 mmap</text>
-  <text x="236" y="377" text-anchor="middle" class="box-title">TP1 mmap</text>
-  <text x="236" y="402" text-anchor="middle" class="box-text">process-local</text>
-  <text x="236" y="421" text-anchor="middle" class="box-text">optional registration</text>
-
-  <rect x="306" y="310" width="120" height="130" rx="8" class="hook"/>
-  <text x="366" y="335" text-anchor="middle" class="eyebrow">RUNTIME &#215;2</text>
-  <text x="366" y="359" text-anchor="middle" class="box-title">DLO backends</text>
-  <text x="366" y="377" text-anchor="middle" class="box-title">+ block hooks</text>
-  <text x="366" y="402" text-anchor="middle" class="box-text">private streams</text>
-  <text x="366" y="421" text-anchor="middle" class="box-text">events + state</text>
-
-  <rect x="440" y="310" width="157" height="130" rx="8" class="gpu"/>
-  <text x="518" y="330" text-anchor="middle" class="eyebrow">TWO DEVICES</text>
-  <rect x="450" y="340" width="137" height="42" rx="6" class="device"/>
-  <text x="518" y="353" text-anchor="middle" class="device-title">TP0 &#183; GPU A0</text>
-  <rect x="457" y="359" width="58" height="16" rx="4" class="slot-current"/>
-  <text x="486" y="370" text-anchor="middle" class="slot-text">SLOT 0 NOW</text>
-  <rect x="521" y="359" width="58" height="16" rx="4" class="slot-next"/>
-  <text x="550" y="370" text-anchor="middle" class="slot-next-text">SLOT 1 NEXT</text>
-  <rect x="450" y="389" width="137" height="42" rx="6" class="device"/>
-  <text x="518" y="402" text-anchor="middle" class="device-title">TP1 &#183; GPU A1</text>
-  <rect x="457" y="408" width="58" height="16" rx="4" class="slot-current"/>
-  <text x="486" y="419" text-anchor="middle" class="slot-text">SLOT 0 NOW</text>
-  <rect x="521" y="408" width="58" height="16" rx="4" class="slot-next"/>
-  <text x="550" y="419" text-anchor="middle" class="slot-next-text">SLOT 1 NEXT</text>
-
-  <path d="M 166 375 L 180 375" class="startup-path"/>
-  <path d="M 292 375 L 306 375" class="runtime-path"/>
-  <path d="M 426 375 L 440 375" class="runtime-path"/>
-
-  <!-- Independent engine B. -->
-  <rect x="660" y="232" width="575" height="250" rx="12" class="engine"/>
-  <text x="683" y="261" class="section">Independent engine B</text>
-  <text x="683" y="281" class="section-sub">TP2 &#183; two rank-local workers &#183; own request schedule and DLO state</text>
-
-  <rect x="683" y="310" width="98" height="130" rx="8" class="loader"/>
-  <text x="732" y="335" text-anchor="middle" class="eyebrow">STARTUP &#215;2</text>
-  <text x="732" y="359" text-anchor="middle" class="box-title">rank-local</text>
-  <text x="732" y="377" text-anchor="middle" class="box-title">loaders</text>
-  <text x="732" y="402" text-anchor="middle" class="box-text">final transforms</text>
-  <text x="732" y="421" text-anchor="middle" class="box-text">publish / join</text>
-
-  <rect x="795" y="310" width="112" height="130" rx="8" class="mapping"/>
-  <text x="851" y="335" text-anchor="middle" class="eyebrow">HOST VIEWS &#215;2</text>
-  <text x="851" y="359" text-anchor="middle" class="box-title">TP0 mmap</text>
-  <text x="851" y="377" text-anchor="middle" class="box-title">TP1 mmap</text>
-  <text x="851" y="402" text-anchor="middle" class="box-text">process-local</text>
-  <text x="851" y="421" text-anchor="middle" class="box-text">optional registration</text>
-
-  <rect x="921" y="310" width="120" height="130" rx="8" class="hook"/>
-  <text x="981" y="335" text-anchor="middle" class="eyebrow">RUNTIME &#215;2</text>
-  <text x="981" y="359" text-anchor="middle" class="box-title">DLO backends</text>
-  <text x="981" y="377" text-anchor="middle" class="box-title">+ block hooks</text>
-  <text x="981" y="402" text-anchor="middle" class="box-text">private streams</text>
-  <text x="981" y="421" text-anchor="middle" class="box-text">events + state</text>
-
-  <rect x="1055" y="310" width="157" height="130" rx="8" class="gpu"/>
-  <text x="1133" y="330" text-anchor="middle" class="eyebrow">TWO DEVICES</text>
-  <rect x="1065" y="340" width="137" height="42" rx="6" class="device"/>
-  <text x="1133" y="353" text-anchor="middle" class="device-title">TP0 &#183; GPU B0</text>
-  <rect x="1072" y="359" width="58" height="16" rx="4" class="slot-current"/>
-  <text x="1101" y="370" text-anchor="middle" class="slot-text">SLOT 0 NOW</text>
-  <rect x="1136" y="359" width="58" height="16" rx="4" class="slot-next"/>
-  <text x="1165" y="370" text-anchor="middle" class="slot-next-text">SLOT 1 NEXT</text>
-  <rect x="1065" y="389" width="137" height="42" rx="6" class="device"/>
-  <text x="1133" y="402" text-anchor="middle" class="device-title">TP1 &#183; GPU B1</text>
-  <rect x="1072" y="408" width="58" height="16" rx="4" class="slot-current"/>
-  <text x="1101" y="419" text-anchor="middle" class="slot-text">SLOT 0 NOW</text>
-  <rect x="1136" y="408" width="58" height="16" rx="4" class="slot-next"/>
-  <text x="1165" y="419" text-anchor="middle" class="slot-next-text">SLOT 1 NEXT</text>
-
-  <path d="M 781 375 L 795 375" class="startup-path"/>
-  <path d="M 907 375 L 921 375" class="runtime-path"/>
-  <path d="M 1041 375 L 1055 375" class="runtime-path"/>
-
-  <!-- The runtimes are deliberately not coupled by a DLO collective. -->
-  <rect x="365" y="516" width="550" height="44" rx="8" class="no-collective"/>
-  <circle cx="395" cy="538" r="12" fill="#dc2626"/>
-  <path d="M 389 532 L 401 544 M 401 532 L 389 544" stroke="#ffffff" stroke-width="2.2"/>
-  <text x="420" y="534" class="no-title">No cross-engine DLO AllGather</text>
-  <text x="420" y="551" class="no-text">shared host pages do not couple request order, progress, or failure domains</text>
-  <text x="640" y="592" text-anchor="middle" class="section-sub">Ordinary TP / SP model collectives, when configured, remain inside each engine</text>
-
-  <!-- Each engine maps both TP-coordinate payloads from the shared host cache. -->
-  <path d="M 455 650 C 412 625, 340 610, 340 570 L 340 525 C 340 478, 278 454, 236 440" class="mapping-path"/>
-  <text x="286" y="617" text-anchor="middle" class="mapping-label">TP0 + TP1 mappings</text>
-  <path d="M 825 650 C 868 625, 940 610, 940 570 L 940 525 C 940 478, 893 454, 851 440" class="mapping-path"/>
-  <text x="994" y="617" text-anchor="middle" class="mapping-label">TP0 + TP1 mappings</text>
-
-  <!-- One node-local physical backing, visually beneath its consumers. -->
-  <rect x="58" y="650" width="242" height="140" rx="10" class="scope"/>
-  <text x="78" y="676" class="eyebrow">SHARING SCOPE</text>
-  <text x="78" y="702" class="box-title">TP2: two coordinate entries</text>
-  <text x="78" y="725" class="box-text">one physical copy per coordinate</text>
-  <text x="78" y="746" class="box-text">A/TP0 + B/TP0 &#8594; TP0 entry</text>
-  <text x="78" y="767" class="box-text">A/TP1 + B/TP1 &#8594; TP1 entry</text>
-
-  <rect x="330" y="650" width="620" height="140" rx="10" class="cache"/>
-  <text x="350" y="676" class="eyebrow">NODE-LOCAL IMMUTABLE HOST WEIGHT CACHE &#183; ONE COPY PER TP COORDINATE</text>
-  <rect x="370" y="692" width="260" height="60" rx="8" class="payload"/>
-  <text x="500" y="717" text-anchor="middle" class="payload-title">TP0 physical payload</text>
-  <text x="500" y="738" text-anchor="middle" class="payload-text">shared by A/TP0 + B/TP0</text>
-  <rect x="650" y="692" width="260" height="60" rx="8" class="payload"/>
-  <text x="780" y="717" text-anchor="middle" class="payload-title">TP1 physical payload</text>
-  <text x="780" y="738" text-anchor="middle" class="payload-text">shared by A/TP1 + B/TP1</text>
-  <text x="640" y="773" text-anchor="middle" class="box-text">Identity includes final layout + TP coordinate; engine and DP replica are excluded</text>
-
-  <rect x="980" y="650" width="242" height="140" rx="10" class="scope"/>
-  <text x="1000" y="676" class="eyebrow">WHAT STAYS PRIVATE</text>
-  <text x="1000" y="702" class="box-title">Per worker / process</text>
-  <text x="1000" y="725" class="box-text">mmap object and validation state</text>
-  <text x="1000" y="746" class="box-text">optional host registration</text>
-  <text x="1000" y="767" class="box-text">hooks, streams, events, GPU slots</text>
-</svg>
@@ -18,8 +18,8 @@ validation_paths:
  - tests/config/**
 upstream_refs:
  - vllm.config
-last_reviewed: 2026-08-11
-last_verified_commit: e356708da4cd39992405686dacfc32b18fcccc7f
+last_reviewed: 2026-08-07
+last_verified_commit: 3d7fc3b9ba3cac88d579d4dc35b78b0b641675fc
 ---

 # vLLM-Omni configuration
@@ -39,66 +39,11 @@ The PR review proposal to capture environment-derived values in configuration
 objects at initialization is an open design question, not a current
 invariant.

-## Current structured-stage reuse behavior
-
-The current RFC #4021 implementation uses upstream vLLM config classes as the
-schema source for AR and generation stages:
-
- `OmniStageLoadConfig` inherits `vllm.config.LoadConfig`.
- `OmniStageCacheConfig` inherits `vllm.config.CacheConfig`.
- `OmniStageSchedulerConfig` inherits `vllm.config.SchedulerConfig`.
- `OmniStageParallelConfig` inherits `vllm.config.ParallelConfig`.
-
-This inheritance preserves the existing load, cache, scheduler, and parallel
-concern boundaries while reducing duplicate downstream declarations. It does
-not, by itself, make every inherited field an effective vLLM-Omni runtime
-option. Construction and runtime consumption are separate surfaces:
-
-| Surface | Current behavior |
-| --- | --- |
-| Structured schema | Upstream dataclass fields are inherited. Their defaults, default factories, and applicable Pydantic validation can participate when the structured object is constructed. |
-| Omni input ownership | Pipeline/deploy/CLI construction continues to accept only fields with an existing structured owner. Direct construction of an inherited sub-config can expose a wider upstream schema. |
-| Engine projection | For AR and generation stages, reusable fields are discovered by intersecting each upstream config dataclass with upstream `EngineArgs`. Only constructor-explicit values are emitted; inherited defaults remain deferred to terminal materialization. |
-| Terminal materialization | The engine-owning process constructs the final upstream `VllmConfig` and performs model-, platform-, rank-, port-, and backend-dependent initialization. |
-
-The generated projection maps include known upstream naming differences such
-as `cache_dtype` to `kv_cache_dtype`, `policy` to `scheduling_policy`, and
-`data_parallel_master_ip` to `data_parallel_address`. A field added to both an
-upstream concern config and `EngineArgs` therefore enters the AR/generation
-projection without requiring a second downstream allowlist. Explicit inherited
-fields that have no projection raise an error rather than being silently
-dropped.
-
-Ownership exclusions remain deliberate. Stage topology owns `scheduler_cls`,
-cache owns `disable_hybrid_kv_cache_manager`, and Omni runtime owns
-`distributed_executor_backend` and `worker_cls`; vLLM's private API-process
-fields are terminal internals. These exclusions prevent one input from being
-constructed or projected through two config concerns. Effective-engine-argument
-tests cover both the dynamic mapping and this exclusion boundary.
-
-`CompilationConfig` and `ProfilerConfig` are direct structured-stage fields:
-`stage.compilation_config` and `stage.profiler_config`. They accept mapping/YAML
-inputs and complete upstream value objects. Mapping inputs are validated and
-materialized as their concrete upstream types when the structured config is
-constructed, so downstream consumers see one resolved representation.
-Prebuilt upstream objects retain their type across the structured projection
-boundary.
-
-Quantization remains on its existing Omni-owned transport contract. This
-reuse step does not adopt upstream `QuantizationConfigArgs`, because the
-current engine-specific quantization split needs to be resolved separately.
-
-Diffusion keeps its existing engine-field ownership and projection surface.
-Shared Python inheritance must not be treated as evidence that an LLM-only
-parallel input is effective for a diffusion stage.
-
 ## Safe-change guide while deferred

 Changes should trace every affected producer to its runtime consumer and test
 the structured, legacy, CLI, and deployment paths that are actually supported.
-For inherited vLLM fields, tests should separately cover constructor behavior
-and effective `EngineArgs` projection. Do not infer a stable contract from
-this placeholder page.
+Do not infer a stable contract from this placeholder page.

 ## Promotion gate

@@ -20,19 +20,13 @@ an implementation contract has a user-facing workflow.
 | Choose serial, batched, step-wise, or streaming diffusion execution | [Execution Modes and Streaming](../user_guide/diffusion/execution_modes.md) | [Diffusion Continuous Batching](../design/feature/diffusion_continuous_batching.md), [Async Diffusion Output](../design/feature/async_diffusion_output.md) |
 | Reclaim stage memory without restarting the server | [Sleep Mode](sleep_mode.md) | Runtime lifecycle behavior is documented in the user guide |

-The [Prefill-Decode Disaggregation design reference](pd_disaggregation.md)
-describes the experimental Qwen3-Omni runtime topology and its current
-integration boundaries. It is not yet a supported launch recipe.
-
 Some runtime designs are deliberately not promoted as standalone User Guide
 features yet:

- Generic, OmniConnector-based
-  [Disaggregated Inference](../design/feature/disaggregated_inference.md) is not
-  yet a standalone User Guide feature. Its
+- [Disaggregated Inference](../design/feature/disaggregated_inference.md) and
  [OmniConnector implementations](../design/index.md#communication) describe
-  transport contracts, while practical connector configuration remains under
-  [Pipeline and deploy configurations](../configuration/stage_configs.md).
+  topology and transport contracts. Practical deployment configuration remains
+  under [Pipeline and deploy configurations](../configuration/stage_configs.md).
 - [Async Chunk](../design/feature/async_chunk.md) and
  [Async Omni Output Materialization](../design/feature/omni_async_output_materialization.md)
  are model- and pipeline-dependent. Use the selected model's deploy
@@ -1,140 +0,0 @@
-# Prefill-Decode (PD) Disaggregation (experimental)
-
-!!! warning "Experimental design reference"
-    PD disaggregation is only partially integrated on the current `main` branch.
-    The runtime contains PD detection and routing scaffolding, but there is no
-    supported deploy overlay or end-to-end validated launch recipe yet. Treat this
-    page as a description of the current design, not as production guidance.
-
-Prefill-decode disaggregation splits the Qwen3-Omni Thinker into two logical
-stages: a prefill worker that processes the prompt and produces KV cache, and a
-decode worker that imports that cache and generates tokens. The Talker and
-Code2Wav stages remain downstream; PD does not disaggregate the entire
-Qwen3-Omni pipeline.
-
-## Current design
-
-### Runtime topology
-
-```mermaid
-flowchart LR
-    client["Client request"] --> orchestrator["Omni Orchestrator"]
-    orchestrator -->|"original prompt<br/>max_tokens = 1"| prefill["Thinker prefill<br/>KV producer"]
-    prefill -->|"KV blocks"| connector[("vLLM KV connector")]
-    prefill -->|"KV routing metadata"| orchestrator
-    orchestrator -->|"original prompt<br/>remote KV parameters"| decode["Thinker decode<br/>KV consumer"]
-    connector -->|"remote KV load"| decode
-    decode -->|"generated text"| client
-```
-
-The orchestrator owns the request lifecycle, while the vLLM KV connector owns
-the bulk KV transfer. The original prompt is submitted to both Thinker stages:
-the decode worker still needs the request tokens and multimodal features to
-construct its request, while the imported KV cache prevents it from repeating
-the completed prefill computation.
-
-For the full Qwen3-Omni pipeline, splitting the Thinker adds one logical stage:
-
-| Stage | Role | Output or handoff |
-| --- | --- | --- |
-| 0 | Thinker prefill | Saves prompt KV and prompt-side multimodal output |
-| 1 | Thinker decode | Loads remote KV, generates text, and produces decode-side conditioning |
-| 2 | Talker | Converts combined Thinker conditioning into codec codes |
-| 3 | Code2Wav | Converts codec codes into audio |
-
-Text remains a final output of the decode stage, and audio remains a final
-output of Code2Wav.
-
-### Request lifecycle
-
-1. At startup, PD detection looks for one stage marked `is_prefill_only` and
-   one downstream stage marked `is_decode_only`. The decode stage must list the
-   prefill stage as an input source.
-2. The orchestrator clones the Thinker sampling parameters for prefill, forces
-   `max_tokens=1`, clears stop conditions, and sets producer-side
-   `kv_transfer_params`. This makes prefill finish by length after exporting KV.
-3. When prefill finishes, the orchestrator captures its KV routing metadata and
-   Qwen multimodal output under the original request ID.
-4. The orchestrator submits the original prompt to decode with consumer-side
-   `kv_transfer_params`. Those parameters identify the transfer, prefill engine,
-   bootstrap endpoint, and, when required by the connector, remote request.
-5. The decode engine imports the KV cache and continues generation. The normal
-   Omni stage path then carries Thinker conditioning through Talker and
-   Code2Wav.
-
-### Control and data paths
-
-PD uses two different connector layers. They solve different handoffs and are
-not interchangeable.
-
-| Path | Payload | Owner |
-| --- | --- | --- |
-| PD control path | Request ID, original prompt, `transfer_id`, bootstrap address, engine ID, and connector-specific request metadata | Omni Orchestrator |
-| PD data path | Thinker attention KV blocks | vLLM KV connector |
-| Downstream model path | Thinker hidden states and embeddings, Talker codec codes, and audio tensors | vLLM-Omni stage connectors and processors |
-
-### Detection and validation
-
-The current runtime accepts one logical prefill/decode pair per pipeline. Each
-stage may use tensor parallelism, but both sides of the pair must satisfy these
-invariants:
-
- both stages define `kv_transfer_config`
- prefill uses `kv_producer` or `kv_both`
- decode uses `kv_consumer` or `kv_both`
- both stages use the same KV connector
- `kv_buffer_device` and `kv_buffer_size` agree when set on both stages
- both stages use the same `tensor_parallel_size`
-
-These checks happen before request processing so an inconsistent PD topology
-fails during engine initialization.
-
-## What is implemented today
-
-| Area | Current state |
-| --- | --- |
-| Pair detection and topology validation | Implemented in the shared PD helper |
-| Prefill sampling preparation | Implemented for sync and async entrypoints |
-| Prefill-to-decode routing | Implemented in the current Orchestrator |
-| Deploy-based configuration | Not yet migrated; the supported `stages:` deploy schema cannot currently expand Qwen3-Omni into the four-stage PD topology |
-| Qwen3 Thinker-to-Talker handoff | Bridge state and embedding-merge helpers exist, but the merge is not wired into the live stage processor |
-| Connector compatibility | Incomplete; the Orchestrator currently requires `remote_request_id`, while the pinned Mooncake connector completes prefill without returning that metadata |
-| Validation coverage | PD entrypoint tests and the PD deploy overlay in online end-to-end tests are temporarily disabled during migration |
-| Performance | No non-regression or throughput/latency result is established for the current path |
-
-Because these pieces are still converging, the previous legacy `stage_args` YAML
-and `vllm serve` command have been removed. The serve CLI now accepts the
-deploy-based `stages:` schema, so retaining the legacy recipe would imply a
-launch path that current `main` rejects.
-
-## Configuration contract under development
-
-The intended deploy integration starts from the
-[default Qwen3-Omni deploy config](gh-file:vllm_omni/deploy/qwen3_omni_moe.yaml)
-and expands its frozen three-stage pipeline into the four stages shown above.
-A complete integration must:
-
- split the Thinker definition without duplicating the Talker or Code2Wav model
-  contracts
- remap downstream stage IDs, input sources, and stage connectors
- inject producer and consumer `kv_transfer_config` values into the two Thinker
-  stages
- allocate separate device sets for prefill and decode
- preserve both final outputs: decode text and Code2Wav audio
- provide connector-specific routing metadata without hard-coding one connector
-  contract
-
-The required number of GPUs depends on tensor-parallel sizes and whether Talker
-and Code2Wav share a device. A basic TP=1 layout uses one device for prefill, one
-for decode, and one for the two downstream audio stages.
-
-## Implementation map
-
- PD detection, validation, and sampling-parameter preparation:
-  [`PDDisaggregationMixin`](gh-file:vllm_omni/entrypoints/pd_utils.py)
- startup detection and connector metadata:
-  [`AsyncOmniEngine`](gh-file:vllm_omni/engine/async_omni_engine.py)
- request-scoped prefill-to-decode routing:
-  [`Orchestrator`](gh-file:vllm_omni/engine/orchestrator.py)
- Qwen3 downstream conditioning helpers:
-  [Qwen3-Omni stage input processor](gh-file:vllm_omni/model_executor/stage_input_processors/qwen3_omni.py)
@@ -53,7 +53,6 @@ th {
 | `LingBotVideoPipeline` | LingBot-Video dense and MoE T2I, T2V, TI2V | `robbyant/lingbot-video-dense-1.3b`, `robbyant/lingbot-video-moe-30b-a3b` | ✅︎ | | | | — |
 | `MiniMaxH3Pipeline` | MiniMax H3 T2VA, first/last-keyframe FL2VA, and mixed-reference Ref2VA | `MiniMaxAI/MiniMax-H3` | ✅︎ | ✅︎ | | | [Published](https://recipes.vllm.ai/MiniMaxAI/MiniMax-H3) |
 | `DreamZeroPipeline` | DreamZero-DROID | `GEAR-Dreams/DreamZero-DROID` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
-| `Pi0Pipeline` | π0 (Pi-Zero) VLA | `lerobot/pi0_base` | ✅︎ | | | | [Repository](https://github.com/vllm-project/vllm-omni/blob/main/recipes/lerobot/Pi0.md) |
 | `HeliosPipeline`, `HeliosPyramidPipeline` | Helios | `BestWishYsh/Helios-Base`, `BestWishYsh/Helios-Mid`, `BestWishYsh/Helios-Distilled` | ✅︎ | ✅︎ | ✅︎ | | — |
 | `MagiHumanPipeline` | MagiHuman | `SII-GAIR/daVinci-MagiHuman-Base-1080p` | ✅︎ | ✅︎ | | | — |
 | `OvisImagePipeline` | Ovis-Image | `OvisAI/Ovis-Image` | ✅︎ | ✅︎ | | ✅︎ | — |
@@ -78,7 +77,6 @@ th {
 | `SoulXSingerPipeline` | SoulX-Singer (SVS) | `Soul-AILab/SoulX-Singer` | ✅︎ | | | | — |
 | `SoulXSingerSVCPipeline` | SoulX-Singer-SVC | `Soul-AILab/SoulX-Singer` (`model-svc.pt`) | ✅︎ | | | | — |
 | `AudioXPipeline` | AudioX | `zhangj1an/AudioX` | ✅︎ | ✅︎ | | | — |
-| `MiniMaxMusic3ForConditionalGeneration` | MiniMax Music 3 (text-to-music) | `MiniMaxAI/MiniMax-Music3` | ✅︎ | | | | — |
 | `Qwen3TTSForConditionalGeneration` | Qwen3-TTS-12Hz-1.7B-CustomVoice | `Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice` | ✅︎ | | | | [Published](https://recipes.vllm.ai/Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice) |
 | `Qwen3TTSForConditionalGeneration` | Qwen3-TTS-12Hz-1.7B-VoiceDesign | `Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
 | `Qwen3TTSForConditionalGeneration` | Qwen3-TTS-12Hz-1.7B-Base | `Qwen/Qwen3-TTS-12Hz-1.7B-Base` | ✅︎ | ✅︎ | ✅︎ | ✅︎ | — |
@@ -141,20 +141,7 @@ The response shape depends on the streaming parameters:

 **Non-streaming (default).** With `stream=false` and no `stream_format`, returns the
 complete clip as binary audio data with an appropriate `Content-Type` header (e.g.
-`audio/wav`). Because the raw-bytes body has no JSON carrier, successful
-non-streaming responses from non-diffusion speech servers report usage through
-response headers:
-
-| Header | Description |
-| --- | --- |
-| `x-vllm-omni-input-tokens` | Total input tokens (`text_tokens` + `audio_tokens`). |
-| `x-vllm-omni-output-tokens` | Generated codec/audio tokens. |
-| `x-vllm-omni-total-tokens` | `input_tokens` + `output_tokens`. |
-| `x-vllm-omni-input-text-tokens` | Tokens from the synthesized text (`input` plus `instructions`). |
-| `x-vllm-omni-input-audio-tokens` | Reference-audio codec frames, non-zero only for in-context voice cloning. |
-
-Diffusion-mode speech servers route through a separate response path and do not
-emit these headers.
+`audio/wav`). The raw-bytes body has no JSON carrier, so no `usage` is reported.

 **Raw audio stream** (`stream_format="audio"`). Streams raw audio bytes (PCM or
 WAV) as they are decoded.
@@ -167,14 +167,6 @@ transfer options such as `control_path`, `control_guidance`,
 `num_conditional_frames`, `show_control_condition`, and `show_input`; see the
 Cosmos3 recipe for complete examples.

-HTTP redirects for `image_reference.image_url` follow vLLM's
-`VLLM_MEDIA_URL_ALLOW_REDIRECTS` setting. Before starting the server, set it to
-`1` (the default) to allow redirects or `0` to reject them. A redirect target
-can differ from the original host, and vLLM-Omni does not yet fully implement
-upstream vLLM's media URL allowlist protection. Deployments should therefore
-treat remote media URLs as untrusted and choose this setting as part of their
-URL access policy.
-
 ### Speech-to-Video

 For models that support audio-driven generation (e.g., Wan2.2-S2V), pass both
@@ -1,127 +1,262 @@
 # Distributed Layerwise Offloading

-Distributed layerwise offloading (DLO) streams DiT blocks through two device
-buffers instead of keeping the complete DiT in HBM. It supports sharded host
-weights with AllGather or independently streamed rank-local weights. The
-no-AllGather path can optionally share node-local host weights through mmap.
+Distributed layerwise offloading (DLO) extends block streaming to multi-device
+deployments. With AllGather enabled, each rank stores roughly `1 / dp_size` of
+the host weights and reconstructs each layer at runtime. Without AllGather,
+each rank streams a complete block independently. Compatible TP1 deployments
+can share checkpoint-backed host pages among processes on the same node;
+otherwise DLO streams the ordinary loader's rank-local tensors. The optional
+runtime-weight cache extends node-local sharing to final ordinary-loader
+layouts, including matching TP coordinates across independent engines. On
+CUDA, an explicit registration budget can make those shared mappings directly
+H2D-copyable and avoid recurrent mmap-to-pinned host packing.

-See the [shared DLO component architecture](../../../design/feature/offloader/distributed_layerwise_offload.md#architecture)
-for ownership and lifecycle coordination. The mode-specific internals are
-documented separately in the
-[AllGather design](../../../design/feature/offloader/distributed_layerwise_offload.md#allgather-design)
-and
-[no-AllGather design](../../../design/feature/offloader/distributed_layerwise_offload.md#no-allgather-design).
+See the [DLO feature design](../../../design/feature/offloader/distributed_layerwise_offload.md)
+for the implementation contract and compatibility matrix.

-## Choose a mode
+## Execution model

-| Mode | Use when | Host weights | Runtime synchronization |
-| --- | --- | --- | --- |
-| DLO AllGather (default) | DP ranks execute the same block path in lockstep | About `1 / dp_size` per rank | DLO weight AllGather |
-| no-AllGather | Ranks or engines must schedule independently | Complete rank-local layout | No DLO weight collective |
-| no-AllGather + host weight cache | Equivalent independent workers share one node | Shared final mmap layout per TP coordinate | No DLO weight collective |
-| host weight cache + host registration | The platform supports registration and recurrent staging is too expensive | Shared registered mmap layout | No DLO weight collective |
+DLO overlaps three operations with a fixed two-block device buffer:

-AllGather is normally the best choice for synchronized DP. Host weight cache mode
-targets independently scheduled replicas, especially repeated TP engines on
-one node.
+```text
+Compute stream:  [Layer N]          [Layer N+1]        [Layer N+2]
+H2D stream:      [H2D shard N+1]    [H2D shard N+2]
+AllGather:       [AG N+1]           [AG N+2]
+Slots:           slot 0: Layer N    slot 1: Layer N+1
+```
+
+AllGather communicates request-independent weight shards, but every member of
+the transfer group must request the same next block in the same collective
+order. A synchronized DP wave may contain different prompts only when those
+requests follow the same denoising and block-execution path.

 ## Usage

 ```bash
-# Sharded host weights with DLO AllGather
+# Four ranks with sharded host weights and AllGather
 vllm serve /path/to/model --omni \
  --enable-distributed-layerwise-offload \
  --data-parallel-size 4

-# Independently streamed rank-local weights
+# Standard-loader rank-local weights, without DLO AllGather
 vllm serve /path/to/model --omni \
  --enable-distributed-layerwise-offload \
+  --data-parallel-size 4 \
  --dlo-no-use-allgather

-# Share final host weights; prefer registered direct H2D under the pinned-memory policy
-vllm serve /path/to/model --omni \
-  --enable-distributed-layerwise-offload \
-  --dlo-no-use-allgather \
-  --dlo-use-host-weight-cache
-
-# Apply an 80 GiB per-worker safety ceiling to the same registration attempt
+# Independently scheduled replicas sharing final runtime weights on one node
 vllm serve /path/to/model --omni \
  --enable-distributed-layerwise-offload \
  --tensor-parallel-size 2 \
  --dlo-no-use-allgather \
-  --dlo-use-host-weight-cache \
-  --dlo-host-weight-cache-pin-limit-gib 80
+  --dlo-enable-runtime-cache \
+  --dlo-runtime-cache-dir /var/cache/vllm-omni/dlo-runtime-weights \
+  --dlo-runtime-cache-pin-limit-gib 80
+
+# Sequence parallel deployment
+vllm serve /path/to/model --omni \
+  --enable-distributed-layerwise-offload \
+  --usp 4
 ```

-## Relevant options
+```python
+from vllm_omni import Omni
+
+omni = Omni(
+    model="/path/to/model",
+    enable_distributed_layerwise_offload=True,
+    dlo_use_allgather=False,
+    dlo_enable_runtime_cache=True,
+    dlo_runtime_cache_dir="/var/cache/vllm-omni/dlo-runtime-weights",
+    dlo_runtime_cache_pin_limit_gib=80,
+)
+```
+
+## Flags

 | Flag | Meaning | Default |
 | --- | --- | --- |
 | `--enable-distributed-layerwise-offload` | Enable DLO | `false` |
-| `--dlo-no-use-allgather` | Stream complete rank-local blocks independently | `false` |
-| `--dlo-use-host-weight-cache` | Use the final-layout cache and prefer registration when pinned memory is enabled | `false` |
-| `--dlo-host-weight-cache-pin-limit-gib GIB` | Optional per-worker registration ceiling; zero means no additional ceiling | `0` |
-| `--dlo-resident-layers N` | Keep eligible leading DiT blocks resident in HBM | `0` |
+| `--data-parallel-size N` | DP ranks and AllGather weight-sharding group | `1` |
+| `--dlo-use-allgather` | Shard host weights and reconstruct with AllGather | `true` |
+| `--dlo-no-use-allgather` | Stream complete rank-local blocks without a DLO weight collective | `false` |
+| `--dlo-enable-runtime-cache` | Normalize final ordinary-loader DiT weights into a node-local mmap cache; requires no-AllGather | `false` |
+| `--dlo-runtime-cache-dir PATH` | Shared local-disk cache root | `~/.cache/vllm-omni/dlo-runtime-weights` |
+| `--dlo-runtime-cache-lock-timeout SECONDS` | Maximum wait for another process publishing the same layout | `600` |
+| `--dlo-runtime-cache-pin-limit-gib GIB` | Per-worker limit for registering the complete read-only runtime-cache mapping for direct H2D; zero keeps bounded host staging | `0` |
+| `--dlo-resident-layers N` | Keep N leading main-DiT blocks on device; requires no-AllGather and model-declared resident paths | `0` |

-No-AllGather DLO automatically uses a compatible checkpoint mmap plan. When
-checkpoint mmap is unavailable, it preserves ordinary pinned loader weights by
-default. `--dlo-use-host-weight-cache` explicitly selects the transform-complete
-final-layout cache. Under the default `pin_cpu_memory=True` policy, DLO attempts
-to register the complete immutable mapping for direct H2D. Setting
-`pin_cpu_memory=False` through structured or programmatic configuration retains
-the staged mmap path instead. A positive pin limit is only an optional safety
-ceiling and does not select the cache. The cache uses
-`~/.cache/vllm-omni/dlo-host-weights`. Programmatic configuration may override
-`dlo_host_weight_cache_dir` and the writer lock timeout; these advanced storage
-controls are intentionally not separate CLI flags.
+## Host-weight loading

-## Operational notes
+The diffusion loader chooses host storage before DLO is enabled. It first
+attempts to build a complete, validated direct-checkpoint mmap plan. If names,
+coverage, shape, dtype, topology, or loader-callback compatibility cannot be
+proven, it runs the ordinary model loader instead. DLO consumes that result and
+does not make a second checkpoint-compatibility decision.

- All workers that should share must see the same local, disk-backed cache
-  directory. Do not use tmpfs or a cross-node filesystem for this version.
- Cache entries are immutable and validated before use. A cache or registration
-  failure keeps the ordinary loader weights or falls back to bounded staging.
- The bounded-staging fallback is a memory-first mode: recurrent pageable-mmap to
-  pinned-slot copies can reduce throughput, especially across concurrent TP
-  engines. Use it intentionally by disabling pinned host memory, or when the
-  active platform cannot register the immutable mapping.
- When a positive registration ceiling is configured, it must cover the
-  complete page-aligned mapping reported in the worker log. Registration is
-  all-or-nothing. Zero applies no additional ceiling.
- CUDA is the first registration backend and requires read-only host
-  registration support for immutable cache mappings. Platforms or devices
-  without that capability continue to use bounded staging; pinned-memory policy
-  requests direct H2D but does not guarantee it.
- Registration is process-local but does not duplicate the underlying file
-  pages. Each worker must still satisfy its platform and OS page-locking limits.
- Shutdown unregisters host ranges before closing their mmap handles.
- Host weight cache v1 has no automatic eviction. Stop all users of an entry
-  before deleting it.
+The shared-mmap optimization in this phase is supported only with TP1. TP
+greater than one falls back before model mutation to ordinary TP-aware loading.
+DLO may still consume those TP-local tensors, but this is a compatibility path:
+it does not share checkpoint-backed runtime weights across DP replicas and
+provides no shared-mmap host-memory guarantee.

-## Scheduling constraints
+When `--dlo-enable-runtime-cache` is set with no-AllGather, ordinary loading
+gains a second storage opportunity. Every rank completes loader callbacks,
+post-load processing, validation, and calibration. It then hashes the final
+CPU DiT parameters and persistent buffers. One process per equivalent runtime
+layout publishes sharded safetensors plus a checksummed manifest; peers
+validate and mmap the same files. DLO replaces only tensor storage, preserves
+the final `Parameter` objects, and releases the private allocator-backed copies
+before installing its streaming hooks.

-AllGather ranks must request blocks in the same collective order. Concurrent DP
-requests may use different prompts, but they must follow the same denoising and
-block-execution path and set the same explicit `num_inference_steps`.
+A positive `--dlo-runtime-cache-pin-limit-gib` deliberately selects this final
+runtime layout even when a direct TP1 checkpoint plan is available. Raw
+checkpoint mappings may require deferred loader transforms such as grouped-QKV
+reordering; registering them would still leave recurrent host work. Publishing
+the fully transformed layout once makes every streamed source directly
+copyable.

-No-AllGather workers do not have this DLO lockstep requirement. Ordinary TP or
-SP model collectives still synchronize ranks within each engine.
+The runtime layout identity excludes DP rank and SP rank. It includes TP world
+size/rank and conservative SP implementation/world-size guards. Therefore two
+independent `TP=2` engines normally create two entries: both TP0 workers map
+one entry and both TP1 workers map the other. Cache sharing does not create a
+process group or an inference-time collective.
+
+With a zero registration budget, cached tensors remain pageable mmap views.
+DLO packs each streamed block into one of two bounded pinned host slots before
+the asynchronous H2D copy. This lowers steady-state PSS but adds recurrent
+host-to-host copies; competing TP ranks or engines can turn DDR contention and
+staging skew into collective wait time.
+
+With a positive budget on CUDA, DLO instead attempts to register the complete
+page-aligned cache mapping as read-only pinned host memory. Registration is
+all-or-nothing: the total mapped range must fit the per-worker budget, and any
+CUDA registration or verification failure rolls back partial registrations and
+uses the bounded staging path. If CUDA cannot roll back a partial registration,
+worker initialization aborts instead of unmapping memory that CUDA may still
+own. On success, each immutable tensor view is copied directly into the
+existing rotating device buffers on the H2D stream, so no private host staging
+slots are allocated.
+
+CUDA registration is process-local, but it does not clone file contents. The
+same physical page-cache pages remain shared among equivalent workers; each
+worker does, however, hold its own registration reference and must satisfy the
+host's page-locking limits. Shutdown synchronizes outstanding GPU work,
+unregisters every range, and only then closes the mmap handles.
+
+The mmap plan skips only dedicated DiT weight sources. Other component sources,
+such as a text encoder loaded through the shared diffusion loader, continue to
+use their ordinary component loader. A checkpoint source that mixes DiT and
+non-DiT weights falls back completely rather than leaving an unplanned
+component uninitialized.
+
+With direct checkpoint mmap, the loader:
+
+1. saves non-persistent buffers such as RoPE frequencies;
+2. moves the normally created transformer to the meta device;
+3. loads checkpoint tensors as mmap views backed by the shared OS page cache;
+4. applies any loader-owned bounded layout adapters while packing blocks;
+5. restores saved non-persistent buffers; and
+6. preserves `post_load_weights()` and `validate_loaded_weights()` lifecycle
+   hooks.
+
+For AllGather with a group larger than one, each process copies only its
+persistent shard and then releases the source mapping. For no-AllGather, each
+process keeps the mapping open and packs complete blocks through two bounded
+pinned staging slots. Processes mapping the same files on one node share the
+immutable pages; no-AllGather still performs a complete-block H2D copy in each
+process.
+
+When the effective DLO group size is one, `dlo_use_allgather=True` does not
+perform a collective and uses the same rank-local transfer behavior.
+
+## Runtime-cache lifecycle
+
+Runtime-cache publication uses a per-layout POSIX file lock, a temporary
+directory on the same filesystem, file and final-tensor SHA-256 validation,
+`fsync`, and an atomic directory rename. A crashed writer releases its kernel
+lock; the next writer removes only that layout's stale temporary directories
+while holding the lock. Lock timeout, unsupported layout, insufficient disk,
+or validation failure keeps the already-valid ordinary tensors.
+
+The cache root is deliberately independent of `VLLM_CACHE_ROOT`, because
+vLLM-Omni may isolate that variable between stage replicas. All consumers that
+should share must receive the same `--dlo-runtime-cache-dir` on a local,
+disk-backed filesystem. Do not place the cache on tmpfs when the goal is lower
+host-memory PSS, and do not use a cross-node directory in this version.
+
+There is no automatic eviction. Budget roughly one final DiT copy per distinct
+TP coordinate, dtype/layout, and model version, plus temporary space while an
+entry is published. Operators may remove entries or the complete cache root
+only after all workers using its mmap files have stopped. The feature is
+opt-in for this reason.
+
+For direct H2D, set the registration limit above the complete page-aligned
+layout size reported in the worker log. Partial registration is intentionally
+unsupported because mixing direct and staged tensors would complicate buffer
+lifetime and make performance topology-dependent. Registration is CUDA-only;
+an insufficient budget, unsupported runtime, OS/CUDA pinning limit, or
+registration error is reported and falls back to bounded staging.
+
+## Declarative topology
+
+Models may declare an `OffloadPlan` instead of embedding offload logic:
+
+```python
+from vllm_omni.diffusion.offloader import OffloadPlan
+
+
+class MyPipeline(nn.Module):
+    _dit_modules = ["transformer"]
+    _offload_plan = OffloadPlan(
+        block_attrs={"transformer": ("blocks",)},
+        offload_submodules={"context_encoder": "layers"},
+    )
+```
+
+When no plan exists, discovery falls back to
+`_layerwise_offload_blocks_attrs` and then heuristic attribute lookup.
+
+## Data-parallel concurrency
+
+With `data_parallel_size > 1` and AllGather enabled, the scheduler can process
+up to `dp_size` requests per denoising step. Every concurrent request must set
+the same explicit `num_inference_steps`; `None` is rejected because every rank
+must enter each collective.

 ## Limitations

- Direct checkpoint mmap currently requires TP1. The host weight cache can share
-  final ordinary-loader layouts between matching TP coordinates at TP1 or TP>1.
- Per-tensor online FP8 linears use the ordinary loader and can run with either
-  DLO transfer path. Other online quantization methods require no-AllGather
-  until their runtime layouts are validated.
- Host weight cache v1 rejects quantized, non-contiguous, aliased/tied,
-  device-only, HSDP, expert-parallel, CFG-parallel, and PP layouts.
- HSDP plus DLO AllGather is unsupported, and host weight cache v1 rejects HSDP.
- The host weight cache changes steady-state host backing, not startup loading:
-  each worker still performs ordinary loading and full content validation.
- Skip-load-on-hit, cache eviction, and cross-node sharing remain follow-up
-  work in
-  [RFC #6195](https://github.com/vllm-project/vllm-omni/issues/6195).
+- Direct checkpoint mmap currently requires TP1. TP greater than one falls
+  back to the ordinary TP-aware loader. With the optional runtime cache,
+  equivalent processes at the same TP coordinate can then share that final
+  rank-local layout in no-AllGather mode. A positive registration budget also
+  chooses the final runtime cache at TP1 so deferred checkpoint transforms are
+  not repeated during inference.
+- HSDP plus AllGather is rejected to avoid double sharding. HSDP without
+  AllGather has limited end-to-end validation.
+- Online quantization uses the ordinary loader with no-AllGather. It remains
+  incompatible with DLO AllGather and runtime-cache v1. Runtime-cache v1 also
+  rejects other quantized, non-contiguous, aliased/tied, device-only, HSDP,
+  expert-parallel, CFG-parallel, and PP layouts.
+- Resident leading layers require `--dlo-no-use-allgather` and a model
+  `OffloadPlan` that declares eligible `resident_dit_paths`.
+- DP concurrency requires an explicit, identical inference-step count.
+
+The runtime cache improves steady-state host PSS, not startup peak: every rank
+loads private weights before remapping, and validation adds full content-hash
+passes. Direct H2D removes recurrent staging but does not change that startup
+contract. Skip-load-on-cache-hit and cache eviction remain follow-up work in
+[RFC #6195](https://github.com/vllm-project/vllm-omni/issues/6195).
+
+Zero-budget staging is also not latency-neutral. In the four-L20X MiniMax-H3
+smoke documented in the L20X runtime-cache matrix in the
+[feature design](../../../design/feature/offloader/distributed_layerwise_offload.md#l20x-runtime-cache-smoke-matrix),
+two independent TP2 engines reduced request-time host PSS by 36.6% with the
+cache, but recurrent mmap-to-pinned staging reduced combined throughput by
+69.3%. The registered path recovered near-private per-engine latency while
+retaining the shared-memory benefit; a matched TP1 run improved mean wave
+latency from 33.87 seconds to 4.88 seconds with byte-identical outputs. See the
+feature design for the full smoke-test caveats and metrics.

 See the [Cosmos3 DistOffload recipe](https://github.com/vllm-project/vllm-omni/blob/main/recipes/cosmos3/Cosmos3-DistOffload.md)
 for an end-to-end example.
@@ -65,7 +65,7 @@ More failure modes (for example, network jitter and network interruption) will b

 ## Source of Conclusions

-The behaviors and conclusions above are summarized from current fault injection validation results on **`Qwen3-Omni`**, **`Wan2.2`**, and **`HunyuanImage-3.0-Instruct`** (DiT-only, `/v1/images/generations`). Automated coverage lives under `tests/dfx/reliability/` (`test_reliability_qwen3_omni.py`, `test_reliability_wan22.py`, `test_reliability_hunyuan_image.py`) and runs weekly via `.buildkite/cuda/test-weekly.yml`.
+The behaviors and conclusions above are summarized from current fault injection validation results on **`Qwen3-Omni`**, **`Wan2.2`**, **`HunyuanImage-3.0-Instruct`** (DiT-only, `/v1/images/generations`), and **`VoxCPM2`** (`/v1/audio/speech`). Automated coverage lives under `tests/dfx/reliability/` (`test_reliability_qwen3_omni.py`, `test_reliability_wan22.py`, `test_reliability_hunyuan_image.py`, `test_reliability_voxcpm2.py`) and runs weekly via `.buildkite/cuda/test-weekly.yml`.

 ### Example: `SIGTERM` fault injection log (Qwen3-Omni)

@@ -179,24 +179,6 @@ python text_to_image.py --model <your-model> --quantization fp8 --ignored-layers
 vllm serve <your-model> --omni --quantization fp8
 ```

-### Distributed layerwise offload
-
-Per-tensor online FP8 linears can be combined with DLO's default AllGather path:
-
-```bash
-vllm serve <your-model> \
-  --omni \
-  --quantization fp8 \
-  --enable-distributed-layerwise-offload
-```
-
-Direct checkpoint mmap is not used for this combination. Each rank runs the
-ordinary loader to create the final FP8 weights and scales before DLO shards
-them, so startup temporarily materializes the complete FP8 model in host memory
-on every rank. Runtime DLO host residency is sharded normally. Use
-`--dlo-no-use-allgather` when the synchronized AllGather request-wave contract
-does not fit the workload.
-
 ## Parameters

 | Parameter | Type | Default | Description |
@@ -1,104 +0,0 @@
-# π0 (Pi-Zero) VLA — OpenPI realtime serving
-
-[π0](https://www.physicalintelligence.company/blog/pi0) is a Vision-Language-Action
-model from Physical Intelligence: multi-camera images + a language instruction +
-robot proprioceptive state → a continuous action chunk via flow-matching denoising
-(it does **not** emit text tokens). This example serves π0 over the OpenPI realtime
-websocket protocol at `/v1/realtime/robot/openpi`.
-
-## Install extras
-
-The core `pip install -e .` does not include the OpenPI client used here:
-
- `openpi-client` (from the [openpi](https://github.com/Physical-Intelligence/openpi)
-  repo: `pip install -e packages/openpi-client`), `websockets`, `msgpack`, `msgpack-numpy`
-
-## Weights
-
-π0 uses the LeRobot `lerobot/pi0_base` checkpoint (HF, ~13 GB). Either let the server
-download it (`MODEL=lerobot/pi0_base`) or point at a local copy
-(`MODEL=/path/to/pi0_base`).
-
-## Run the server
-
-```bash
-vllm serve lerobot/pi0_base --omni --port 8000 \
-    --served-model-name pi0 \
-    --deploy-config vllm_omni/deploy/pi0.yaml \
-    --enforce-eager --disable-log-stats
-```
-
-The deploy config (`vllm_omni/deploy/pi0.yaml`) declares a single diffusion stage
-(`Pi0Pipeline`), float32, `max_num_seqs: 1`, and the `policy_server_config` handshake
-metadata (3 cameras, `joint_position`, action horizon 50, action dim 32).
-
-### Commonly adjusted `model_config` keys
-
-These live under `stages[0].model_config` in `vllm_omni/deploy/pi0.yaml`:
-
-| Key | Default | Meaning |
-|---|---|---|
-| `chunk_size` | `50` | Action-chunk length (timesteps) the model predicts per inference. |
-| `num_inference_steps` | `10` | Flow-matching Euler denoising steps. |
-| `max_action_dim` | `32` | Action dimensionality (state/action are padded to this). |
-| `max_state_dim` | `32` | Proprioceptive-state dimensionality (zero-padded to this). |
-| `image_resolution` | `[224, 224]` | Per-camera input size (square; SigLIP). |
-| `tokenizer_max_length` | `48` | Max PaliGemma prompt tokens. |
-| `max_cameras` | `3` | Camera slots the model attends to (real + `-1`-padded). |
-| `image_feature_keys` | 3 `observation.images.*` keys | Camera order the model attends to. |
-| `image_key_map` | `{}` | Map raw obs camera keys → `image_feature_keys` (empty = verbatim). |
-
-The `policy_server_config` block below them is the OpenPI handshake metadata
-advertised to the client; keep its `action_horizon` / `action_dim` /
-`image_resolution` in sync with the `model_config` values above. The e2e test
-`tests/e2e/online_serving/test_pi0_expansion.py::test_pi0_openpi_online` connects to a
-live server and asserts the advertised metadata matches these `pi0.yaml` values.
-
-## Run the client
-
-```bash
-python examples/online_serving/pi0/openpi_client.py --host 127.0.0.1 --port 8000 \
-    --prompt "pick up the red block and place it in the bin"
-```
-
-It connects, prints the server metadata, sends robot observations, and prints the
-returned `[action_horizon, action_dim] = [50, 32]` action chunks. Replace the blank
-cameras / zero state in `_make_dummy_obs` with real frames (HWC uint8) and
-proprioceptive state to drive a robot.
-
-### Observation format
-
-The client sends a flat dict per inference:
-
-```python
-{
-    "observation.images.base_0_rgb":       np.uint8[H, W, 3],
-    "observation.images.left_wrist_0_rgb": np.uint8[H, W, 3],
-    "observation.images.right_wrist_0_rgb":np.uint8[H, W, 3],
-    "state":   np.float32[state_dim],   # zero-padded to max_state_dim=32 server-side
-    "prompt":  "pick up the red block",
-    "session_id": "<uuid>",             # accepted but ignored (π0 is stateless)
-}
-```
-
-Camera keys must match the server's `image_feature_keys` (the checkpoint's
-`input_features` order). If your robot uses different camera names, set
-`model_config.image_key_map` in `pi0.yaml` to map raw obs keys → feature keys.
-
-## Correctness
-
-π0's flow-matching kernel is bit-for-bit matched to the LeRobot `PI0Policy` reference
-(`max|Δ| = 7.15e-07`, CPU/float32, fixed noise; see
-`tests/diffusion/models/pi0/test_pi0_parity.py::test_pi0_vllm_omni_vs_lerobot`) and is
-version-stable across transformers releases (the version-stability checks in
-`tests/diffusion/models/pi0/test_pi0_units.py`). An OpenPI websocket online-serving e2e
-lives in `tests/e2e/online_serving/test_pi0_expansion.py::test_pi0_openpi_online`.
-
-## Limitations
-
- **Normalization stats**: `lerobot/pi0_base` uses identity normalization, which is
-  fully supported (state passes through, actions are returned in the model's space).
-  Per-dataset `norm_stats` declared in a checkpoint's `config.json` are honored, but
-  stats stored only in LeRobot's `policy_preprocessor.json` companion safetensors are
-  not yet loaded — a fine-tuned checkpoint that relies on those would need that bridge
-  added before its actions are in real-world units.
@@ -1,119 +0,0 @@
-#!/usr/bin/env python3
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""Minimal OpenPI client for the π0 (Pi-Zero) VLA policy server.
-
-Connects to a running π0 server (see README.md), sends a single robot
-observation (3 cameras + state + language instruction), and prints the returned
-continuous action chunk ``[action_horizon, action_dim]``.
-
-The server speaks the OpenPI websocket protocol:
-    connect -> server sends msgpack(policy_server_config metadata)
-    infer   -> client sends msgpack(obs), server replies msgpack(actions ndarray)
-
-Usage:
-    python openpi_client.py --host 127.0.0.1 --port 8000 \
-        --prompt "pick up the red block"
-"""
-
-from __future__ import annotations
-
-import argparse
-import logging
-import sys
-import uuid
-from pathlib import Path
-from typing import Any
-
-import numpy as np
-
-try:
-    # Avoid shadowing the installed ``openpi_client`` package with this dir.
-    example_dir = str(Path(__file__).resolve().parent)
-    removed = sys.path and sys.path[0] == example_dir
-    if removed:
-        sys.path.pop(0)
-    try:
-        from openpi_client.websocket_client_policy import WebsocketClientPolicy
-    finally:
-        if removed:
-            sys.path.insert(0, example_dir)
-except ImportError as exc:  # pragma: no cover - runtime dependency guard
-    raise ImportError("π0 OpenPI example requires `openpi-client`.") from exc
-
-logger = logging.getLogger("pi0_openpi_client")
-
-DEFAULT_HOST = "127.0.0.1"
-DEFAULT_PORT = 8000
-DEFAULT_PATH = "/v1/realtime/robot/openpi"
-DEFAULT_PROMPT = "pick up the red block and place it in the bin"
-
-# π0 / pi0_base camera identities (must match the server's image_feature_keys).
-CAMERA_KEYS = (
-    "observation.images.base_0_rgb",
-    "observation.images.left_wrist_0_rgb",
-    "observation.images.right_wrist_0_rgb",
-)
-STATE_DIM = 32
-
-
-def _make_dummy_obs(prompt: str, session_id: str, image_size: int = 224) -> dict[str, Any]:
-    """Build a single π0 observation with blank cameras + zero state.
-
-    Replace the zeros with real camera frames (HWC uint8) and proprioceptive
-    state to drive an actual robot.
-    """
-    obs: dict[str, Any] = {cam: np.zeros((image_size, image_size, 3), dtype=np.uint8) for cam in CAMERA_KEYS}
-    obs["state"] = np.zeros(STATE_DIM, dtype=np.float32)
-    obs["prompt"] = prompt
-    obs["session_id"] = session_id
-    return obs
-
-
-def _ws_uri(host: str, port: int, path: str) -> str:
-    """Build the full websocket URI. WebsocketClientPolicy takes the whole URI as
-    ``host`` (it has no separate path arg)."""
-    if host.startswith(("ws://", "wss://")):
-        return host
-    return f"ws://{host}:{port}{path}"
-
-
-def _as_action_array(response: Any) -> np.ndarray:
-    if isinstance(response, dict) and response.get("type") == "error":
-        raise RuntimeError(f"Inference failed: {response.get('message', response)}")
-    return np.asarray(response, dtype=np.float32)
-
-
-def main() -> int:
-    parser = argparse.ArgumentParser(description="π0 OpenPI client")
-    parser.add_argument("--host", default=DEFAULT_HOST)
-    parser.add_argument("--port", type=int, default=DEFAULT_PORT)
-    parser.add_argument("--path", default=DEFAULT_PATH)
-    parser.add_argument("--prompt", default=DEFAULT_PROMPT)
-    parser.add_argument("--num-steps", type=int, default=2, help="Inference observations to send.")
-    args = parser.parse_args()
-    logging.basicConfig(level=logging.INFO, format="%(message)s")
-
-    policy = WebsocketClientPolicy(host=_ws_uri(args.host, args.port, args.path))
-    metadata = dict(policy.get_server_metadata())
-    logger.info("Server metadata: %s", metadata)
-
-    session_id = str(uuid.uuid4())
-    for step in range(args.num_steps):
-        obs = _make_dummy_obs(args.prompt, session_id)
-        actions = _as_action_array(policy.infer(obs))
-        logger.info(
-            "[step %d] actions shape=%s mean=%.4f std=%.4f",
-            step,
-            actions.shape,
-            float(actions.mean()),
-            float(actions.std()),
-        )
-        if not np.isfinite(actions).all():
-            raise RuntimeError("Server returned non-finite actions.")
-    logger.info("Done — π0 OpenPI round-trip OK.")
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())
@@ -79,7 +79,6 @@ plugins:
        user_guide/diffusion/step_execution.md: user_guide/diffusion/execution_modes.md
        user_guide/diffusion/cpu_offload_diffusion.md: user_guide/diffusion/cpu_offload.md
        user_guide/quantization/index.md: user_guide/quantization/overview.md
-        configuration/pd_disaggregation.md: features/pd_disaggregation.md
        design/feature/diffusion_request_level_batching.md: design/feature/diffusion_continuous_batching.md
        design/feature/diffusion_step_execution.md: design/feature/diffusion_continuous_batching.md
        design/feature/cpu_offloading.md: design/feature/offloader/README.md
@@ -81,7 +81,7 @@ MammothModa2 generation parameters as a JSON object through `--extra-body`:
 ```bash
 python examples/offline_inference/text_to_image/text_to_image.py \
  --model ./MammothModa2-Preview \
-  --deploy-config vllm_omni/deploy/mammoth_moda2.yaml \
+  --stage-configs-path vllm_omni/deploy/mammoth_moda2.yaml \
  --prompt "A stylish woman riding a motorcycle in NYC, movie poster style" \
  --height 1024 \
  --width 1024 \
@@ -132,44 +132,6 @@ ls -lh mammoth_t2i.png
 python -c "from PIL import Image; print(Image.open('mammoth_t2i.png').size)"
 ```

-### 1x AMD MI300X, MammothModa2 Preview
-
-#### Environment
-
- OS: Linux 6.8.0-134-generic, x86_64
- Container: official ROCm image built from `docker/Dockerfile.rocm`
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- Driver / runtime: AMD 6.19.14.31400000 / ROCm 7.2.53211
- GPU: one AMD Instinct MI300X, `gfx942:sramecc+:xnack-`, 191.69 GiB visible HBM
- vLLM version: 0.27.0+rocm723
- vLLM Omni version or commit: `73e1368c7bb940efe1a025859c9d6c8eeeb2e3f0`
- Installed vLLM Omni package metadata: `0.27.0rc2.dev44+g55abdade9.rocm`
-
-#### Offline Commands
-
-The checked run used the committed stage split, with `gpu_memory_utilization` set to 0.5 for AR and 0.3 for DiT:
-
-```bash
-python3 examples/offline_inference/text_to_image/text_to_image.py \
-    --model bytedance-research/MammothModa2-Preview \
-    --deploy-config vllm_omni/deploy/mammoth_moda2.yaml \
-    --prompt "A stylish woman riding a motorcycle in NYC, movie poster style" \
-    --height 1024 \
-    --width 1024 \
-    --seed 42 \
-    --extra-body '{"text_guidance_scale": 4.0, "cfg_range": [0.0, 1.0], "num_inference_steps": 50}' \
-    --enable-diffusion-pipeline-profiler \
-    --log-stats \
-    --output mammoth_t2i.png
-```
-
-#### Verification
-
-The first request took 85.224 seconds. The AR stage generated 4,161 visual tokens in 72.996 seconds, and the DiT stage took 12.163 seconds. AR weight loading used 21.4 GiB and took 8.250 seconds. DiT weight loading used 5.49 GiB and took 1.824 seconds. The largest one second whole device memory sample was 106.57 GiB, including the AR KV cache reserved by the 0.5 memory setting.
-
-The output was a valid 1024 by 1024 RGB PNG.
-
 ## MammothModa2-Dev unified inference

 MammothModa2-Dev uses a Qwen3-VL AR backbone, while MammothModa2-Preview uses
@@ -416,10 +416,9 @@ For example, keep the first main block's attention projections in BF16 with:
 ```

 The structured option replaces `--quantization fp8`. Online FP8 can be used
-with H3 layerwise offload and with both DLO transfer paths. The default
-AllGather path uses the ordinary loader to finalize FP8 weights and scales
-before sharding them across ranks. `--dlo-no-use-allgather` instead retains
-complete rank-local tensors and avoids the synchronized request-wave contract.
+with H3 layerwise offload and with distributed layerwise offload's full-weight
+per-rank path (`--dlo-no-use-allgather`). The sharded DLO AllGather path is not
+supported for runtime-created FP8 weights.

 ## AMD ROCm (gfx942 / gfx950)

@@ -871,9 +870,8 @@ vllm serve "${MODEL_ROOT}/FL2VA" \
  H3 native `tile` mode only.
 - A U2 x Ring2 hybrid currently fails with an attention-mask length mismatch; use
  pure Ulysses.
- Online FP8 with DLO AllGather temporarily materializes the complete FP8 model
-  in host memory on every rank during startup before retaining only each rank's
-  shard. Size startup host memory for that transient peak.
+- Runtime-created FP8 weights do not support the sharded DLO AllGather path;
+  use `--dlo-no-use-allgather` when combining online FP8 with DLO.
 - TeaCache and Cache-DiT cannot be enabled on the same server.
 - Image+audio Ref2VA accepts exactly one image and one audio reference.
 - Video Ref2VA accepts one or more video files, but not an additional standalone
@@ -1,173 +0,0 @@
-# MiniMax Music 3
-
-> Text-to-music: lyrics plus a style caption in, a 32 kHz stereo song out
-
-## Summary
-
- Vendor: MiniMax
- Model: `MiniMaxAI/MiniMax-Music3`
- Task: Text-to-music generation (vocals and instrumental together)
- Mode: Online serving through the OpenAI-compatible `/v1/audio/speech` API
- Maintainer: @linyueqian
-
-## When to use this recipe
-
-Use it to serve `MiniMaxAI/MiniMax-Music3` for full-song generation up to six
-minutes. This is not a TTS model with a musical voice: there is no reference
-speaker, no `voice`, and no `temperature`. What you control is the lyrics, the
-caption, the seed, and the length.
-
-## References
-
- Model card: [MiniMaxAI/MiniMax-Music3](https://huggingface.co/MiniMaxAI/MiniMax-Music3)
- Demo samples: [MiniMax Music 3 demo](https://minimax-ai.github.io/music3-demo/)
- Deploy config: `vllm_omni/deploy/minimax_music3.yaml` (single GPU),
-  `vllm_omni/deploy/minimax_music3_2gpu.yaml` (split across two)
-
-## Architecture
-
-| Component | Spec |
-|---|---|
-| Backbone | Qwen3 decoder, 36 layers, hidden 4096, GQA 32/8, vocab 200k |
-| Depth decoder | 4 layers, hidden 4096, 16 heads; 7 residual codebooks of 1024 |
-| Frame | 8 codebooks; `c0` in the backbone vocabulary, `c1..c7` in the depth decoder |
-| AR guidance | Classifier-free, scale 1.5, `c0` masked to the conditioned branch's top 50 |
-| Acoustic stage | Flow-matching transformer (36 layers, dim 2048) plus a DAC-style decoder, 512x upsampling |
-| Solver | Euler, 30 steps, acoustic guidance scale 1.7 |
-| Window | 200 frames per acoustic chunk, 100-frame hop |
-| Frame rate | 25 frames per second |
-| Context length | 10,240 tokens |
-| Output | 32 kHz stereo WAV |
-
-Generation is two-stage. The backbone predicts one audio frame per decode step,
-each frame eight RVQ codebooks deep. Every 200 frames are handed to the
-flow-matching transformer, which solves a latent that the decoder turns into
-waveform.
-
-**Guidance doubles the rows.** Every request decodes twice: one row sees the
-real prompt, the other sees the same prompt with the caption-and-lyrics span
-replaced by `<|audio_cfg|>`. Both rows hold their own KV cache for the whole
-song, so a request costs twice the KV its length suggests. Size the pool for
-rows, not requests.
-
-## Hardware Support
-
-## GPU
-
-### 1x H200 141GB
-
-#### Environment
-
- OS: Linux
- Python: 3.12
- CUDA: 12.9
- vLLM version: 0.27.0
- vLLM-Omni version or commit: branch `feat/minimax-music3`
-
-#### Command
-
-```bash
-vllm serve MiniMaxAI/MiniMax-Music3 \
-    --host 0.0.0.0 --port 8000 \
-    --trust-remote-code --omni
-```
-
-The default deploy config `vllm_omni/deploy/minimax_music3.yaml` is discovered
-automatically from the checkpoint's `model_type=minimax_music3`. It colocates
-both stages on device 0.
-
-#### Verification
-
-```bash
-curl -X POST http://localhost:8000/v1/audio/speech \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "MiniMaxAI/MiniMax-Music3",
-    "input": "[Verse]\nWalking down the empty street at midnight\nStreetlights flicker like a broken dream\n[Chorus]\nAnd I keep on walking\nTill the morning finds me",
-    "instructions": "A melancholic lo-fi hip-hop track at 85 BPM in F minor: mellow Rhodes piano riff, soft vinyl crackle, dusty boom-bap drums with a laid-back swing, warm upright bass. Intimate bedroom production, gentle tape saturation, no bright cymbals.",
-    "seed": 1,
-    "max_new_tokens": 750
-  }' \
-  --output song.wav
-```
-
-Expected: a 30 second 32 kHz stereo WAV. `max_new_tokens` counts audio frames
-at 25 per second, so 750 frames is at most 30 seconds. It is a cap, not a
-target: the model ends the song itself when it emits the audio-end token, and a
-response shorter than the cap is the model finishing rather than a truncation.
-
-#### Notes
-
- **Memory:** stage 0 takes `gpu_memory_utilization: 0.6` for the bf16 backbone
-  and its KV pool; stage 1 takes 0.2 for the transformer and decoder weights,
-  which sit outside any KV budget.
- **Key flags:** none required. Guidance, the solver step count and the window
-  geometry are fixed by the checkpoint and are not request parameters.
- **Known limitations:**
-  - The external API is non-streaming; `stream: true` is rejected.
-  - `max_model_len` is 10,240, so a maximum-length prompt (5,000 tokens) and a
-    maximum-length song (9,000 frames) cannot both fit. Long captions shorten
-    the maximum song. This is a property of the checkpoint.
-  - The acoustic stage runs in float32. bfloat16 is accepted but measurably
-    degrades the solver.
-
-### 2x H200 141GB
-
-Use `minimax_music3_2gpu.yaml` to put the acoustic stage on the second device:
-
-```bash
-vllm serve MiniMaxAI/MiniMax-Music3 \
-    --host 0.0.0.0 --port 8000 \
-    --trust-remote-code --omni \
-    --deploy-config minimax_music3_2gpu.yaml
-```
-
-Only the placement differs; both layouts run the acoustic stage in float32.
-
-## Writing the request
-
-Two fields carry the request, and both are required.
-
- `input` is the **lyrics**. Structure tags (`[Verse]`, `[Chorus]`, `[Bridge]`,
-  `[Outro]`) steer the arrangement and are part of the prompt contract.
-  **Put a tag on its own line.** Normalization keeps only the tags on a line
-  that starts with one and drops the rest of that line, so lyrics written next
-  to a tag are silently lost:
-
-  ```text
-  "[Verse]\nWalking down the street"   ->   [start] [verse] Walking down the street
-  "[Verse] Walking down the street"    ->   [start] [verse]
-  ```
-
- `instructions` is the **caption** describing genre, instrumentation, tempo,
-  mood and production. It is the strongest control available: vague captions
-  give generic arrangements. Long structured captions work well, since the
-  model was trained on descriptions covering global attributes, emotional
-  progression, vocal detail and arrangement.
-
-A request is deterministic in its seed: the same lyrics, caption, seed and
-length return the same audio. Changing only the seed gives another take of the
-same song.
-
-## Request parameters
-
-| Parameter | Type | Default | Description |
-|---|---|---|---|
-| `input` | string | required | Lyrics, non-empty. Tags on their own lines |
-| `instructions` | string | required | Caption, non-empty |
-| `seed` | int | `0` | Non-negative. Fixes the output for a given request |
-| `max_new_tokens` | int | `9000` | Cap on audio frames, 25 per second. Maximum 9,000 |
-| `response_format` | string | `"wav"` | Output container |
-
-## Parameters the model rejects
-
-The request is refused rather than silently ignored, so a mistake is visible
-immediately.
-
-| Parameter | Reason |
-|---|---|
-| `temperature`, `top_p`, `top_k`, `repetition_penalty` | Sampling is fixed: guidance at 1.5, then a seeded top-k 50 draw |
-| `voice` | There is no speaker to select; the vocal comes from the caption |
-| `ref_audio`, `ref_text`, `language`, `task_type` | No reference-audio conditioning and no language tag |
-| `speed` | Only `1.0`. Tempo belongs in the caption, e.g. "at 92 BPM" |
-| `stream: true` | The external API is non-streaming |
@@ -71,35 +71,6 @@ single-GPU entry point; `minicpmo_4_5_2gpu.yaml` is the recommended
 two-GPU profile. The removed fused two-stage implementation is not retained as
 a fallback because it would duplicate state machines and correctness paths.

-### Graph execution
-
-Graph boundaries follow each stage's state model:
-
-| Stage | Graph mode on Ascend | Eager boundary |
-| --- | --- | --- |
-| 0 Thinker | vLLM `PIECEWISE` | multimodal preprocessing and output routing |
-| 1 Talker | vLLM `PIECEWISE` | conditioning preprocessing, codec sampling, request-state updates |
-| 2 Code2Wav | inner exact-shape NPUGraph for the CFM DiT estimator | encoder, timestep embedding, HiFT/RNG, request parsing, stream-state commit |
-
-Stage 2 keeps `enforce_eager: true` for the generation runner because its
-Python request metadata and per-request cache dictionaries are not valid
-outer-graph inputs. The backend captures only the deterministic CFM DiT
-estimator, with an exact graph key for each tensor shape. Timestep embedding
-stays eager because the upstream implementation creates a CPU frequency tensor;
-HiFT stays eager because it generates random phase. The graph cache stores up
-to 32 entries by default, after which unseen shapes run eagerly.
-
-An ACL graph capture failure is fatal to that Stage-2 process because older
-torch-npu releases can leave allocator and RNG capture state invalid. Restart
-the service after a capture failure. To run without capture, set
-`code2wav_enable_npu_graph: false` under the Stage-2
-`platforms.npu.stages[].additional_config` block before startup. Tune the cache
-limit there with `code2wav_max_npu_graphs`. Graph mode also requires
-`ASCEND_LAUNCH_BLOCKING` to be unset or set to `0`.
-
-The inner NPUGraph is independent of the outer runner setting, so do not use a
-global `--enforce-eager` override when Stage 0/1 `PIECEWISE` replay is desired.
-
 ## GPU

 ### 1 x GPU (default — single command)
@@ -240,9 +211,9 @@ speech output (TTS)"** checkbox on / off.
  model processes still share one CUDA device.
 - `--trust-remote-code` is required — the HF repo ships a custom
  `MiniCPMO` config / model class.
- Stage 0 Thinker and Stage 1 Talker enable vLLM CUDA Graphs. Stage 2 keeps its
-  request orchestration eager while using inner CFM and HiFT CUDA Graphs. On
-  Ascend, the exact-shape CFM DiT estimator instead uses inner NPUGraph replay.
+- Stage 0 Thinker and Stage 1 Talker enable vLLM CUDA Graphs. Stage 2 remains
+  eager because its request-owned Flow/HiFT caches and variable chunk/cache
+  shapes are not yet exposed through a static exact-shape graph wrapper.
 - All default stages use `max_num_seqs: 4` to reduce cross-process GPU
  contention. Talker AR
  state and Code2Wav caches are request-owned; Code2Wav batches only
@@ -253,8 +224,8 @@ speech output (TTS)"** checkbox on / off.
 - `StageRequestStats.batch_size` is a request-scoped placeholder, not the
  scheduler's execution batch.
 - Single-GPU co-location trades throughput for hardware density: Stage 0/1
-  CUDA Graph replay and Stage 2 vocoder kernels compete across three CUDA
-  contexts. Use the 8x4090 config or a custom multi-GPU mapping for
+  CUDA Graph replay and eager Stage 2 vocoder kernels compete across three
+  CUDA contexts. Use the 8x4090 config or a custom multi-GPU mapping for
  throughput-sensitive serving.

 ### 8 x RTX 4090 24GB (consumer-GPU layout)
@@ -252,53 +252,3 @@ curl -X POST http://localhost:8091/v1/audio/speech \
    - stage_id: 1
      gpu_memory_utilization: 0.15
  ```
-
-### 1x AMD MI300X, 1.7B checkpoints
-
-#### Environment
-
- OS: Linux 6.8.0-134-generic, x86_64
- Container: official ROCm image built from `docker/Dockerfile.rocm`
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- Driver / runtime: AMD 6.19.14.31400000 / ROCm 7.2.53211
- GPU: one AMD Instinct MI300X, `gfx942:sramecc+:xnack-`, 191.69 GiB visible HBM
- vLLM version: 0.27.0+rocm723
- vLLM Omni version or commit: `3fecb6953ca8dc51210cc0421ef24552267a41ef`
- Installed vLLM Omni package metadata: `0.27.0rc2.dev44+g55abdade9.rocm`
- ONNX Runtime: onnxruntime-rocm 1.22.2.post3 with `ROCMExecutionProvider`
- transformers: 5.15.0
-
-#### Command
-
-The MI300X tests also worked without `enforce_eager`. To use that setup, remove `enforce_eager: true` from stage 1 under `platforms.rocm` in `vllm_omni/deploy/qwen3_tts.yaml`.
-
-Set this environment variable before you start Qwen3-TTS:
-
-```bash
-export MIOPEN_FIND_MODE=FAST
-```
-
-#### Verification
-
-Each row records one offline request with one prompt and batch size one. Request time is the logged `e2e_wall_time_ms`, and stage 1 time is the logged `e2e_stage_1_wall_time_ms`.
-
-| Checkpoint | Code2Wav configuration | `MIOPEN_FIND_MODE` | Request time | Stage 1 time | Output duration | Real time factor | Maximum sampled device memory |
-| --- | --- | --- | ---: | ---: | ---: | ---: | ---: |
-| CustomVoice | Eager | Unset | 1.683 s | 1.613 s | 5.68 s | 0.30 | 60.91 GiB |
-| CustomVoice | Graphs allowed | `FAST` | 1.326 s | 1.259 s | 5.68 s | 0.23 | 62.19 GiB |
-| VoiceDesign | Eager | Unset | 1.195 s | 1.180 s | 4.40 s | 0.27 | 60.91 GiB |
-| VoiceDesign | Graphs allowed | `FAST` | 1.127 s | 1.054 s | 4.40 s | 0.26 | 62.21 GiB |
-| Base | Eager | Unset | 13.330 s | 13.263 s | 4.32 s | 3.09 | 61.74 GiB |
-| Base | Graphs allowed | `FAST` | 14.314 s | 14.245 s | 4.80 s | 2.98 | 62.84 GiB |
-
-All six processes finished successfully, and each output passed the 24 kHz mono WAV checks.
-
-#### Notes
-
- The current ROCm config keeps Code2Wav in eager mode.
- The tested alternative allows graphs and sets [`MIOPEN_FIND_MODE=FAST`](https://rocm.docs.amd.com/projects/MIOpen/en/develop/reference/env_variables.html). MIOpen uses a saved kernel choice when available and uses its immediate fallback otherwise.
- CustomVoice used a graph for one of four logged Code2Wav batches. VoiceDesign used a graph for two of four batches. Base used no graphs for its four logged batches.
- The recorded request time with graphs allowed was lower for CustomVoice and VoiceDesign and higher for Base. The Base runs generated different amounts of audio, so their request times are not a direct comparison.
- Each setup was tested with one request, so the timing is not a performance benchmark.
- GPU memory was sampled once per second with `rocm-smi`.
@@ -49,7 +49,6 @@ recipes/
 | [`Helios/Helios.md`](./Helios/Helios.md) | Text-to-video, image-to-video, and video-to-video generation | 1x NVIDIA H20 |
 | [`inclusionAI/Ming-flash-omni-2.0.md`](./inclusionAI/Ming-flash-omni-2.0.md) | Online serving for multimodal chat + standalone TTS | 4x H100 / 1x H100 80GB |
 | [`inclusionAI/Ming-omni-tts.md`](./inclusionAI/Ming-omni-tts.md) | Offline + online dense Ming TTS/audio generation | 1x H100 80GB / 1x AMD MI300X (ROCm 7.2) |
-| [`k2-fsa/OmniVoice.md`](./k2-fsa/OmniVoice.md) | Offline and online multilingual text-to-speech | 1x AMD MI300X (ROCm 7.2) |
 | [`IndexTeam/IndexTTS-2.md`](./IndexTeam/IndexTTS-2.md) | Online serving for voice-cloned TTS with optional emotion control | 1x L4 24GB or larger CUDA GPU |
 | [`IndexTeam/IndexTTS-2_5.md`](./IndexTeam/IndexTTS-2_5.md) | Online + offline multilingual voice-cloned TTS with native speed and emotion control | 1x NVIDIA H20 96GB |
 | [`MiniMaxAI/MiniMax-H3.md`](./MiniMaxAI/MiniMax-H3.md) | T2VA, FL2VA, image+audio Ref2VA, and multi-video Ref2VA serving | 1x GPU with CPU offload / 4x B300 |
@@ -74,9 +73,9 @@ recipes/
 | [`Qwen/Qwen-Image-2512.md`](./Qwen/Qwen-Image-2512.md) | Text-to-image serving with step-wise continuous batching replay and ModelOpt FP8 / mixed FP8/NVFP4 | 1x A800 80GB / 2x B200 |
 | [`Qwen/Qwen-Image-Edit.md`](./Qwen/Qwen-Image-Edit.md) | Text-guided single-image editing | 1x or 2x H200 141GB |
 | [`Qwen/Qwen3-Omni.md`](./Qwen/Qwen3-Omni.md) | Online serving for multimodal chat | 1x A100 80GB |
-| [`Qwen/Qwen3-TTS.md`](./Qwen/Qwen3-TTS.md) | Text-to-speech serving (CustomVoice / VoiceDesign / Base) | 1x H100/A100 80GB / 1x AMD MI300X |
-| [`SenseNova/SenseNova-U1.md`](./SenseNova/SenseNova-U1.md) | Unified image generation and understanding | 1x or 2x H200 / 1x AMD MI300X |
-| [`StabilityAI/Stable-Audio-Open.md`](./StabilityAI/Stable-Audio-Open.md) | Offline + online text-to-audio generation (Stable Audio Open) | 1x RTX 4090 24GB / 1x AMD MI300X |
+| [`Qwen/Qwen3-TTS.md`](./Qwen/Qwen3-TTS.md) | Text-to-speech serving (CustomVoice / VoiceDesign / Base) | 1x H100/A100 80GB |
+| [`SenseNova/SenseNova-U1.md`](./SenseNova/SenseNova-U1.md) | Unified image generation and understanding | 1x H200 (144GB) |
+| [`StabilityAI/Stable-Audio-Open.md`](./StabilityAI/Stable-Audio-Open.md) | Offline + online text-to-audio generation (Stable Audio Open) | 1x RTX 4090 24GB |
 | [`Tencent/Covo-Audio-Chat.md`](./Tencent/Covo-Audio-Chat.md) | Online serving for audio chat | 1x A100 80GB |
 | [`Tencent/HunyuanImage-3.0-Instruct.md`](./Tencent/HunyuanImage-3.0-Instruct.md) | DiT-only text-to-image serving and benchmark, including ModelOpt mixed FP8/NVFP4 | 4x H100/H800 80GB / 2x B200 |
 | [`Wan-AI/Wan2.2-T2V.md`](./Wan-AI/Wan2.2-T2V.md) | Text-to-video serving (Wan2.2 14B) with Skip-Softmax sparse attention | 1x datacenter Blackwell (B200/B300) |
@@ -178,48 +178,6 @@ with the expected 1536×2720 resolution.
 - The LLM transformer uses `QKVParallelLinear` and `MergedColumnParallelLinear`
  for fused QKV and gate/up projections with TP support.

-### 1x AMD MI300X 192GB
-
-#### Environment
-
- OS: Linux 6.8.0-134-generic, x86_64
- Container: official ROCm image built from `docker/Dockerfile.rocm`
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- Driver / runtime: AMD 6.19.14.31400000 / ROCm 7.2.53211
- GPU: one AMD Instinct MI300X, `gfx942:sramecc+:xnack-`, 191.69 GiB visible HBM
- vLLM version: 0.27.0+rocm723
- vLLM Omni version or commit: `a704c8759c96e123c0d7c89b11f120b1c0f120cf`
- Installed vLLM Omni package metadata: `0.27.0rc2.dev44+g55abdade9.rocm`
-
-#### Command
-
-```bash
-python3 examples/offline_inference/text_to_image/text_to_image.py \
-    --model SenseNova/SenseNova-U1-8B-MoT \
-    --prompt "Close portrait of an elderly woman by a farmhouse window, textured skin, gentle smile, warm natural light, emotional documentary look." \
-    --width 1536 \
-    --height 2720 \
-    --seed 42 \
-    --num-inference-steps 50 \
-    --cfg-scale 4.0 \
-    --extra-body '{"think": true, "cfg_norm": "none", "timestep_shift": 3.0, "t_eps": 0.02}' \
-    --enable-diffusion-pipeline-profiler \
-    --log-stats \
-    --output sensenova_u1_mi300x.png
-```
-
-#### Verification
-
-The command completed and wrote a valid 1536 by 2720 RGB PNG.
-
-#### Notes
-
- Generation with 50 inference steps took 34.011 seconds.
- Model loading used 32.774 GiB and took 13.987 seconds.
- The internal profiler recorded 35.67 GB reserved and 35.10 GB allocated for the request.
- The highest one second whole device memory sample was 37.93 GiB.
-
 ## Online Serving

 SenseNova-U1 supports all four modalities via the OpenAI-compatible
@@ -164,50 +164,7 @@ PY
  `torchaudio==2.11.0+cu130`.
 - The `NIXL is not available`, `GLOO_SOCKET_IFNAME`, and `torchsde` boundary
  warnings observed during validation did not prevent successful generation.
- The RTX 4090 entry was validated on one 24 GB GPU. The MI300X entry below covers one 192 GB GPU.
+- This recipe was validated on one RTX 4090 24GB GPU only. Other GPU counts,
+  ROCm, XPU, and NPU setups are not covered here.
 - Long generations, higher inference-step counts, and non-WAV response formats
  were not benchmarked in this recipe.
-
-#### 1x AMD MI300X 192GB
-
-##### Environment
-
- OS: Linux 6.8.0-134-generic, x86_64
- Container: official ROCm image built from `docker/Dockerfile.rocm`
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- Driver / runtime: AMD 6.19.14.31400000 / ROCm 7.2.53211
- GPU: one AMD Instinct MI300X, `gfx942:sramecc+:xnack-`, 191.69 GiB visible HBM
- vLLM version: 0.27.0+rocm723
- vLLM Omni version or commit: `73e1368c7bb940efe1a025859c9d6c8eeeb2e3f0`
- Installed vLLM Omni package metadata: `0.27.0rc2.dev44+g55abdade9.rocm`
-
-##### Commands
-
-```bash
-python3 examples/offline_inference/text_to_audio/text_to_audio.py \
-    --model stabilityai/stable-audio-open-1.0 \
-    --prompt "A gentle piano melody with soft room ambience" \
-    --negative-prompt "Low quality, distorted, noisy" \
-    --seed 42 \
-    --guidance-scale 7.0 \
-    --audio-length 10.0 \
-    --num-inference-steps 50 \
-    --cache-backend tea_cache \
-    --enable-diffusion-pipeline-profiler \
-    --output stable_audio_10s.wav
-```
-
-##### Verification
-
-The command completed and wrote a valid 44.1 kHz stereo WAV with 10.00 seconds of audio.
-
-##### Notes
-
- TeaCache ran with `rel_l1_thresh=0.2`.
- Model loading used 2.7891 GiB and took 3.706 seconds.
- Generation took 4.750 seconds, which gives a real time factor of 0.475 for the 10.00 second output.
- The internal profiler recorded 15.65 GB reserved and 9.68 GB allocated for the request.
- The highest one second whole device memory sample was 19.61 GiB.
- The output RMS was 0.0887, and its peak absolute amplitude was 0.5761.
- The full process took 384 seconds, including startup and compilation.
@@ -189,10 +189,8 @@ wait

 - `--dlo-use-allgather` (default) requires all concurrent requests to have the same
  `num_inference_steps` (AllGather is a collective that needs all ranks synchronized)
- Online per-tensor FP8 linears with DLO AllGather use the ordinary loader
-  instead of direct checkpoint mmap, then shard the finalized FP8 weights and
-  scales. Size host memory for a transient complete FP8 model on every rank at
-  startup
+- Online quantization (FP8) is incompatible with the DLO+AllGather mmap path;
+  use `--dlo-no-use-allgather` or disable online quantization
 - Tensor Parallel > 1 is rejected in the DLO+AllGather mmap path. TP with
  `--dlo-no-use-allgather` retains the standard TP loader and should be treated
  as an experimental compatibility path
@@ -170,35 +170,6 @@ python examples/online_serving/text_to_speech/ming_tts/openai_speech_client.py \
 `--ref-audio` matches upstream `use_spk_emb=True`; do not add `--ref-text`
 for the dialect case.

-**Offline smoke result on ROCm 7.2:**
-
-A later offline smoke run used the same MI300X model with the following software:
-
- GPU: one AMD Instinct MI300X with 191.69 GiB visible HBM and gfx942
- Kernel: Linux 6.8.0-134-generic, x86_64
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- ROCm or HIP: 7.2.53211
- vLLM: 0.27.0+rocm723
- vLLM Omni checkout: `73e1368c7bb940efe1a025859c9d6c8eeeb2e3f0`
-
-```bash
-python3 examples/offline_inference/text_to_speech/ming_tts/end2end.py \
-    --model inclusionAI/Ming-omni-tts-0.5B \
-    --deploy-config vllm_omni/deploy/ming_tts.yaml \
-    --case basic \
-    --text "你好，这是 AMD MI300X ROCm 配方验证。" \
-    --output-dir ./ming_tts_output \
-    --output-name ming_omni_tts_mi300x.wav \
-    --enforce-eager \
-    --log-stats \
-    --metadata-json ./ming_tts_manifest.json
-```
-
-The checked run produced a valid 44.1 kHz mono WAV with 4.80 seconds of audio, RMS 0.1449, and peak absolute amplitude 0.8621. The first request took 20.406 seconds, which gives a real time factor of 4.25. The stage weight loads used 1.31 GiB and 1.47 GiB and took 1.756 seconds and 1.264 seconds. The largest one second whole device memory sample was 91.05 GiB, including the reserved KV cache. Both stages used eager mode, and the AR stage selected `TRITON_ATTN`.
-
-The run used the official ROCm image built from `docker/Dockerfile.rocm`.
-
 ## Notes

 - The official ROCm image includes the platform dependencies.
@@ -1,98 +0,0 @@
-# OmniVoice
-
-> Offline and online text to speech with automatic voice, voice design, and reference voice cloning
-
-## Summary
-
- Vendor: k2-fsa
- Model: `k2-fsa/OmniVoice`
- Task: Multilingual text to speech
- Mode: Offline inference for all modes and online serving for automatic voice
- Maintainer: Community
-
-## References
-
- [Hugging Face model](https://huggingface.co/k2-fsa/OmniVoice)
- [Offline text to speech examples](../../docs/user_guide/examples/offline_inference/text_to_speech.md#omnivoice)
- [Online text to speech examples](../../docs/user_guide/examples/online_serving/text_to_speech.md#omnivoice)
- [Deploy config](../../vllm_omni/deploy/omnivoice.yaml)
-
-## Requirements
-
-Download the checkpoint before the first run:
-
-```bash
-huggingface-cli download k2-fsa/OmniVoice
-```
-
-Automatic voice and voice design require `transformers>=4.57.0`. Reference voice cloning requires `transformers>=5.3.0`.
-
-The vLLM Omni deploy config runs the OmniVoice generator and decoder inside one float32 diffusion stage.
-
-## ROCm
-
-### 1x AMD MI300X
-
-#### Environment
-
- GPU: one AMD Instinct MI300X with 191.69 GiB visible HBM and gfx942
- Kernel: Linux 6.8.0-134-generic, x86_64
- Python: 3.12.13
- PyTorch: 2.11.0+gitd0c8b1f
- ROCm or HIP: 7.2.53211
- vLLM: 0.27.0+rocm723
- vLLM Omni checkout: `73e1368c7bb940efe1a025859c9d6c8eeeb2e3f0`
- transformers: 5.15.0
-
-The run used the official ROCm image built from `docker/Dockerfile.rocm`.
-
-#### Automatic voice
-
-Run the offline example from the repository root:
-
-```bash
-python3 examples/offline_inference/text_to_speech/omnivoice/end2end.py \
-    --model k2-fsa/OmniVoice \
-    --deploy-config vllm_omni/deploy/omnivoice.yaml \
-    --text "Hello, this is OmniVoice running on one AMD MI300X." \
-    --seed 42 \
-    --output omnivoice_mi300x.wav
-```
-
-The checked run produced a valid 24 kHz mono WAV with 3.60 seconds of audio, RMS 0.1194, and peak absolute amplitude 0.9129. The first request took 0.685 seconds, which gives a real time factor of 0.19. Model loading used 3.5739 GiB and took 9.564 seconds. The request used 4.56 GB reserved and 3.64 GB allocated according to the internal profiler, while the largest one second whole device sample was 8.33 GiB. The log states that the OmniVoice Triton kernels loaded.
-
-#### Reference voice cloning
-
-The offline interface also accepts a reference WAV and its transcript:
-
-```bash
-python3 examples/offline_inference/text_to_speech/omnivoice/end2end.py \
-    --model k2-fsa/OmniVoice \
-    --text "Hello, this is a cloned voice." \
-    --ref-audio ref.wav \
-    --ref-text "This is the reference transcription." \
-    --output cloned.wav
-```
-
-#### Voice design
-
-```bash
-python3 examples/offline_inference/text_to_speech/omnivoice/end2end.py \
-    --model k2-fsa/OmniVoice \
-    --text "Hello, this is a designed voice." \
-    --instruct "female, low pitch, British accent" \
-    --output designed.wav
-```
-
-## Online serving
-
-Online serving currently supports automatic voice:
-
-```bash
-vllm serve k2-fsa/OmniVoice --omni --port 8091 --trust-remote-code
-```
-
-```bash
-cd examples/online_serving/text_to_speech/omnivoice
-python speech_client.py --text "Hello, how are you?"
-```
@@ -1,108 +0,0 @@
-# Pi0 (π0, Pi-Zero) VLA
-
-> Vision-Language-Action policy serving over the OpenPI realtime websocket
-
-## Summary
-
- Vendor: Physical Intelligence (checkpoint redistributed by LeRobot)
- Model: `lerobot/pi0_base`
- Task: Vision-Language-Action — multi-camera images + a language instruction +
-  robot proprioceptive state → a continuous action chunk
-  `[action_horizon, action_dim] = [50, 32]` via flow-matching denoising. π0 does
-  **not** emit text tokens.
- Mode: Online serving via the OpenPI realtime robot API
-  (`/v1/realtime/robot/openpi` websocket)
- Maintainer: Community
-
-## When to use this recipe
-
-Use this recipe to deploy `lerobot/pi0_base` as a robot policy server: a robot
-client connects over the OpenPI websocket and, per control step, sends an
-observation (camera frames + state + instruction) and receives an action chunk.
-
-A single pipeline class (`Pi0Pipeline`) serves the policy. π0 is stateless across
-calls (no KV reuse), so the protocol's `session_id` / `reset` are accepted but
-ignored.
-
-## Weights
-
-π0 uses the LeRobot `lerobot/pi0_base` checkpoint (HF, ~13 GB). Either let the
-server download it (`MODEL=lerobot/pi0_base`) or point at a local copy.
-
-## Hardware Support
-
-### 1x H200 141GB (Online serving)
-
-#### Environment
-
- OS: Ubuntu 22.04+
- Python: 3.12+
- Driver / runtime: NVIDIA CUDA environment
- vLLM / vLLM-Omni: use the commit you are deploying from
- OpenPI client extras (for the example client): `openpi-client`
-  (`pip install -e packages/openpi-client` from the
-  [openpi](https://github.com/Physical-Intelligence/openpi) repo), `websockets`,
-  `msgpack`, `msgpack-numpy`
-
-#### Command
-
-```bash
-vllm serve lerobot/pi0_base \
-  --omni \
-  --host 0.0.0.0 --port 8000 \
-  --served-model-name pi0 \
-  --deploy-config vllm_omni/deploy/pi0.yaml \
-  --enforce-eager --disable-log-stats
-```
-
-The deploy config (`vllm_omni/deploy/pi0.yaml`) declares a single diffusion stage
-(`Pi0Pipeline`), `float32`, `max_num_seqs: 1`, and the `policy_server_config`
-handshake metadata (3 cameras, `joint_position`, action horizon 50, action dim
-32) that enables the OpenPI realtime serving layer.
-
-#### Verification
-
-Run the example client (sends a synthetic observation and prints the returned
-action chunk):
-
-```bash
-python examples/online_serving/pi0/openpi_client.py \
-  --host 127.0.0.1 --port 8000 \
-  --prompt "pick up the red block and place it in the bin"
-```
-
-It connects, prints the server metadata, sends robot observations, and prints the
-returned `[action_horizon, action_dim] = [50, 32]` action chunks. The observation
-is a flat dict per inference:
-
-```python
-{
-    "observation.images.base_0_rgb":        np.uint8[H, W, 3],
-    "observation.images.left_wrist_0_rgb":  np.uint8[H, W, 3],
-    "observation.images.right_wrist_0_rgb": np.uint8[H, W, 3],
-    "state":   np.float32[state_dim],   # zero-padded to max_state_dim=32 server-side
-    "prompt":  "pick up the red block",
-    "session_id": "<uuid>",             # accepted but ignored (π0 is stateless)
-}
-```
-
-Camera keys must match the server's `image_feature_keys` (the checkpoint's
-`input_features` order). If your robot uses different camera names, set
-`model_config.image_key_map` in `pi0.yaml` to map raw obs keys → feature keys.
-
-## Correctness
-
-π0's flow-matching kernel is bit-for-bit matched to the LeRobot `PI0Policy`
-reference (`max|Δ| = 7.15e-07`, CPU/float32, fixed noise; see
-`tests/diffusion/models/pi0/test_pi0_parity.py::test_pi0_vllm_omni_vs_lerobot`).
-
-## References
-
- Model blog: <https://www.physicalintelligence.company/blog/pi0>
- Checkpoint: <https://huggingface.co/lerobot/pi0_base>
- Pipeline: [`vllm_omni/diffusion/models/pi0/pipeline_pi0.py`](../../vllm_omni/diffusion/models/pi0/pipeline_pi0.py)
- Deploy config: [`vllm_omni/deploy/pi0.yaml`](../../vllm_omni/deploy/pi0.yaml)
- Example client: [`examples/online_serving/pi0/`](../../examples/online_serving/pi0/)
- Tests: CPU units + LeRobot parity in [`tests/diffusion/models/pi0/`](../../tests/diffusion/models/pi0/),
-  OpenPI websocket e2e in
-  [`tests/e2e/online_serving/test_pi0_expansion.py`](../../tests/e2e/online_serving/test_pi0_expansion.py)
@@ -9,17 +9,13 @@ from pathlib import Path
 import pytest

 from vllm_omni.benchmarks.data_modules.daily_omni_dataset import (
-    DailyOmniDataset,
    daily_omni_local_qa_json,
    daily_omni_local_videos_dir,
-    ensure_daily_omni_hub_root,
    resolve_daily_omni_local_root,
 )

 pytestmark = [pytest.mark.core_model, pytest.mark.cpu, pytest.mark.benchmark]

-_HUB_ID = "liarliar/Daily-Omni"
-

 def _write_qa_json(root: Path) -> Path:
    qa = root / "qa.json"
@@ -27,28 +23,6 @@ def _write_qa_json(root: Path) -> Path:
    return qa


-def _stage_offline_hub_cache(tmp_path: Path, monkeypatch: pytest.MonkeyPatch, *, with_qa: bool = True) -> Path:
-    """Build a hub cache holding ``liarliar/Daily-Omni`` and cut off all network access.
-
-    Reproduces the air-gapped NPU runner: the dataset is pre-staged with ``hf download``, so it
-    only exists in the *hub* cache -- never in the ``datasets`` cache that ``load_dataset`` reads.
-    """
-    constants = pytest.importorskip("huggingface_hub").constants
-
-    cache = tmp_path / "hub"
-    repo_dir = cache / "datasets--liarliar--Daily-Omni"
-    snapshot = repo_dir / "snapshots" / "deadbeef"
-    snapshot.mkdir(parents=True)
-    if with_qa:
-        _write_qa_json(snapshot)
-    (repo_dir / "refs").mkdir()
-    (repo_dir / "refs" / "main").write_text("deadbeef", encoding="utf-8")
-
-    monkeypatch.setattr(constants, "HF_HUB_CACHE", str(cache))
-    monkeypatch.setattr(constants, "HF_HUB_OFFLINE", True)
-    return snapshot
-
-
 def test_resolve_local_root_returns_none_for_hub_id() -> None:
    assert resolve_daily_omni_local_root("liarliar/Daily-Omni") is None
    assert resolve_daily_omni_local_root("") is None
@@ -116,37 +90,3 @@ def test_missing_qa_json_reports_none(tmp_path: Path) -> None:
    root = resolve_daily_omni_local_root(str(tmp_path))
    assert daily_omni_local_qa_json(root) is None
    assert daily_omni_local_videos_dir(root) is None
-
-
-def test_hub_id_resolves_to_offline_cache(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
-    snapshot = _stage_offline_hub_cache(tmp_path, monkeypatch)
-
-    root = ensure_daily_omni_hub_root(_HUB_ID)
-
-    assert root == snapshot.resolve()
-    assert daily_omni_local_qa_json(root) == (snapshot / "qa.json").resolve()
-
-
-def test_dataset_loads_hub_id_without_network(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
-    """The regression this file exists for: a Hub id must not require ``load_dataset``."""
-    snapshot = _stage_offline_hub_cache(tmp_path, monkeypatch)
-    # Any use of `datasets` here means we went back to the path that cannot read the hub cache.
-    monkeypatch.setattr("vllm_omni.benchmarks.data_modules.daily_omni_dataset.load_dataset", None)
-
-    dataset = DailyOmniDataset(dataset_path=_HUB_ID, random_seed=0)
-
-    assert dataset.qa_json_path == (snapshot / "qa.json").resolve()
-    assert [row["video_id"] for row in dataset.data] == ["vid0"]
-
-
-def test_uncached_hub_id_reports_both_attempts(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
-    _stage_offline_hub_cache(tmp_path, monkeypatch, with_qa=False)
-    monkeypatch.setattr("vllm_omni.benchmarks.data_modules.daily_omni_dataset.load_dataset", None)
-
-    with pytest.raises(RuntimeError) as excinfo:
-        DailyOmniDataset(dataset_path="someone-else/Daily-Omni", random_seed=0)
-
-    message = str(excinfo.value)
-    assert "someone-else/Daily-Omni" in message
-    assert "snapshot_download failed" in message
-    assert "--daily-omni-qa-json" in message
@@ -1,119 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
-"""Guard the NPU pytest circular import through DiffusionOutput."""
-
-from __future__ import annotations
-
-import ast
-import os
-import subprocess
-import sys
-from pathlib import Path
-
-import pytest
-
-pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
-REPO_ROOT = Path(__file__).resolve().parents[2]
-
-
-def _top_level_import_modules(path: Path) -> set[str]:
-    tree = ast.parse(path.read_text(encoding="utf-8"))
-    modules: set[str] = set()
-    for node in tree.body:
-        if isinstance(node, ast.ImportFrom) and node.module:
-            modules.add(node.module)
-        elif isinstance(node, ast.Import):
-            modules.update(alias.name for alias in node.names)
-    return modules
-
-
-def _assert_isolated_import_succeeds(script: str) -> None:
-    env = os.environ.copy()
-    env.pop("CUDA_VISIBLE_DEVICES", None)
-    env.pop("HIP_VISIBLE_DEVICES", None)
-    result = subprocess.run(
-        [sys.executable, "-c", script],
-        cwd=REPO_ROOT,
-        env=env,
-        capture_output=True,
-        text=True,
-        timeout=120,
-        check=False,
-    )
-    assert result.returncode == 0, result.stderr
-
-
-def test_config_package_does_not_eagerly_import_pipeline_registry() -> None:
-    imported = _top_level_import_modules(REPO_ROOT / "vllm_omni" / "config" / "__init__.py")
-    assert "vllm_omni.config.config_factory" not in imported
-    assert "vllm_omni.config.pipeline_registry" not in imported
-
-
-def test_config_package_lazy_exports_still_resolve() -> None:
-    from vllm_omni.config import StageConfigFactory, register_pipeline
-    from vllm_omni.config.config_factory import StageConfigFactory as DirectStageConfigFactory
-    from vllm_omni.config.pipeline_registry import register_pipeline as direct_register_pipeline
-
-    assert StageConfigFactory is DirectStageConfigFactory
-    assert register_pipeline is direct_register_pipeline
-
-
-def test_npu_platform_defers_minimax_h3_encoder_patch() -> None:
-    src = (REPO_ROOT / "vllm_omni" / "platforms" / "npu" / "platform.py").read_text(encoding="utf-8")
-    init_start = src.index("def __init__(self) -> None:")
-    runtime_start = src.index("def init_diffusion_model_runner_runtime")
-    init_src = src[init_start:runtime_start]
-    runtime_src = src[runtime_start:]
-    assert "apply_minimax_h3_qwen3vl_patch" not in init_src
-    assert "apply_minimax_h3_qwen3vl_patch" in runtime_src
-    assert "apply_minimax_h3_qwen3vl_swiglu_patch" in runtime_src
-
-
-def test_lora_config_import_does_not_load_pipeline_registry() -> None:
-    """Import LoRAConfig and diffusion.data in a fresh interpreter.
-
-    The NPU failure was ``data.py`` importing LoRAConfig while still loading,
-    then ``vllm_omni.config`` re-entering ``pipeline_registry`` → PI0 →
-    DiffusionOutput. This child process must not preload those exports.
-    """
-    _assert_isolated_import_succeeds(
-        """
-import sys
-
-from vllm_omni.config.lora import LoRAConfig
-
-if "vllm_omni.config.pipeline_registry" in sys.modules:
-    raise SystemExit("LoRAConfig import loaded pipeline_registry")
-if "vllm_omni.diffusion.models.pi0.pipeline_pi0" in sys.modules:
-    raise SystemExit("LoRAConfig import loaded PI0_PIPELINE")
-
-from vllm_omni.diffusion.data import DiffusionOutput
-
-if DiffusionOutput is None or LoRAConfig is None:
-    raise SystemExit("expected LoRAConfig and DiffusionOutput")
-"""
-    )
-
-
-def test_pipeline_registry_does_not_import_pi0_runtime() -> None:
-    """Resolving the pi0 topology must not import its runtime pipeline.
-
-    ``pipeline_pi0`` imports ``diffusion.data``. Importing it while
-    ``diffusion.data`` is only partially initialized closes the Ascend plugin's
-    circular-import loop, so the registry must depend only on a lightweight
-    pipeline-config module.
-    """
-    _assert_isolated_import_succeeds(
-        """
-import sys
-
-from vllm_omni.config.pipeline_registry import resolve_pipeline_config
-
-if "vllm_omni.diffusion.models.pi0.pipeline_pi0" in sys.modules:
-    raise SystemExit("pipeline_registry imported the pi0 runtime pipeline")
-
-pipeline = resolve_pipeline_config("pi0")
-if pipeline is None or pipeline.model_type != "pi0":
-    raise SystemExit("expected the registered pi0 pipeline config")
-"""
-    )
@@ -5,20 +5,11 @@
 from __future__ import annotations

 from dataclasses import fields
-from inspect import Parameter, signature
 from pathlib import Path

-import msgspec
 import pytest
 from pydantic import ValidationError
-from pydantic.fields import FieldInfo
 from transformers import Qwen3OmniMoeConfig
-from vllm.config import CacheConfig as VllmCacheConfig
-from vllm.config import CompilationConfig as VllmCompilationConfig
-from vllm.config import LoadConfig as VllmLoadConfig
-from vllm.config import ParallelConfig as VllmParallelConfig
-from vllm.config import ProfilerConfig as VllmProfilerConfig
-from vllm.config import SchedulerConfig as VllmSchedulerConfig

 from tests.helpers.stage_config import get_deploy_config_path
 from vllm_omni.config import omni_config as omni_config_module
@@ -49,7 +40,6 @@ from vllm_omni.config.stage_config import (
    merge_pipeline_deploy,
 )
 from vllm_omni.diffusion.diffusion_kv.config import DiffusionKVCacheMode
-from vllm_omni.engine.stage_engine_startup import _serialize_stage_config
 from vllm_omni.engine.stage_init_utils import build_legacy_engine_args_dict

 pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
@@ -253,29 +243,9 @@ def test_from_pipeline_config_rejects_unowned_deploy_engine_extras(engine_extras
    ],
    ids=["global", "ar-stage", "generation-stage"],
 )
-def test_from_pipeline_config_accepts_upstream_cache_dtype_for_llm_stages(cli_overrides, stage_id):
-    stage = _from_pipeline_key("qwen3_tts", cli_overrides=cli_overrides).stage_by_id(stage_id)
-
-    assert stage.cache_config.cache_dtype == "fp8"
-    assert "cache_dtype" in stage.cache_config._omni_explicit_fields
-
-
-@pytest.mark.parametrize("field_name", ["data_parallel_master_ip", "data_parallel_address"])
-def test_from_pipeline_config_normalizes_nested_parallel_config_aliases(field_name):
-    pipeline = _resolve_pipeline_or_skip("qwen3_tts")
-    deploy = DeployConfig(
-        stages=[
-            StageDeployConfig(
-                stage_id=0,
-                engine_extras={"parallel_config": {field_name: "10.0.0.1"}},
-            )
-        ]
-    )
-
-    stage = VllmOmniConfig.from_pipeline_config(pipeline, user_deploy_config=deploy).stage_by_id(0)
-
-    assert stage.parallel_config.data_parallel_master_ip == "10.0.0.1"
-    assert "data_parallel_master_ip" in stage.parallel_config._omni_explicit_fields
+def test_from_pipeline_config_rejects_diffusion_only_cli_fields_for_llm_stages(cli_overrides, stage_id):
+    with pytest.raises(ValueError, match=rf"Stage {stage_id} .*no structured config owner: kv_cache_dtype"):
+        _from_pipeline_key("qwen3_tts", cli_overrides=cli_overrides)


 def test_from_pipeline_config_accepts_diffusion_only_cli_fields_for_diffusion_stage():
@@ -467,8 +437,6 @@ def test_vllm_omni_stage_config_public_fields_use_typed_stage_realizations():
        "connector_config",
        "runtime_config",
        "parallel_config",
-        "compilation_config",
-        "profiler_config",
        "quantization_config",
    }
    assert "diffusion_config" not in public_fields
@@ -487,47 +455,10 @@ def test_runtime_config_fields_match_structured_runtime_scope():
        "num_gpus",
        "log_level",
        "log_stats",
+        "profiler_config",
    }


-def test_upstream_config_compatible_fields_materialize_mapping_inputs():
-    compilation_config = {"backend": "eager"}
-    profiler_config = {"profiler": "cuda"}
-
-    stage_config = VllmOmniARStageConfig(
-        stage_pipeline_config=_resolve_pipeline_or_skip("qwen3_tts").stages[0],
-        compilation_config=compilation_config,
-        profiler_config=profiler_config,
-    )
-
-    assert isinstance(stage_config.compilation_config, VllmCompilationConfig)
-    assert stage_config.compilation_config.backend == "eager"
-    assert isinstance(stage_config.profiler_config, VllmProfilerConfig)
-    assert stage_config.profiler_config.profiler == "cuda"
-
-
-def test_upstream_config_compatible_fields_validate_mapping_inputs():
-    with pytest.raises(ValidationError):
-        VllmOmniARStageConfig(
-            stage_pipeline_config=_resolve_pipeline_or_skip("qwen3_tts").stages[0],
-            profiler_config={"delay_iterations": -1},
-        )
-
-
-def test_upstream_config_compatible_fields_keep_prebuilt_config_types():
-    compilation_config = VllmCompilationConfig(backend="eager")
-    profiler_config = VllmProfilerConfig(profiler="cuda")
-
-    stage_config = VllmOmniARStageConfig(
-        stage_pipeline_config=_resolve_pipeline_or_skip("qwen3_tts").stages[0],
-        compilation_config=compilation_config,
-        profiler_config=profiler_config,
-    )
-
-    assert stage_config.compilation_config is compilation_config
-    assert stage_config.profiler_config is profiler_config
-
-
 def test_sub_config_fields_match_structured_scopes():
    assert {f.name for f in fields(OmniStageModelConfig)} == {
        "model",
@@ -558,6 +489,7 @@ def test_sub_config_fields_match_structured_scopes():
        "enforce_eager",
        "max_cudagraph_capture_size",
        "enable_flashinfer_autotune",
+        "compilation_config",
        "enable_multithread_weight_load",
        "num_weight_load_threads",
        "disable_autocast",
@@ -567,25 +499,29 @@ def test_sub_config_fields_match_structured_scopes():
        "model_subdir",
        "tokenizer_subdir",
    }
-    vllm_load_fields = {f.name for f in fields(VllmLoadConfig)}
-    assert issubclass(OmniStageLoadConfig, VllmLoadConfig)
-    assert {f.name for f in fields(OmniStageLoadConfig)} == vllm_load_fields | {
+    assert {f.name for f in fields(OmniStageLoadConfig)} == {
        "tokenizer",
+        "download_dir",
        "skip_tokenizer_init",
+        "load_format",
        "tokenizer_mode",
        "config_format",
        "skip_mm_profiling",
    }
-    assert OmniStageLoadConfig(load_format="PT").load_format == "pt"
-    assert issubclass(OmniStageCacheConfig, VllmCacheConfig)
-    assert {f.name for f in fields(OmniStageCacheConfig)} == {f.name for f in fields(VllmCacheConfig)} | {
+    assert {f.name for f in fields(OmniStageCacheConfig)} == {
+        "kv_cache_memory_bytes",
+        "gpu_memory_utilization",
+        "enable_prefix_caching",
        "disable_hybrid_kv_cache_manager",
        "mm_processor_cache_gb",
        "mamba_ssm_cache_dtype",
    }
-    assert issubclass(OmniStageSchedulerConfig, VllmSchedulerConfig)
-    assert {f.name for f in fields(OmniStageSchedulerConfig)} == {f.name for f in fields(VllmSchedulerConfig)} | {
+    assert {f.name for f in fields(OmniStageSchedulerConfig)} == {
+        "max_num_seqs",
+        "max_num_batched_tokens",
        "max_model_len",
+        "enable_chunked_prefill",
+        "async_scheduling",
    }
    assert {f.name for f in fields(OmniStageConnectorConfig)} == {
        "async_chunk",
@@ -594,10 +530,17 @@ def test_sub_config_fields_match_structured_scopes():
        "output_connectors",
        "input_connectors",
    }
-    vllm_parallel_fields = {f.name for f in fields(VllmParallelConfig)}
-    assert issubclass(OmniStageParallelConfig, VllmParallelConfig)
-    assert {f.name for f in fields(OmniStageParallelConfig)} == vllm_parallel_fields
-    assert {f.name for f in fields(OmniStageDiffusionParallelConfig)} == vllm_parallel_fields | {
+    assert {f.name for f in fields(OmniStageParallelConfig)} == {
+        "pipeline_parallel_size",
+        "data_parallel_size",
+        "tensor_parallel_size",
+        "enable_expert_parallel",
+        "world_size",
+    }
+    assert {f.name for f in fields(OmniStageDiffusionParallelConfig)} == {
+        "pipeline_parallel_size",
+        "data_parallel_size",
+        "tensor_parallel_size",
        "sequence_parallel_size",
        "ulysses_degree",
        "ring_degree",
@@ -611,86 +554,20 @@ def test_sub_config_fields_match_structured_scopes():
        "mask_sp_padding",
        "hsdp_shard_size",
        "hsdp_replicate_size",
+        "enable_expert_parallel",
+        "world_size",
    }


-def test_inherited_sub_configs_initialize_transport_safe_derived_fields():
-    scheduler_config = OmniStageSchedulerConfig(max_num_batched_tokens=4096)
-    assert scheduler_config.max_num_encoder_input_tokens == 4096
-    assert scheduler_config.encoder_cache_size == 4096
-
-    unresolved_scheduler_config = OmniStageSchedulerConfig()
-    assert unresolved_scheduler_config.max_num_encoder_input_tokens is None
-    assert unresolved_scheduler_config.encoder_cache_size is None
-
-    parallel_config = OmniStageParallelConfig(data_parallel_size=3, data_parallel_rank=2)
-    assert parallel_config.data_parallel_index == 2
-
-    diffusion_parallel_config = OmniStageDiffusionParallelConfig(data_parallel_size=3, data_parallel_rank=2)
-    assert diffusion_parallel_config.data_parallel_index == 2
-
-
-@pytest.mark.parametrize(
-    "config_cls",
-    [
-        OmniStageLoadConfig,
-        OmniStageCacheConfig,
-        OmniStageSchedulerConfig,
-        OmniStageParallelConfig,
-        OmniStageDiffusionParallelConfig,
-    ],
-)
-def test_inherited_sub_configs_are_keyword_only(config_cls):
-    assert all(parameter.kind is Parameter.KEYWORD_ONLY for parameter in signature(config_cls).parameters.values())
-
-    with pytest.raises(TypeError):
-        config_cls(1)
-
-
-@pytest.mark.parametrize(
-    "config_cls",
-    [
-        OmniStageLoadConfig,
-        OmniStageCacheConfig,
-        OmniStageSchedulerConfig,
-        OmniStageParallelConfig,
-        OmniStageDiffusionParallelConfig,
-    ],
-)
-def test_inherited_sub_configs_remain_msgpack_transport_safe(config_cls):
-    config = config_cls()
-
-    uninitialized_fields = [
-        config_field.name
-        for config_field in fields(config)
-        if isinstance(getattr(config, config_field.name), FieldInfo)
-    ]
-    assert uninitialized_fields == []
-
-    msgspec.msgpack.encode(_serialize_stage_config(config))
-
-
-def test_structured_llm_stage_registration_payloads_remain_msgpack_transport_safe():
-    omni_config = _from_pipeline_key("qwen3_tts")
-
-    for stage_config in omni_config.stage_configs:
-        payload = {
-            "stage_id": stage_config.stage_id,
-            "stage_config": _serialize_stage_config(stage_config),
-        }
-        msgspec.msgpack.encode(payload)
-
-
 def test_diffusion_parallel_config_fields_cover_legacy_surface():
    from vllm_omni.diffusion.data import DiffusionParallelConfig

    legacy_fields = {f.name for f in fields(DiffusionParallelConfig)}
    structured_fields = {f.name for f in fields(OmniStageDiffusionParallelConfig)}
    expected_upstream_fields = {"mask_sp_padding"}
-    vllm_parallel_fields = {f.name for f in fields(VllmParallelConfig)}

    assert legacy_fields | expected_upstream_fields <= structured_fields
-    assert structured_fields - legacy_fields - expected_upstream_fields - vllm_parallel_fields == set()
+    assert structured_fields - legacy_fields - expected_upstream_fields == {"world_size"}


 def test_diffusion_parallel_config_keeps_current_diffusion_parallel_surface():
@@ -908,52 +785,6 @@ def test_structured_diffusion_config_rejects_invalid_compile_granularity():
        omni_config_module._DiffusionConfigProjection(diffusion_compile_granularity="block")


-@pytest.mark.parametrize(
-    ("enable_dlo", "use_allgather"),
-    [(False, False), (True, True)],
-)
-@pytest.mark.parametrize(
-    ("field_name", "field_value"),
-    [
-        ("dlo_use_host_weight_cache", True),
-        ("dlo_host_weight_cache_pin_limit_gib", 1.0),
-    ],
-)
-def test_structured_diffusion_config_restricts_host_weight_cache_to_no_allgather_dlo(
-    enable_dlo,
-    use_allgather,
-    field_name,
-    field_value,
-):
-    with pytest.raises(ValueError, match="requires no-AllGather distributed layerwise offload"):
-        omni_config_module._DiffusionConfigProjection(
-            enable_distributed_layerwise_offload=enable_dlo,
-            dlo_use_allgather=use_allgather,
-            **{field_name: field_value},
-        )
-
-
-def test_structured_diffusion_config_requires_cache_for_registration_ceiling():
-    with pytest.raises(ValueError, match="requires dlo_use_host_weight_cache=True"):
-        omni_config_module._DiffusionConfigProjection(
-            enable_distributed_layerwise_offload=True,
-            dlo_use_allgather=False,
-            dlo_host_weight_cache_pin_limit_gib=1.0,
-        )
-
-
-@pytest.mark.parametrize(
-    "field_name",
-    [
-        "dlo_host_weight_cache_lock_timeout",
-        "dlo_host_weight_cache_pin_limit_gib",
-    ],
-)
-def test_structured_diffusion_config_rejects_non_finite_host_weight_cache_values(field_name):
-    with pytest.raises(ValueError, match=rf"{field_name} must be finite"):
-        omni_config_module._DiffusionConfigProjection(**{field_name: float("inf")})
-
-
 def test_from_pipeline_config_matches_stage_config_to_omegaconf_behavior_for_representative_stage():
    pipeline = _resolve_pipeline_or_skip("qwen3_tts")
    legacy_stage = merge_pipeline_deploy(pipeline, _load_default_deploy(pipeline))[0]
@@ -1247,21 +1078,3 @@ def test_diffusion_config_projection_keeps_mapping_quantization_config_serializa
    cfg = omni_config_module._DiffusionConfigProjection.from_kwargs(quantization_config=quantization_config)

    assert cfg.quantization_config == quantization_config
-
-
-def test_diffusion_quantization_mapping_reaches_terminal_config(monkeypatch):
-    from vllm_omni.diffusion.data import OmniDiffusionConfig
-
-    quantization_config = {"method": "int8", "activation_scheme": "dynamic"}
-    cfg = omni_config_module._DiffusionConfigProjection.from_kwargs(
-        quantization_config=quantization_config,
-    )
-
-    # Exercise the terminal construction performed by the future typed startup
-    # path without probing ports or loading remote model metadata.
-    monkeypatch.setattr(OmniDiffusionConfig, "_resolve_master_port", lambda _self: 29500)
-    monkeypatch.setattr(OmniDiffusionConfig, "enrich_config", lambda _self: None)
-    cfg.enrich_config()
-
-    assert cfg.quantization_config is not None
-    assert cfg.quantization_config.get_name() == "int8"
@@ -4,10 +4,7 @@ from types import SimpleNamespace
 import pytest
 from vllm.v1.engine import FinishReason

-from vllm_omni.core.sched.omni_scheduler_mixin import (
-    DEFAULT_INPUT_WAIT_TIMEOUT_S,
-    OmniSchedulerMixin,
-)
+from vllm_omni.core.sched.omni_scheduler_mixin import OmniSchedulerMixin
 from vllm_omni.core.sched.output import OmniChunkRecvHandle

 pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
@@ -25,25 +22,13 @@ def test_schedule_lifecycle_helpers_process_and_restore_both_input_paths():
    scheduler.requests = {"request": object()}
    scheduler._consume_pending_connector_output = lambda mode: calls.append(("consume", mode))
    scheduler._process_pending_input_timeouts = lambda: calls.append(("timeouts",))
-
-    def _collect_timed_out(timeout_s):
-        calls.append(("chunk-timeouts", timeout_s))
-        return set()
-
-    def _collect_failed_sends():
-        calls.append(("failed-sends",))
-        return {}
-
    scheduler.chunk_transfer_adapter = SimpleNamespace(
-        receives_chunks=True,
        process_pending_chunks=lambda waiting, running, scheduler_requests: calls.append(
            ("process", waiting, running, scheduler_requests)
        ),
        restore_queues=lambda waiting, running, scheduler_requests: calls.append(
            ("restore-chunks", waiting, running, scheduler_requests)
        ),
-        collect_timed_out_request_ids=_collect_timed_out,
-        collect_failed_send_request_ids=_collect_failed_sends,
    )
    scheduler.input_coordinator = SimpleNamespace(
        restore_queues=lambda waiting: calls.append(("restore-full", waiting))
@@ -56,10 +41,6 @@ def test_schedule_lifecycle_helpers_process_and_restore_both_input_paths():
        ("consume", "ar"),
        ("timeouts",),
        ("process", scheduler.waiting, scheduler.running, scheduler.requests),
-        # The chunk deadline runs after chunks are applied, so a chunk that
-        # arrived this cycle resets the clock before it is measured (R1.1).
-        ("chunk-timeouts", DEFAULT_INPUT_WAIT_TIMEOUT_S),
-        ("failed-sends",),
        ("restore-chunks", scheduler.waiting, scheduler.running, scheduler.requests),
        ("restore-full", scheduler.waiting),
    ]
@@ -1,10 +1,6 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""Unit coverage for the two stage-input timeout hooks.
-
-``_process_pending_input_timeouts`` covers the full-payload transport;
-``_process_pending_chunk_timeouts`` covers async-chunk. Both share
-``VLLM_OMNI_INPUT_WAIT_TIMEOUT_S``.
+"""Unit coverage for _process_pending_input_timeouts.

 Verifies that the mixin correctly *delegates* timed-out requests to the
 base scheduler's ``finish_requests`` API with ``RequestStatus.FINISHED_ERROR``.
@@ -16,10 +12,6 @@ this file only asserts the wiring from the mixin to that API.

 from __future__ import annotations

-import importlib
-import logging
-import math
-import os
 from types import SimpleNamespace

 import pytest
@@ -110,248 +102,5 @@ def test_process_pending_input_timeouts_disabled_when_timeout_zero(monkeypatch):
    assert scheduler.finish_calls == []


-class _FakeChunkAdapter:
-    """Stands in for OmniChunkTransferAdapter's timeout-collection surface."""
-
-    def __init__(self, timed_out_ids, *, receives_chunks=True):
-        self._timed_out_ids = set(timed_out_ids)
-        self.receives_chunks = receives_chunks
-        self.calls = []
-
-    def collect_timed_out_request_ids(self, timeout_s):
-        self.calls.append(timeout_s)
-        return set(self._timed_out_ids)
-
-
-class _FakeChunkScheduler(OmniSchedulerMixin):
-    def __init__(self, requests, adapter):
-        self.requests = requests
-        self.chunk_transfer_adapter = adapter
-        self.finish_calls = []
-
-    def finish_requests(self, req_ids, status):
-        self.finish_calls.append((set(req_ids), status))
-
-
-def test_process_pending_chunk_timeouts_delegates_to_finish_requests():
-    """A request stalled in WAITING_FOR_CHUNK is failed, not left parked.
-
-    This is the async-chunk half of the safety net (RFC #4855 R1.1); before it
-    existed, the chunk path had no deadline at all and issue #3833 could park a
-    request forever.
-    """
-    req_id = "stalled-stream"
-    adapter = _FakeChunkAdapter(timed_out_ids={req_id})
-    scheduler = _FakeChunkScheduler({req_id: SimpleNamespace(request_id=req_id)}, adapter)
-
-    scheduler._process_pending_chunk_timeouts()
-
-    assert len(adapter.calls) == 1
-    assert adapter.calls[0] > 0
-    assert len(scheduler.finish_calls) == 1
-    finished_ids, status = scheduler.finish_calls[0]
-    assert finished_ids == {req_id}
-    assert getattr(status, "name", str(status)).endswith("FINISHED_ERROR")
-
-
-def test_process_pending_chunk_timeouts_skips_already_freed_request():
-    adapter = _FakeChunkAdapter(timed_out_ids={"already-freed"})
-    scheduler = _FakeChunkScheduler(requests={}, adapter=adapter)
-
-    scheduler._process_pending_chunk_timeouts()
-
-    assert scheduler.finish_calls == []
-
-
-def test_process_pending_chunk_timeouts_noop_on_a_producer_stage():
-    """Stage 0 does not receive chunks, so it has nothing to expire."""
-    adapter = _FakeChunkAdapter(timed_out_ids={"r1"}, receives_chunks=False)
-    scheduler = _FakeChunkScheduler({"r1": SimpleNamespace(request_id="r1")}, adapter)
-
-    scheduler._process_pending_chunk_timeouts()
-
-    assert adapter.calls == []
-    assert scheduler.finish_calls == []
-
-
-def test_process_pending_chunk_timeouts_noop_without_adapter():
-    class _NoAdapter(OmniSchedulerMixin):
-        def __init__(self):
-            self.requests = {}
-            self.chunk_transfer_adapter = None
-            self.finish_calls = []
-
-        def finish_requests(self, req_ids, status):
-            self.finish_calls.append((set(req_ids), status))
-
-    scheduler = _NoAdapter()
-    scheduler._process_pending_chunk_timeouts()
-    assert scheduler.finish_calls == []
-
-
-def test_process_pending_chunk_timeouts_disabled_when_timeout_zero(monkeypatch):
-    """The shared knob disables both transports' nets together."""
-    from vllm_omni.core.sched import omni_scheduler_mixin
-
-    monkeypatch.setattr(omni_scheduler_mixin, "DEFAULT_INPUT_WAIT_TIMEOUT_S", 0.0)
-
-    adapter = _FakeChunkAdapter(timed_out_ids={"r1"})
-    scheduler = _FakeChunkScheduler({"r1": SimpleNamespace(request_id="r1")}, adapter)
-    scheduler._process_pending_chunk_timeouts()
-
-    assert adapter.calls == []
-    assert scheduler.finish_calls == []
-
-
-class _FakeSendFailureAdapter:
-    """Stands in for the adapter's drained send-failure map (R1.2)."""
-
-    def __init__(self, failures):
-        self._failures = dict(failures)
-        self.collect_calls = 0
-
-    def collect_failed_send_request_ids(self):
-        self.collect_calls += 1
-        failures, self._failures = self._failures, {}
-        return failures
-
-
-class _FakeSendFailureScheduler(OmniSchedulerMixin):
-    def __init__(self, requests, adapter):
-        self.requests = requests
-        self.chunk_transfer_adapter = adapter
-
-
-def test_log_failed_chunk_sends_reports_a_live_request(caplog):
-    adapter = _FakeSendFailureAdapter({"req-live": "OSError: No space left on device"})
-    scheduler = _FakeSendFailureScheduler({"req-live": SimpleNamespace(request_id="req-live")}, adapter)
-
-    with caplog.at_level(logging.ERROR):
-        scheduler._log_failed_chunk_sends()
-
-    assert "req-live" in caplog.text
-    assert "No space left on device" in caplog.text
-
-
-def test_log_failed_chunk_sends_still_reports_a_departed_request(caplog):
-    """``collect_*`` drains the map, so a failure for a request that already left
-    this scheduler is the one place a drop could still go unrecorded."""
-    adapter = _FakeSendFailureAdapter({"req-gone": "OSError: No space left on device"})
-    scheduler = _FakeSendFailureScheduler({}, adapter)
-
-    with caplog.at_level(logging.WARNING):
-        scheduler._log_failed_chunk_sends()
-
-    assert adapter.collect_calls == 1
-    assert "req-gone" in caplog.text
-    assert "No space left on device" in caplog.text
-    assert "finished or aborted" in caplog.text
-
-
-def test_log_failed_chunk_sends_noop_without_adapter():
-    scheduler = _FakeSendFailureScheduler({}, None)
-    scheduler._log_failed_chunk_sends()
-
-
-# --- R1.4: VLLM_OMNI_INPUT_WAIT_TIMEOUT_S is parsed at import time, so each of
-# these reloads the module under a different environment. One knob now gates the
-# deadline on both transports, which is why a silent 0 matters more than it did
-# when only the full-payload path used it.
-
-
-@pytest.fixture
-def reload_with(monkeypatch):
-    """Reload the mixin module under a given env value, then put it back.
-
-    ``importlib.reload`` re-executes into the live module dict, so whatever the
-    last test parsed is what every later test in the session sees -- ``-k
-    zero_warns`` alone would leave ``DEFAULT_INPUT_WAIT_TIMEOUT_S = 0.0`` and
-    silently disable both deadlines. The rejection cases are worse: the
-    ValueError fires *after* the rebind, so ``nan`` lingers in the module.
-    ``monkeypatch`` restores the environment but not the module, so reload once
-    more on the way out.
-    """
-    import vllm_omni.core.sched.omni_scheduler_mixin as mod
-
-    def _reload(raw):
-        if raw is None:
-            monkeypatch.delenv("VLLM_OMNI_INPUT_WAIT_TIMEOUT_S", raising=False)
-        else:
-            monkeypatch.setenv("VLLM_OMNI_INPUT_WAIT_TIMEOUT_S", raw)
-        return importlib.reload(mod)
-
-    yield _reload
-
-    monkeypatch.undo()
-    importlib.reload(mod)
-
-
-@pytest.mark.parametrize(
-    "raw,expected",
-    [
-        (None, 600.0),
-        ("30", 30.0),
-        ("0.5", 0.5),
-        ("0", 0.0),  # explicit opt-out is still honoured
-    ],
-)
-def test_timeout_accepts_valid_values(reload_with, raw, expected):
-    mod = reload_with(raw)
-    assert mod.DEFAULT_INPUT_WAIT_TIMEOUT_S == expected
-
-
-@pytest.mark.parametrize("raw", ["-1", "-600", "-0.5"])
-def test_negative_timeout_is_rejected(reload_with, raw):
-    """`-1` is a common "no limit" idiom; substituting 600 would give an operator
-    who meant "never time out" mysterious failures ten minutes in. Fail loudly."""
-    with pytest.raises(ValueError, match="negative"):
-        reload_with(raw)
-
-
-@pytest.mark.parametrize("raw", ["nan", "inf", "-inf", "1e999"])
-def test_non_finite_timeout_is_rejected(reload_with, raw):
-    """nan compares false against every deadline check and inf is unreachable;
-    both disable the net while looking like a number."""
-    with pytest.raises(ValueError, match="not finite"):
-        reload_with(raw)
-
-
-def test_zero_warns_and_names_both_transports(reload_with, caplog):
-    with caplog.at_level(logging.WARNING):
-        mod = reload_with("0")
-    assert mod.DEFAULT_INPUT_WAIT_TIMEOUT_S == 0.0
-    text = caplog.text
-    assert "DISABLED" in text
-    assert "WAITING_FOR_INPUT" in text and "WAITING_FOR_CHUNK" in text
-
-
-def test_unparsable_timeout_falls_back(reload_with):
-    mod = reload_with("not-a-number")
-    assert mod.DEFAULT_INPUT_WAIT_TIMEOUT_S == 600.0
-
-
-def test_a_rejected_value_does_not_outlive_its_test(monkeypatch):
-    """The reason ``reload_with`` reloads on the way out, done here by hand.
-
-    ``nan`` is bound before the ValueError fires, so the reload that rejects it
-    still leaves it in the module -- and nan compares false against every
-    deadline check, which is exactly the silent-disable this item exists to
-    stop. Restoring the environment is not enough; the module has to be reloaded
-    too.
-    """
-    import vllm_omni.core.sched.omni_scheduler_mixin as mod
-
-    ambient = float(os.environ.get("VLLM_OMNI_INPUT_WAIT_TIMEOUT_S", "600"))
-
-    monkeypatch.setenv("VLLM_OMNI_INPUT_WAIT_TIMEOUT_S", "nan")
-    with pytest.raises(ValueError, match="not finite"):
-        importlib.reload(mod)
-    assert math.isnan(mod.DEFAULT_INPUT_WAIT_TIMEOUT_S)
-
-    monkeypatch.undo()
-    importlib.reload(mod)
-    assert mod.DEFAULT_INPUT_WAIT_TIMEOUT_S == ambient
-
-
 if __name__ == "__main__":
    pytest.main([__file__, "-v"])
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-stage BAGEL (TP=1, CFG=1), t2i Task",
@@ -61,7 +61,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-stage BAGEL (TP=1, CFG=1), i2i Task",
@@ -113,7 +113,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Two-stage BAGEL (bagel.yaml), TP=1, CFG=1, t2i Task",
@@ -164,7 +164,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Two-stage BAGEL (bagel.yaml), TP=1, CFG=1, i2i Task",
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "tts"
    ],
    "server_params": {
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-device baseline with enforce-eager (no torch.compile)",
@@ -64,7 +64,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-device with torch.compile (default, no enforce-eager)",
@@ -117,7 +117,7 @@
          "num_cards": 2
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "CFG-parallel=2 with enforce-eager",
@@ -172,7 +172,7 @@
          "num_cards": 2
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "CFG-parallel=2 with torch.compile",
@@ -226,7 +226,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "CacheDiT with enforce-eager",
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-device baseline (two input images)",
@@ -76,7 +76,7 @@
          "num_cards": 4
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Ulysses SP=2 + CFG=2 + VAE patch parallel=4",
@@ -125,7 +125,7 @@
          "num_cards": 4
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Ulysses SP=2 + CFG=2 + CacheDiT",
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-device baseline",
@@ -73,7 +73,7 @@
          "num_cards": 4
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Ulysses SP=2 + CFG=2 + VAE patch parallel=4",
@@ -121,7 +121,7 @@
          "num_cards": 4
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Ulysses SP=2 + CFG=2 + CacheDiT",
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "diffusion"
    ],
    "description": "Single-device baseline",
@@ -10,7 +10,7 @@
          "num_cards": 1
        }
      },
-      "local_model",
+      "full_model",
      "tts"
    ],
    "server_params": {
@@ -7,10 +7,12 @@ The detailed RFC document is maintained outside this repository (local/internal
 - Qwen3 tests: `test_reliability_qwen3_omni.py`
 - Wan2.2 tests: `test_reliability_wan22.py`
 - HunyuanImage tests: `test_reliability_hunyuan_image.py`
+- VoxCPM2 tests: `test_reliability_voxcpm2.py`

 ```bash
 pytest --collect-only tests/dfx/reliability
 pytest -s -v tests/dfx/reliability/test_reliability_qwen3_omni.py -m slow
 pytest -s -v tests/dfx/reliability/test_reliability_wan22.py -m slow
 pytest -s -v tests/dfx/reliability/test_reliability_hunyuan_image.py -m slow
+pytest -s -v tests/dfx/reliability/test_reliability_voxcpm2.py -m slow
 ```
@@ -976,68 +976,3 @@ def test_voices_delete_invalid_requests(
            "err_message": err_message,
        }
    )
-
-
-# ─── POST /v1/audio/speech · MiniMax Music 3 ───
-
-# Text-to-music on the speech endpoint. Most of the speech contract does not
-# apply: there is no speaker, no reference audio and no temperature, and the
-# lyrics alone are not a request. Each case below pins a rejection so a caller
-# who assumes the usual TTS shape finds out immediately.
-_MINIMAX_MUSIC3_SPEECH = [
-    pytest.param(
-        OmniServerParams(
-            model="MiniMaxAI/MiniMax-Music3",
-            stage_config_path=get_deploy_config_path("minimax_music3.yaml"),
-            server_args=_SPEECH_SERVER_ARGS,
-        ),
-        id="minimax_music3",
-        marks=hardware_marks(res={"cuda": "H100"}),
-    ),
-]
-
-_MINIMAX_MUSIC3_LYRICS = "[Verse]\nWalking down the empty street at midnight"
-_MINIMAX_MUSIC3_CAPTION = "A melancholic lo-fi hip-hop track at 85 BPM in F minor."
-# 25 frames per second, capped at six minutes of song.
-_MINIMAX_MUSIC3_MAX_FRAMES = 9000
-
-
-@pytest.mark.parametrize(
-    "overrides, err_message",
-    [
-        pytest.param({"instructions": None}, ("instructions",), id="caption_missing"),
-        pytest.param({"instructions": "   "}, ("instructions",), id="caption_blank"),
-        pytest.param({"input": "   "}, ("input",), id="lyrics_blank"),
-        pytest.param({"voice": "alloy"}, ("voice",), id="voice_rejected"),
-        pytest.param({"speed": 1.5}, ("speed",), id="speed_rejected"),
-        pytest.param(
-            {"max_new_tokens": _MINIMAX_MUSIC3_MAX_FRAMES + 1},
-            ("max_new_tokens", str(_MINIMAX_MUSIC3_MAX_FRAMES)),
-            id="max_new_tokens_over_cap",
-        ),
-        pytest.param({"max_new_tokens": 0}, ("max_new_tokens",), id="max_new_tokens_zero"),
-    ],
-)
-@pytest.mark.parametrize("omni_server", _MINIMAX_MUSIC3_SPEECH, indirect=True)
-def test_minimax_music3_speech_invalid_field_values(
-    omni_server: OmniServer,
-    openai_client: OpenAIClientHandler,
-    overrides: dict[str, object],
-    err_message: str | tuple[str, ...],
-) -> None:
-    body: dict[str, Any] = {
-        "model": omni_server.model,
-        "input": _MINIMAX_MUSIC3_LYRICS,
-        "instructions": _MINIMAX_MUSIC3_CAPTION,
-        "seed": 1,
-        "max_new_tokens": 250,
-        "response_format": "wav",
-    }
-    for key, value in overrides.items():
-        if value is None:
-            body.pop(key, None)
-        else:
-            body[key] = value
-    openai_client.send_audio_speech_http_request(
-        {"json": body, "timeout": 120, "err_code": 400, "err_message": err_message}
-    )
@@ -71,7 +71,7 @@ def _default_oom_device_spec() -> str:

 OOM_INJECTION_CONFIG = {
    "device": _default_oom_device_spec(),
-    "target_mem_ratio": 0.95,
+    "target_mem_ratio": 0.92,
    "hold_seconds": 0,
    "startup_timeout_sec": 20,
    "strict": False,
@@ -0,0 +1,508 @@
+"""
+VoxCPM2 reliability integration tests.
+
+Scenarios follow ``docs/user_guide/fault_injection_reliability_matrix.md``.
+"""
+
+from __future__ import annotations
+
+import concurrent.futures
+import os
+import time
+from pathlib import Path
+from typing import Any
+
+import pytest
+
+from tests.dfx.conftest import (
+    assert_fault_exception,
+    create_reliability_omni_server_params,
+    resolve_oom_device_spec,
+)
+from tests.dfx.reliability.helpers import (
+    FaultInjector,
+    assert_no_server_tree_process_residual_and_gpu_release,
+    extract_openai_error_contract_from_bytes,
+    get_health_raw,
+    inject_gpu_oom,
+    make_process_kill_fault_injector,
+    make_server_root_kill_fault_injector,
+    make_server_tree_kill_fault_injector,
+    post_json_raw,
+    run_fault_injection_with_rate_load,
+    stop_gpu_oom_hogs,
+    worker_residual_timeout_after_kill_signal,
+)
+from tests.helpers.mark import hardware_test
+from tests.helpers.media import load_test_audio_data_url
+from tests.helpers.runtime import OpenAIClientHandler
+
+RELIABILITY_SCENARIOS: list[dict[str, Any]] = [
+    {
+        "test_name": "voxcpm2_reliability_default",
+        "server_params": {
+            "model": "openbmb/VoxCPM2",
+            "stage_config_name": "voxcpm2.yaml",
+            # VoxCPM2 uses a custom HF tokenizer; CLI default trust_remote_code=False
+            # would otherwise override deploy YAML settings during server startup.
+            "server_args": ["--trust-remote-code", "--disable-log-stats"],
+        },
+    },
+]
+
+DEPLOY_CONFIGS_DIR = Path(__file__).resolve().parent.parent.parent.parent / "vllm_omni" / "deploy"
+OOM_INJECTION_CONFIG = {
+    "target_mem_ratio": 0.96,
+    "hold_seconds": 0,
+    "startup_timeout_sec": 20,
+    "strict": False,
+}
+FAULT_ERROR_KEYWORDS = (
+    "the request failed",
+    "oom",
+    "out of memory",
+    "cuda",
+    "orchestrator",
+    "timeout",
+    "connection",
+    "500",
+    "503",
+)
+PROCESS_KILL_ERROR_KEYWORDS = (
+    "timeout",
+    "did not complete within",
+    "connection",
+    "engine",
+    "orchestrator",
+    "dead",
+    "internal",
+    "500",
+    "503",
+)
+RUNTIME_WORKER_PATTERN = "VLLM::"
+SERVE_SIGNAL_PARAMS = [
+    pytest.param("SIGTERM", id="sigterm"),
+    pytest.param("SIGINT", id="sigint"),
+    pytest.param("SIGKILL", id="sigkill"),
+]
+SERVE_WITH_LOAD_SIGNAL_PARAMS = [
+    pytest.param("SIGTERM", id="sigterm"),
+    pytest.param("SIGINT", id="sigint"),
+    pytest.param("SIGKILL", id="sigkill"),
+]
+TREE_SIGNAL_PARAMS = [
+    pytest.param("SIGTERM", id="sigterm"),
+    pytest.param("SIGKILL", id="sigkill"),
+]
+TREE_WITH_LOAD_SIGNAL_PARAMS = [
+    pytest.param("SIGTERM", id="sigterm"),
+    pytest.param("SIGKILL", id="sigkill"),
+]
+WORKER_SIGNAL_FAULT_PARAMS = [
+    pytest.param(
+        make_process_kill_fault_injector(
+            grep_patterns=RUNTIME_WORKER_PATTERN,
+            signal_name="SIGTERM",
+            limit=1,
+            post_kill_wait_seconds=2.0,
+        ),
+        id="runtime_process_chain_sigterm",
+    ),
+    pytest.param(
+        make_process_kill_fault_injector(
+            grep_patterns=RUNTIME_WORKER_PATTERN,
+            signal_name="SIGKILL",
+            limit=1,
+            post_kill_wait_seconds=2.0,
+        ),
+        id="runtime_process_chain_sigkill",
+    ),
+]
+
+VOXCPM2_PARAMS = create_reliability_omni_server_params(RELIABILITY_SCENARIOS, DEPLOY_CONFIGS_DIR)
+INFLIGHT_INJECTION_REQUEST_RATE = 0.3
+INFLIGHT_INJECTION_REQUEST_COUNT = 10
+_SPEECH_FF_HTTP_TIMEOUT_SEC = 30
+_SPEECH_FF_MAX_ELAPSED_SEC = 30
+
+
+def _speech_request_config(omni_server: Any) -> dict[str, Any]:
+    return {
+        "model": omni_server.model,
+        "input": "The weather is nice today, perfect for a walk in the park.",
+        "stream": False,
+        "timeout": 120.0,
+        "response_format": "wav",
+        "voice": "default",
+    }
+
+
+def _submit_speech_request(openai_client: OpenAIClientHandler, omni_server: Any) -> None:
+    openai_client.send_audio_speech_request(_speech_request_config(omni_server), request_num=1)
+
+
+def _oom_speech_request_config(omni_server: Any) -> dict[str, Any]:
+    base = "The weather is nice today, perfect for a walk in the park. "
+    return {
+        "model": omni_server.model,
+        "input": base * 80,
+        "max_new_tokens": 4096,
+        "ref_audio": load_test_audio_data_url("qwen3_tts/clone_2.wav"),
+        "stream": False,
+        "timeout": 300.0,
+        "response_format": "wav",
+        "voice": "default",
+    }
+
+
+def _submit_oom_speech_request(openai_client: OpenAIClientHandler, omni_server: Any) -> None:
+    openai_client.send_audio_speech_request(_oom_speech_request_config(omni_server), request_num=1)
+
+
+def _assert_post_fault_speech_fast_fail(host: str, port: int, *, model: str, scenario: str) -> None:
+    payload = {
+        "model": model,
+        "input": "post-fault fast-fail check",
+        "voice": "default",
+    }
+    start = time.monotonic()
+    try:
+        status, body = post_json_raw(host, port, "/v1/audio/speech", payload, timeout_sec=20)
+        elapsed = time.monotonic() - start
+        assert elapsed < 15, f"[{scenario} fast_fail] /v1/audio/speech did not fail fast: {elapsed:.2f}s"
+        assert status >= 500, (
+            f"[{scenario} fast_fail] expected server-side error after fault, got status={status}, body={body[:200]!r}"
+        )
+    except Exception:  # noqa: BLE001
+        elapsed = time.monotonic() - start
+        assert elapsed < 15, f"[{scenario} fast_fail] exception was too slow after fault: {elapsed:.2f}s"
+
+
+def _assert_post_fault_health_terminal(host: str, port: int, *, scenario: str) -> None:
+    deadline = time.monotonic() + 20.0
+    last_observation = ""
+    while time.monotonic() < deadline:
+        try:
+            status, body = get_health_raw(host, port, timeout_sec=5)
+            last_observation = f"http={status}, body={body[:200]!r}"
+            if status == 503:
+                return
+        except Exception as exc:  # noqa: BLE001
+            last_observation = f"exception={exc!r}"
+            return
+        time.sleep(0.5)
+    pytest.fail(f"[{scenario} health] no terminal post-fault health observed: {last_observation}")
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skip(reason="issue#4285")
+@hardware_test(res={"cuda": "L4"}, num_cards=1)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_gpu_oom_speech_request_failure(
+    omni_server_function,
+    openai_client_function,
+) -> None:
+    stage_config_path = getattr(omni_server_function, "stage_config_path", None)
+    device_spec = resolve_oom_device_spec(OOM_INJECTION_CONFIG, stage_config_path)
+    handle = inject_gpu_oom(
+        device=device_spec,
+        target_mem_ratio=OOM_INJECTION_CONFIG["target_mem_ratio"],
+        hold_seconds=OOM_INJECTION_CONFIG["hold_seconds"],
+        startup_timeout_sec=OOM_INJECTION_CONFIG["startup_timeout_sec"],
+        strict=OOM_INJECTION_CONFIG["strict"],
+    )
+    try:
+        try:
+            _submit_oom_speech_request(openai_client_function, omni_server_function)
+        except Exception as exc:
+            assert_fault_exception(exc, FAULT_ERROR_KEYWORDS)
+        else:
+            pytest.fail("expected /v1/audio/speech request failure during GPU OOM injection")
+    finally:
+        stop_gpu_oom_hogs(handle)
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skip(reason="issue#4285")
+@hardware_test(res={"cuda": "L4"}, num_cards=1)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_gpu_oom_speech_error_contract(
+    omni_server_function,
+) -> None:
+    stage_config_path = getattr(omni_server_function, "stage_config_path", None)
+    device_spec = resolve_oom_device_spec(OOM_INJECTION_CONFIG, stage_config_path)
+    handle = inject_gpu_oom(
+        device=device_spec,
+        target_mem_ratio=OOM_INJECTION_CONFIG["target_mem_ratio"],
+        hold_seconds=OOM_INJECTION_CONFIG["hold_seconds"],
+        startup_timeout_sec=OOM_INJECTION_CONFIG["startup_timeout_sec"],
+        strict=OOM_INJECTION_CONFIG["strict"],
+    )
+    host = omni_server_function.host
+    port = omni_server_function.port
+    oom_config = _oom_speech_request_config(omni_server_function)
+    payload = {
+        "model": oom_config["model"],
+        "input": oom_config["input"],
+        "voice": oom_config["voice"],
+        "max_new_tokens": oom_config["max_new_tokens"],
+        "ref_audio": oom_config["ref_audio"],
+    }
+    try:
+        status, body = post_json_raw(host, port, "/v1/audio/speech", payload, timeout_sec=300)
+    finally:
+        stop_gpu_oom_hogs(handle)
+
+    assert status >= 500, f"expected speech runtime error under OOM, got status={status}"
+    speech_error = extract_openai_error_contract_from_bytes(body)
+    assert speech_error is not None, f"speech error payload not OpenAI-compatible: {body[:300]!r}"
+    assert "code" in speech_error, f"speech error lacks code field: {speech_error!r}"
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize(
+    "fault_injector",
+    WORKER_SIGNAL_FAULT_PARAMS,
+    indirect=True,
+)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_request_failure(
+    omni_server_after_fault_function,
+    openai_client_function,
+) -> None:
+    try:
+        _submit_speech_request(openai_client_function, omni_server_after_fault_function)
+    except Exception as exc:
+        assert_fault_exception(exc, PROCESS_KILL_ERROR_KEYWORDS)
+    else:
+        pytest.fail("expected /v1/audio/speech request failure after process-kill injection")
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize(
+    "fault_injector",
+    WORKER_SIGNAL_FAULT_PARAMS,
+    indirect=True,
+)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_fast_fail_and_concurrent(
+    omni_server_after_fault_function,
+    openai_client_function,
+) -> None:
+    host = omni_server_after_fault_function.host
+    port = omni_server_after_fault_function.port
+    model = omni_server_after_fault_function.model
+    payload = {
+        "model": model,
+        "input": "Say hello in one short sentence.",
+        "voice": "default",
+    }
+
+    start = time.monotonic()
+    try:
+        status, body = post_json_raw(
+            host,
+            port,
+            "/v1/audio/speech",
+            payload,
+            timeout_sec=_SPEECH_FF_HTTP_TIMEOUT_SEC,
+        )
+        elapsed = time.monotonic() - start
+        assert elapsed <= _SPEECH_FF_MAX_ELAPSED_SEC + 1.0, (
+            f"[process_kill fast_fail] /v1/audio/speech did not fail fast after fault: {elapsed:.2f}s"
+        )
+        assert status >= 500, (
+            "[process_kill fast_fail] expected server-side failure after fault, "
+            f"got status={status}, body={body[:200]!r}"
+        )
+    except Exception:
+        elapsed = time.monotonic() - start
+        assert elapsed <= _SPEECH_FF_MAX_ELAPSED_SEC + 1.0, (
+            f"[process_kill fast_fail] /v1/audio/speech exception was too slow after fault: {elapsed:.2f}s"
+        )
+
+    start = time.monotonic()
+    with concurrent.futures.ThreadPoolExecutor(max_workers=3) as executor:
+        futures = [
+            executor.submit(_submit_speech_request, openai_client_function, omni_server_after_fault_function)
+            for _ in range(3)
+        ]
+        done, pending = concurrent.futures.wait(
+            futures,
+            timeout=30,
+            return_when=concurrent.futures.ALL_COMPLETED,
+        )
+
+    elapsed = time.monotonic() - start
+    assert not pending, f"[process_kill concurrent] some fault-time speech requests hung: pending={len(pending)}"
+    assert elapsed < 30, f"[process_kill concurrent] fault-time speech request convergence is too slow: {elapsed:.2f}s"
+
+    fault_observed = False
+    for future in done:
+        try:
+            future.result()
+        except Exception as exc:
+            fault_observed = True
+            assert_fault_exception(exc, PROCESS_KILL_ERROR_KEYWORDS)
+    assert fault_observed, (
+        "[process_kill concurrent] expected at least one /v1/audio/speech request to fail after fault"
+    )
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize(
+    "fault_injector",
+    WORKER_SIGNAL_FAULT_PARAMS,
+    indirect=True,
+)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_worker_with_load_request_failure(
+    omni_server_function,
+    openai_client_function,
+    fault_injector: FaultInjector,
+) -> None:
+    scenario = "kill_worker_with_load"
+    load_result = run_fault_injection_with_rate_load(
+        submit_request=lambda: _submit_speech_request(openai_client_function, omni_server_function),
+        inject_fault=lambda: fault_injector(omni_server_function),
+        num_requests=INFLIGHT_INJECTION_REQUEST_COUNT,
+        request_rate=INFLIGHT_INJECTION_REQUEST_RATE,
+        completion_timeout_sec=120.0,
+    )
+    assert load_result["failure_observed"], (
+        f"[{scenario}] expected at least one load request failure after fault; load_result={load_result}"
+    )
+    host = omni_server_function.host
+    port = omni_server_function.port
+    _assert_post_fault_speech_fast_fail(host, port, model=omni_server_function.model, scenario=scenario)
+    _assert_post_fault_health_terminal(host, port, scenario=scenario)
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@hardware_test(res={"cuda": "L4"}, num_cards=1)
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize("signal_name", SERVE_SIGNAL_PARAMS)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_serve_root_no_load_fast_fail_and_cleanup(
+    omni_server_function,
+    signal_name: str,
+) -> None:
+    scenario = f"kill_serve_root_no_load_{signal_name.lower()}"
+    injector = make_server_root_kill_fault_injector(signal_name=signal_name, post_kill_wait_seconds=2.0)
+    injector(omni_server_function)
+    host = omni_server_function.host
+    port = omni_server_function.port
+    _assert_post_fault_speech_fast_fail(host, port, model=omni_server_function.model, scenario=scenario)
+    _assert_post_fault_health_terminal(host, port, scenario=scenario)
+    assert_no_server_tree_process_residual_and_gpu_release(
+        omni_server_function,
+        scenario=scenario,
+        timeout_sec=worker_residual_timeout_after_kill_signal(signal_name),
+    )
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@hardware_test(res={"cuda": "L4"}, num_cards=1)
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize("signal_name", SERVE_WITH_LOAD_SIGNAL_PARAMS)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_serve_root_with_load_fast_fail_and_cleanup(
+    omni_server_function,
+    openai_client_function,
+    signal_name: str,
+) -> None:
+    scenario = f"kill_serve_root_with_load_{signal_name.lower()}"
+    injector = make_server_root_kill_fault_injector(signal_name=signal_name, post_kill_wait_seconds=2.0)
+    load_result = run_fault_injection_with_rate_load(
+        submit_request=lambda: _submit_speech_request(openai_client_function, omni_server_function),
+        inject_fault=lambda: injector(omni_server_function),
+        num_requests=INFLIGHT_INJECTION_REQUEST_COUNT,
+        request_rate=INFLIGHT_INJECTION_REQUEST_RATE,
+        completion_timeout_sec=300.0,
+    )
+    assert load_result["failure_observed"], (
+        f"[{scenario}] expected at least one load request failure after fault; load_result={load_result}"
+    )
+    host = omni_server_function.host
+    port = omni_server_function.port
+    _assert_post_fault_speech_fast_fail(host, port, model=omni_server_function.model, scenario=scenario)
+    _assert_post_fault_health_terminal(host, port, scenario=scenario)
+    assert_no_server_tree_process_residual_and_gpu_release(
+        omni_server_function,
+        scenario=scenario,
+        timeout_sec=worker_residual_timeout_after_kill_signal(signal_name),
+    )
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize("signal_name", TREE_SIGNAL_PARAMS)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_tree_no_load_fast_fail_and_cleanup(
+    omni_server_function,
+    signal_name: str,
+) -> None:
+    scenario = f"kill_serve_tree_no_load_{signal_name.lower()}"
+    injector = make_server_tree_kill_fault_injector(
+        signal_name=signal_name,
+        post_kill_wait_seconds=2.0,
+        inter_kill_wait_seconds=0.1,
+    )
+    injector(omni_server_function)
+    host = omni_server_function.host
+    port = omni_server_function.port
+    _assert_post_fault_speech_fast_fail(host, port, model=omni_server_function.model, scenario=scenario)
+    _assert_post_fault_health_terminal(host, port, scenario=scenario)
+    assert_no_server_tree_process_residual_and_gpu_release(
+        omni_server_function,
+        scenario=scenario,
+        timeout_sec=worker_residual_timeout_after_kill_signal(signal_name),
+    )
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.skipif(os.name == "nt", reason="process-kill injection helper is POSIX-only")
+@pytest.mark.parametrize("signal_name", TREE_WITH_LOAD_SIGNAL_PARAMS)
+@pytest.mark.parametrize("omni_server_function", VOXCPM2_PARAMS, indirect=True)
+def test_reliability_fault_process_kill_tree_with_load_fast_fail_and_cleanup(
+    omni_server_function,
+    openai_client_function,
+    signal_name: str,
+) -> None:
+    scenario = f"kill_serve_tree_with_load_{signal_name.lower()}"
+    injector = make_server_tree_kill_fault_injector(
+        signal_name=signal_name,
+        post_kill_wait_seconds=2.0,
+        inter_kill_wait_seconds=0.1,
+    )
+    load_result = run_fault_injection_with_rate_load(
+        submit_request=lambda: _submit_speech_request(openai_client_function, omni_server_function),
+        inject_fault=lambda: injector(omni_server_function),
+        num_requests=INFLIGHT_INJECTION_REQUEST_COUNT,
+        request_rate=INFLIGHT_INJECTION_REQUEST_RATE,
+        completion_timeout_sec=120.0,
+    )
+    assert load_result["failure_observed"], (
+        f"[{scenario}] expected at least one load request failure after fault; load_result={load_result}"
+    )
+    host = omni_server_function.host
+    port = omni_server_function.port
+    _assert_post_fault_speech_fast_fail(host, port, model=omni_server_function.model, scenario=scenario)
+    _assert_post_fault_health_terminal(host, port, scenario=scenario)
+    assert_no_server_tree_process_residual_and_gpu_release(
+        omni_server_function,
+        scenario=scenario,
+        timeout_sec=worker_residual_timeout_after_kill_signal(signal_name),
+    )
@@ -0,0 +1,69 @@
+"""
+VoxCPM2 stability: OmniServer + ``vllm bench serve --omni`` for a fixed duration.
+
+Configuration: ``tests/dfx/stability/tests/test_voxcpm2.json``.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import pytest
+
+from tests.dfx.conftest import (
+    create_benchmark_indices,
+    create_test_parameter_mapping,
+    create_unique_server_params,
+    load_configs,
+)
+from tests.dfx.stability.helpers import _run_one_vllm_bench_batch, run_stability_benchmark_loop
+
+STABILITY_DIR = Path(__file__).resolve().parent.parent
+DEPLOY_CONFIGS_DIR = STABILITY_DIR / "deploy"
+CONFIG_FILE_PATH = str(STABILITY_DIR / "tests" / "test_voxcpm2.json")
+DEFAULT_NUM_PROMPTS_PER_BATCH = 20
+STABILITY_SERVER_TIMEOUT_ARGS = ["--stage-init-timeout", "600"]
+
+try:
+    BENCHMARK_CONFIGS = load_configs(CONFIG_FILE_PATH)
+except FileNotFoundError:
+    BENCHMARK_CONFIGS = []
+
+test_params = create_unique_server_params(BENCHMARK_CONFIGS, DEPLOY_CONFIGS_DIR) if BENCHMARK_CONFIGS else []
+server_to_benchmark_mapping = create_test_parameter_mapping(BENCHMARK_CONFIGS) if BENCHMARK_CONFIGS else {}
+benchmark_indices = create_benchmark_indices(BENCHMARK_CONFIGS, server_to_benchmark_mapping)
+
+
+@pytest.mark.slow
+@pytest.mark.tts
+@pytest.mark.parametrize("omni_server", test_params, indirect=True)
+@pytest.mark.parametrize("stability_benchmark_params", benchmark_indices, indirect=True)
+def test_stability_voxcpm2(omni_server, stability_benchmark_params):
+    test_name = stability_benchmark_params["test_name"]
+    params = stability_benchmark_params["params"]
+    duration_sec = params.get("duration_sec", 300)
+    num_prompts_per_batch = params.get("num_prompts_per_batch", DEFAULT_NUM_PROMPTS_PER_BATCH)
+    request_rate = params.get("request_rate")
+    max_concurrency = params.get("max_concurrency")
+
+    bench_params = {
+        k: v
+        for k, v in params.items()
+        if k not in ("duration_sec", "request_rate", "max_concurrency", "num_prompts_per_batch")
+    }
+
+    result = run_stability_benchmark_loop(
+        host=omni_server.host,
+        port=omni_server.port,
+        model=omni_server.model,
+        duration_sec=duration_sec,
+        params=bench_params,
+        request_rate=request_rate,
+        max_concurrency=max_concurrency,
+        result_dir=str(STABILITY_DIR),
+        num_prompts_per_batch=num_prompts_per_batch,
+        run_one_batch=_run_one_vllm_bench_batch,
+    )
+
+    assert result.get("failed", 0) == 0, f"[{test_name}] Failed requests detected: {result.get('errors', [])}"
+    assert result.get("completed", 0) > 0, f"[{test_name}] No requests completed"
@@ -1,297 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-from types import SimpleNamespace
-from unittest.mock import Mock
-
-import pytest
-import torch
-from vllm.v1.kv_cache_interface import FullAttentionSpec
-
-import vllm_omni.diffusion.diffusion_kv.initialization as initialization
-import vllm_omni.diffusion.vllm_config as diffusion_vllm_config
-from vllm_omni.diffusion.diffusion_kv.config import DiffusionKVCacheMode
-from vllm_omni.diffusion.sched.request_scheduler import RequestScheduler
-from vllm_omni.quantization import build_quant_config
-
-pytestmark = [pytest.mark.core_model, pytest.mark.cpu, pytest.mark.diffusion]
-
-
-class _Executor:
-    def __init__(self) -> None:
-        self.specs = [{"layer0": object()}, {"layer0": object()}]
-        self.memory = [1024, 1024]
-        self.configs = None
-        self.profile_requests = None
-
-    def get_kv_cache_specs(self):
-        return self.specs
-
-    def determine_available_kv_memory(self, profile_requests):
-        self.profile_requests = profile_requests
-        return self.memory
-
-    def set_kv_cache_configs(self, configs) -> None:
-        self.configs = configs
-
-
-def _od_config(**overrides):
-    values = dict(
-        model="test-diffusion",
-        diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER,
-        dtype=torch.bfloat16,
-        max_model_len=64,
-        quantization_config=None,
-        tf_model_config=None,
-        enforce_eager=True,
-        is_moe=False,
-        additional_config={},
-        profiler_config=None,
-        parallel_config=SimpleNamespace(
-            tensor_parallel_size=1,
-            pipeline_parallel_size=1,
-            data_parallel_size=1,
-            enable_expert_parallel=False,
-        ),
-        gpu_memory_utilization=0.9,
-        kv_cache_memory_bytes=None,
-        max_num_batched_tokens=64,
-        num_gpus=1,
-    )
-    values.update(overrides)
-    return SimpleNamespace(**values)
-
-
-def test_control_plane_builds_worker_and_scheduler_configs(monkeypatch) -> None:
-    executor = _Executor()
-    od_config = _od_config(num_gpus=2)
-    vllm_config = SimpleNamespace(
-        model_config=SimpleNamespace(max_model_len=4096),
-        max_in_flight_tokens=128,
-    )
-    worker_configs = [object(), object()]
-    scheduler_kv_cache_config = object()
-    profile_requests = [object()]
-    monkeypatch.setattr(initialization, "create_diffusion_vllm_config", lambda *args, **kwargs: vllm_config)
-    build = Mock(return_value=(worker_configs, scheduler_kv_cache_config))
-    monkeypatch.setattr(
-        initialization,
-        "build_native_kv_cache_configs",
-        build,
-    )
-    monkeypatch.setattr(initialization, "resolve_kv_cache_block_sizes", lambda *args: (16, 16))
-
-    result = initialization.initialize_diffusion_kv_control_plane(
-        executor,
-        od_config,
-        profile_requests=profile_requests,
-        device=torch.device("cpu"),
-    )
-
-    build.assert_called_once_with(vllm_config, executor.specs, executor.memory)
-    assert result == (scheduler_kv_cache_config, 16, 16, vllm_config)
-    assert executor.configs == worker_configs
-    assert executor.profile_requests is profile_requests
-
-
-def test_dense_mode_skips_worker_kv_initialization(monkeypatch) -> None:
-    executor = _Executor()
-    od_config = _od_config(diffusion_kv_mode=DiffusionKVCacheMode.DENSE_LEGACY)
-    get_device = Mock(side_effect=AssertionError("dense mode must not resolve a KV device"))
-    monkeypatch.setattr(initialization.current_omni_platform, "get_torch_device", get_device)
-
-    result = initialization.initialize_diffusion_kv_control_plane(executor, od_config)
-
-    assert result is None
-    assert executor.configs is None
-    get_device.assert_not_called()
-
-
-def test_control_plane_rejects_rank_count_mismatch(monkeypatch) -> None:
-    executor = _Executor()
-    executor.memory = [1024]
-    od_config = _od_config(num_gpus=2)
-    monkeypatch.setattr(initialization, "create_diffusion_vllm_config", lambda *args, **kwargs: object())
-
-    with pytest.raises(ValueError, match="rank count mismatch"):
-        initialization.initialize_diffusion_kv_control_plane(
-            executor,
-            od_config,
-            profile_requests=[object()],
-            device=torch.device("cpu"),
-        )
-
-
-def test_paged_control_plane_requires_prepared_profile_request() -> None:
-    with pytest.raises(ValueError, match="prepared memory profile requests"):
-        initialization.initialize_diffusion_kv_control_plane(
-            _Executor(),
-            _od_config(),
-            device=torch.device("cpu"),
-        )
-
-
-def test_native_config_builder_produces_matching_worker_and_scheduler_pools() -> None:
-    od_config = _od_config()
-    vllm_config = diffusion_vllm_config.create_diffusion_vllm_config(torch.device("cpu"), od_config)
-    spec = FullAttentionSpec(
-        block_size=vllm_config.cache_config.block_size,
-        num_kv_heads=2,
-        head_size=8,
-        dtype=torch.bfloat16,
-        non_causal=True,
-    )
-    num_blocks = 32
-
-    worker_configs, scheduler_kv_cache_config = initialization.build_native_kv_cache_configs(
-        vllm_config,
-        [{"layer0": spec}],
-        [spec.page_size_bytes * num_blocks],
-    )
-
-    assert len(worker_configs) == 1
-    assert worker_configs[0].num_blocks == num_blocks
-    assert scheduler_kv_cache_config.num_blocks == num_blocks
-    assert worker_configs[0].kv_cache_groups == scheduler_kv_cache_config.kv_cache_groups
-
-
-def test_native_control_plane_config_initializes_scheduler() -> None:
-    od_config = _od_config()
-    vllm_config = diffusion_vllm_config.create_diffusion_vllm_config(torch.device("cpu"), od_config)
-    spec = FullAttentionSpec(
-        block_size=vllm_config.cache_config.block_size,
-        num_kv_heads=2,
-        head_size=8,
-        dtype=torch.bfloat16,
-        non_causal=True,
-    )
-    _, scheduler_kv_cache_config = initialization.build_native_kv_cache_configs(
-        vllm_config,
-        [{"layer0": spec}],
-        [spec.page_size_bytes * 32],
-    )
-
-    scheduler = RequestScheduler()
-    scheduler.initialize(
-        od_config,
-        kv_cache_config=scheduler_kv_cache_config,
-        scheduler_block_size=spec.block_size,
-        hash_block_size=spec.block_size,
-        kv_vllm_config=vllm_config,
-    )
-
-    assert scheduler._diffusion_kv_manager is not None
-    scheduler.close()
-
-
-def test_dense_config_keeps_new_paged_sizing_inputs_inactive() -> None:
-    od_config = _od_config(
-        diffusion_kv_mode=DiffusionKVCacheMode.DENSE_LEGACY,
-        parallel_config=SimpleNamespace(
-            tensor_parallel_size=1,
-            pipeline_parallel_size=4,
-            data_parallel_size=1,
-            enable_expert_parallel=False,
-        ),
-        gpu_memory_utilization=0.5,
-        kv_cache_memory_bytes=4096,
-        max_num_batched_tokens=64,
-    )
-    vllm_config = diffusion_vllm_config.create_base_diffusion_vllm_config(torch.device("cpu"), od_config)
-    original_pp_size = vllm_config.parallel_config.pipeline_parallel_size
-    original_memory_utilization = vllm_config.cache_config.gpu_memory_utilization
-    original_memory_bytes = vllm_config.cache_config.kv_cache_memory_bytes
-    original_token_budget = vllm_config.scheduler_config.max_num_batched_tokens
-
-    diffusion_vllm_config.configure_diffusion_vllm_config(vllm_config, od_config)
-
-    assert vllm_config.parallel_config.pipeline_parallel_size == original_pp_size
-    assert vllm_config.cache_config.gpu_memory_utilization == original_memory_utilization
-    assert vllm_config.cache_config.kv_cache_memory_bytes == original_memory_bytes
-    assert vllm_config.scheduler_config.max_num_batched_tokens == original_token_budget
-
-
-def test_paged_config_forwards_gpu_memory_utilization_to_native_cache_config() -> None:
-    vllm_config = diffusion_vllm_config.create_diffusion_vllm_config(
-        torch.device("cpu"),
-        _od_config(gpu_memory_utilization=0.42),
-    )
-
-    assert vllm_config.cache_config.gpu_memory_utilization == 0.42
-
-
-def test_diffusion_vllm_model_config_supplies_dtype_for_quant_methods() -> None:
-    quantization_config = build_quant_config(
-        {
-            "quant_method": "modelopt",
-            "quant_algo": "FP8",
-            "ignore": [],
-        }
-    )
-    model_config = diffusion_vllm_config._make_diffusion_vllm_model_config(
-        _od_config(
-            diffusion_kv_mode=DiffusionKVCacheMode.DENSE_LEGACY,
-            quantization_config=quantization_config,
-        )
-    )
-
-    assert model_config.dtype is torch.bfloat16
-    assert model_config.quantization == "modelopt"
-    assert model_config.quantization_config is quantization_config
-    assert model_config.is_quantized
-    assert model_config.is_moe is False
-    assert model_config.is_nvfp4_quantized() is False
-    assert model_config.original_max_model_len == model_config.max_model_len
-
-    unquantized_model_config = diffusion_vllm_config._make_diffusion_vllm_model_config(
-        _od_config(
-            diffusion_kv_mode=DiffusionKVCacheMode.DENSE_LEGACY,
-            quantization_config=None,
-        )
-    )
-    assert not unquantized_model_config.is_quantized
-
-
-def test_minus_one_max_model_len_uses_model_limit() -> None:
-    od_config = SimpleNamespace(
-        max_model_len=-1,
-        tf_model_config=SimpleNamespace(max_position_embeddings=4096),
-        diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER,
-    )
-
-    assert diffusion_vllm_config.resolve_diffusion_max_model_len(od_config) == 4096
-
-    model_config = diffusion_vllm_config._make_diffusion_vllm_model_config(
-        _od_config(
-            max_model_len=-1,
-            tf_model_config=SimpleNamespace(max_position_embeddings=4096),
-        )
-    )
-    assert model_config.max_model_len == 4096
-    assert model_config.original_max_model_len == -1
-
-
-def test_minus_one_max_model_len_is_auto_fitted_by_native_cache_sizing() -> None:
-    vllm_config = diffusion_vllm_config.create_diffusion_vllm_config(
-        torch.device("cpu"),
-        _od_config(
-            max_model_len=-1,
-            tf_model_config=SimpleNamespace(max_position_embeddings=4096),
-        ),
-    )
-    spec = FullAttentionSpec(
-        block_size=vllm_config.cache_config.block_size,
-        num_kv_heads=2,
-        head_size=8,
-        dtype=torch.bfloat16,
-        non_causal=True,
-    )
-
-    initialization.build_native_kv_cache_configs(
-        vllm_config,
-        [{"layer0": spec}],
-        [spec.page_size_bytes * 16],
-    )
-
-    assert vllm_config.model_config.original_max_model_len == -1
-    assert vllm_config.model_config.max_model_len == 256
@@ -1,211 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-import pytest
-import torch
-import vllm.v1.core.single_type_kv_cache_manager as native_kv_managers
-from vllm.v1.kv_cache_interface import FullAttentionSpec, KVCacheConfig, KVCacheGroupSpec, KVCacheTensor
-
-from vllm_omni.diffusion.diffusion_kv.manager import DiffusionKVAdmissionError, DiffusionKVCacheManager
-from vllm_omni.diffusion.diffusion_kv.request import DiffusionKVContext, DiffusionKVRequest
-
-pytestmark = [pytest.mark.core_model, pytest.mark.cpu, pytest.mark.diffusion]
-
-BLOCK_SIZE = 4
-
-
-def _config(num_blocks: int) -> KVCacheConfig:
-    native_kv_managers.register_all_kvcache_specs(None)
-    spec = FullAttentionSpec(
-        block_size=BLOCK_SIZE,
-        num_kv_heads=2,
-        head_size=8,
-        dtype=torch.bfloat16,
-    )
-    return KVCacheConfig(
-        num_blocks=num_blocks,
-        kv_cache_tensors=[KVCacheTensor(size=spec.page_size_bytes * num_blocks, shared_by=["layer0"])],
-        kv_cache_groups=[KVCacheGroupSpec(layer_names=["layer0"], kv_cache_spec=spec)],
-    )
-
-
-def _request(public_id: str, sequence_id: int, *, seq_len: int = 8, kv_contexts=()) -> DiffusionKVRequest:
-    return DiffusionKVRequest(
-        f"{public_id}/diffusion-kv/{sequence_id}",
-        sequence_id=sequence_id,
-        prefix_len=4,
-        target_len=4,
-        seq_len=seq_len,
-        kv_contexts=kv_contexts,
-    )
-
-
-def _manager(num_blocks: int, *, max_model_len: int = 64) -> DiffusionKVCacheManager:
-    return DiffusionKVCacheManager(
-        _config(num_blocks),
-        max_model_len=max_model_len,
-        scheduler_block_size=BLOCK_SIZE,
-        hash_block_size=BLOCK_SIZE,
-    )
-
-
-def test_reserve_and_free_multi_cfg_request() -> None:
-    manager = _manager(8)
-    free_before = manager.native_manager.block_pool.get_num_free_blocks()
-    requests = (_request("public", 0), _request("public", 1))
-
-    metadata = manager.reserve_request("public", requests)
-
-    assert metadata is not None
-    assert metadata.request_id == "public"
-    assert metadata.allocation_generation == 1
-    assert [sequence.sequence_id for sequence in metadata.sequences] == [0, 1]
-    assert [sequence.prefix_len for sequence in metadata.sequences] == [4, 4]
-    assert [sequence.target_len for sequence in metadata.sequences] == [4, 4]
-    assert [sequence.seq_len for sequence in metadata.sequences] == [8, 8]
-    assert [len(sequence.block_ids[0]) for sequence in metadata.sequences] == [2, 2]
-    assert all(sequence.context_ids == () for sequence in metadata.sequences)
-    assert metadata.contexts == ()
-    assert metadata == manager.get_metadata("public")
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before - 4
-
-    manager.free_request("public")
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before
-
-
-def test_impossible_cfg_allocation_rolls_back_and_fails_fast() -> None:
-    manager = _manager(3)
-    free_before = manager.native_manager.block_pool.get_num_free_blocks()
-
-    with pytest.raises(DiffusionKVAdmissionError, match="cannot fit even when the block pool is empty"):
-        manager.reserve_request(
-            "public",
-            (_request("public", 0), _request("public", 1)),
-        )
-
-    assert manager.has_request("public") is False
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before
-
-
-def test_impossible_cfg_allocation_fails_fast_while_pool_is_busy() -> None:
-    manager = _manager(5)
-    assert manager.reserve_request("running", (_request("running", 0),)) is not None
-    free_before = manager.native_manager.block_pool.get_num_free_blocks()
-
-    with pytest.raises(DiffusionKVAdmissionError, match="required_blocks=6, available_blocks=4"):
-        manager.reserve_request(
-            "impossible",
-            tuple(_request("impossible", sequence_id) for sequence_id in range(3)),
-        )
-
-    assert manager.has_request("impossible") is False
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before
-
-
-def test_temporary_capacity_pressure_returns_none() -> None:
-    manager = _manager(3)
-    assert manager.reserve_request("running", (_request("running", 0),)) is not None
-
-    allocation = manager.reserve_request("waiting", (_request("waiting", 0),))
-
-    assert allocation is None
-    assert manager.has_request("waiting") is False
-
-
-def test_partial_cfg_rollback_can_retry_after_capacity_is_released() -> None:
-    manager = _manager(5)
-    assert manager.reserve_request("running", (_request("running", 0),)) is not None
-    waiting = (_request("waiting", 0), _request("waiting", 1))
-
-    assert manager.reserve_request("waiting", waiting) is None
-    assert manager.has_request("waiting") is False
-
-    manager.free_request("running")
-    metadata = manager.reserve_request("waiting", waiting)
-
-    assert metadata is not None
-    assert [len(sequence.block_ids[0]) for sequence in metadata.sequences] == [2, 2]
-
-
-def test_reallocated_request_gets_a_new_allocation_generation() -> None:
-    manager = _manager(3)
-    request = (_request("public", 0),)
-
-    first = manager.reserve_request("public", request)
-    assert first is not None
-    manager.free_request("public")
-    second = manager.reserve_request("public", request)
-
-    assert second is not None
-    assert second.allocation_generation == first.allocation_generation + 1
-
-
-def test_cfg_allocation_exception_rolls_back(monkeypatch) -> None:
-    manager = _manager(8)
-    free_before = manager.native_manager.block_pool.get_num_free_blocks()
-    native_allocate = manager.native_manager.allocate_slots
-    num_calls = 0
-
-    def fail_second_allocation(*args, **kwargs):
-        nonlocal num_calls
-        num_calls += 1
-        if num_calls == 2:
-            raise RuntimeError("injected allocation failure")
-        return native_allocate(*args, **kwargs)
-
-    monkeypatch.setattr(manager.native_manager, "allocate_slots", fail_second_allocation)
-
-    with pytest.raises(RuntimeError, match="injected allocation failure"):
-        manager.reserve_request(
-            "public",
-            (_request("public", 0), _request("public", 1)),
-        )
-
-    assert manager.has_request("public") is False
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before
-
-
-def test_rejects_independent_context_until_role_routing_is_implemented() -> None:
-    manager = _manager(8)
-    context = DiffusionKVContext(context_id="text", cache_role="cross.text", num_tokens=8)
-
-    with pytest.raises(DiffusionKVAdmissionError, match="DiffusionKVContext"):
-        manager.reserve_request(
-            "public",
-            (_request("public", 0, kv_contexts=(context,)),),
-        )
-
-
-def test_rejects_sequence_longer_than_admission_bound() -> None:
-    manager = _manager(8, max_model_len=8)
-
-    with pytest.raises(DiffusionKVAdmissionError, match="exceeds max_model_len"):
-        manager.reserve_request("public", (_request("public", 0, seq_len=9),))
-
-
-@pytest.mark.parametrize(
-    "requests",
-    [
-        (_request("public", 1),),
-        (_request("public", 1), _request("public", 0)),
-        (_request("public", 0), _request("public", 2)),
-    ],
-)
-def test_rejects_ambiguous_sequence_order(requests) -> None:
-    manager = _manager(8)
-
-    with pytest.raises(ValueError, match="ordered by contiguous sequence_id"):
-        manager.reserve_request("public", requests)
-
-
-def test_close_releases_every_public_request() -> None:
-    manager = _manager(8)
-    free_before = manager.native_manager.block_pool.get_num_free_blocks()
-    assert manager.reserve_request("first", (_request("first", 0),)) is not None
-    assert manager.reserve_request("second", (_request("second", 0),)) is not None
-
-    manager.close()
-
-    assert manager.native_manager.block_pool.get_num_free_blocks() == free_before
-    assert manager.has_request("first") is False
-    assert manager.has_request("second") is False
@@ -107,8 +107,11 @@ def test_request_rejects_non_context_values() -> None:
 def _manager(*, enable_caching: bool) -> KVCacheManager:
    # vLLM 0.26 lazily registers its built-in KV specs. Another test may have
    # already registered an out-of-tree spec, making the registry non-empty
-    # before FullAttentionSpec is registered. Initialize the native mappings.
-    native_kv_managers.register_all_kvcache_specs(None)
+    # before FullAttentionSpec is registered. Initialize the native mappings
+    # explicitly; vLLM 0.25 has no registration hook and needs no setup.
+    register_all = getattr(native_kv_managers, "register_all_kvcache_specs", None)
+    if register_all is not None:
+        register_all(None)

    spec = FullAttentionSpec(
        block_size=BLOCK_SIZE,
@@ -1,259 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
-from contextlib import nullcontext
-from types import SimpleNamespace
-from unittest.mock import Mock
-
-import pytest
-import torch
-from torch import nn
-from vllm.v1.kv_cache_interface import FullAttentionSpec, KVCacheConfig, KVCacheGroupSpec, KVCacheTensor
-
-import vllm_omni.diffusion.worker.diffusion_worker as diffusion_worker_module
-from vllm_omni.diffusion.attention.layer import Attention
-from vllm_omni.diffusion.diffusion_kv.config import DiffusionKVCacheMode
-from vllm_omni.diffusion.worker.diffusion_model_runner import DiffusionModelRunner
-from vllm_omni.diffusion.worker.diffusion_worker import DiffusionWorker
-
-pytestmark = [pytest.mark.core_model, pytest.mark.cpu, pytest.mark.diffusion]
-
-
-class _Backend:
-    @classmethod
-    def indexes_kv_by_block_stride(cls) -> bool:
-        return True
-
-
-def _attention(*, enabled: bool) -> Attention:
-    attention = Attention.__new__(Attention)
-    nn.Module.__init__(attention)
-    attention.paged_kv_cache_role = "primary" if enabled else None
-    attention.paged_kv_cache_dtype = torch.bfloat16
-    attention.num_kv_heads = 2
-    attention.head_size = 8
-    attention.causal = False
-    attention.attn_backend = _Backend
-    return attention
-
-
-def _runner(attention: Attention) -> DiffusionModelRunner:
-    runner = object.__new__(DiffusionModelRunner)
-    runner.od_config = SimpleNamespace(diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER)
-    runner.vllm_config = SimpleNamespace(
-        cache_config=SimpleNamespace(block_size=4),
-        model_config=SimpleNamespace(dtype=torch.float16),
-    )
-    pipeline = nn.Module()
-    pipeline.image_attention = attention
-    runner.pipeline = pipeline
-    return runner
-
-
-def test_runner_discovers_native_spec_from_loaded_attention() -> None:
-    specs = _runner(_attention(enabled=True)).get_kv_cache_spec()
-
-    assert set(specs) == {"image_attention"}
-    spec = specs["image_attention"]
-    assert isinstance(spec, FullAttentionSpec)
-    assert spec.block_size == 4
-    assert spec.num_kv_heads == 2
-    assert spec.head_size == 8
-    assert spec.dtype is torch.bfloat16
-    assert spec.indexes_kv_by_block_stride is True
-    assert spec.non_causal is True
-
-
-def test_runner_rejects_paged_mode_without_cache_enabled_attention() -> None:
-    runner = _runner(_attention(enabled=False))
-
-    with pytest.raises(RuntimeError, match="no cache-enabled Attention"):
-        runner.get_kv_cache_spec()
-
-
-def test_runner_retains_matching_rank_local_config() -> None:
-    runner = _runner(_attention(enabled=True))
-    spec = runner.get_kv_cache_spec()["image_attention"]
-    config = KVCacheConfig(
-        num_blocks=8,
-        kv_cache_tensors=[KVCacheTensor(size=spec.page_size_bytes * 8, shared_by=["image_attention"])],
-        kv_cache_groups=[KVCacheGroupSpec(layer_names=["image_attention"], kv_cache_spec=spec)],
-    )
-
-    runner.set_kv_cache_config(config)
-
-    assert runner.kv_cache_config is config
-
-
-def test_runner_rejects_rank_local_config_for_different_layers() -> None:
-    runner = _runner(_attention(enabled=True))
-    spec = runner.get_kv_cache_spec()["image_attention"]
-    config = KVCacheConfig(
-        num_blocks=8,
-        kv_cache_tensors=[KVCacheTensor(size=spec.page_size_bytes * 8, shared_by=["other_attention"])],
-        kv_cache_groups=[KVCacheGroupSpec(layer_names=["other_attention"], kv_cache_spec=spec)],
-    )
-
-    with pytest.raises(ValueError, match="layer mismatch"):
-        runner.set_kv_cache_config(config)
-
-
-def test_worker_selects_its_rank_local_config() -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.rank = 1
-    worker.od_config = SimpleNamespace(num_gpus=2)
-    worker.model_runner = SimpleNamespace(set_kv_cache_config=lambda config: setattr(worker, "installed", config))
-    configs = [object(), object()]
-
-    worker.set_kv_cache_configs(configs)
-
-    assert worker.installed is configs[1]
-    with pytest.raises(ValueError, match="rank count mismatch"):
-        worker.set_kv_cache_configs(configs[:1])
-
-
-def test_worker_honors_explicit_kv_memory_budget(monkeypatch) -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.rank = 0
-    worker.vllm_config = SimpleNamespace(cache_config=SimpleNamespace(kv_cache_memory_bytes=4096))
-    worker.init_snapshot = SimpleNamespace(free_memory=16384)
-    worker.requested_memory = 8192
-    worker.model_runner = SimpleNamespace(profile_run=Mock(), model_memory_usage=1024)
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", lambda value: [value])
-    profile_request = object()
-
-    assert worker.determine_available_kv_memory(profile_request) == [4096]
-    worker.model_runner.profile_run.assert_called_once_with(profile_request)
-
-
-def test_worker_treats_zero_kv_memory_budget_as_profiled_auto_sizing(monkeypatch) -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.rank = 0
-    worker.init_snapshot = SimpleNamespace(free_memory=900)
-    worker.requested_memory = 750
-    worker.vllm_config = SimpleNamespace(
-        cache_config=SimpleNamespace(
-            kv_cache_memory_bytes=0,
-            gpu_memory_utilization=0.75,
-        )
-    )
-    worker.model_runner = SimpleNamespace(profile_run=Mock(), model_memory_usage=100)
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", lambda value: [value])
-    profile_result = SimpleNamespace(
-        non_kv_cache_memory=400,
-        after_profile=SimpleNamespace(free_memory=500),
-    )
-    monkeypatch.setattr(
-        diffusion_worker_module,
-        "memory_profiling",
-        lambda snapshot, weights_memory: nullcontext(profile_result),
-    )
-    profile_request = object()
-
-    assert worker.determine_available_kv_memory(profile_request) == [350]
-    worker.model_runner.profile_run.assert_called_once_with(profile_request)
-
-
-@pytest.mark.parametrize("non_kv_cache_memory", [750, 800])
-def test_worker_rejects_non_positive_profiled_kv_memory(monkeypatch, non_kv_cache_memory: int) -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.rank = 0
-    worker.init_snapshot = SimpleNamespace(free_memory=900)
-    worker.requested_memory = 750
-    worker.vllm_config = SimpleNamespace(
-        cache_config=SimpleNamespace(
-            kv_cache_memory_bytes=None,
-            gpu_memory_utilization=0.75,
-        )
-    )
-    worker.model_runner = SimpleNamespace(profile_run=Mock(), model_memory_usage=100)
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", lambda value: [value])
-    profile_result = SimpleNamespace(
-        non_kv_cache_memory=non_kv_cache_memory,
-        after_profile=SimpleNamespace(free_memory=100),
-    )
-    monkeypatch.setattr(
-        diffusion_worker_module,
-        "memory_profiling",
-        lambda snapshot, weights_memory: nullcontext(profile_result),
-    )
-    profile_request = object()
-
-    with pytest.raises(
-        RuntimeError,
-        match=(
-            "rank 0: RuntimeError: No memory remains for Diffusion KV cache after profiling: "
-            f"requested_memory=750 bytes, non_kv_cache_memory={non_kv_cache_memory} bytes"
-        ),
-    ):
-        worker.determine_available_kv_memory(profile_request)
-
-    worker.model_runner.profile_run.assert_called_once_with(profile_request)
-
-
-def test_worker_profiles_activation_headroom_instead_of_current_residency(monkeypatch) -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.rank = 0
-    worker.init_snapshot = SimpleNamespace(free_memory=1200)
-    worker.requested_memory = 1000
-    worker.vllm_config = SimpleNamespace(
-        cache_config=SimpleNamespace(
-            kv_cache_memory_bytes=None,
-            gpu_memory_utilization=0.8,
-        )
-    )
-    worker.model_runner = SimpleNamespace(profile_run=Mock(), model_memory_usage=250)
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", lambda value: [value])
-    profile_result = SimpleNamespace(
-        non_kv_cache_memory=650,
-        after_profile=SimpleNamespace(free_memory=550),
-    )
-    profiling_calls = []
-
-    def profile(snapshot, *, weights_memory):
-        profiling_calls.append((snapshot, weights_memory))
-        return nullcontext(profile_result)
-
-    monkeypatch.setattr(
-        diffusion_worker_module,
-        "memory_profiling",
-        profile,
-    )
-    profile_request = object()
-
-    assert worker.determine_available_kv_memory(profile_request) == [350]
-    assert profiling_calls == [(worker.init_snapshot, 250)]
-    worker.model_runner.profile_run.assert_called_once_with(profile_request)
-
-
-def test_worker_rejects_profile_without_preload_snapshot(monkeypatch) -> None:
-    worker = object.__new__(DiffusionWorker)
-    worker.init_snapshot = None
-    worker.requested_memory = None
-    worker.vllm_config = SimpleNamespace(cache_config=SimpleNamespace(kv_cache_memory_bytes=None))
-    worker.model_runner = SimpleNamespace()
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", lambda value: [value])
-
-    with pytest.raises(RuntimeError, match="snapshot was not captured before model loading"):
-        worker.determine_available_kv_memory(object())
-
-
-def test_rank_probe_gathers_local_failure_before_raising(monkeypatch) -> None:
-    gathered = []
-
-    def gather(local_result):
-        gathered.append(local_result)
-        return [local_result, (True, "peer-result")]
-
-    def fail_local_probe():
-        raise ValueError("local failure")
-
-    monkeypatch.setattr(diffusion_worker_module, "_all_gather_rank_values", gather)
-
-    with pytest.raises(RuntimeError, match="rank 0: ValueError: local failure"):
-        diffusion_worker_module._run_and_gather_rank_values(
-            "test probe",
-            fail_local_probe,
-        )
-
-    assert gathered == [(False, "ValueError: local failure")]
@@ -85,15 +85,19 @@ def test_new_request_data_carries_scheduler_allocation_atomically() -> None:
    assert new_req.diffusion_kv_metadata is metadata


-def test_paged_request_without_metadata_fails_before_request_forward() -> None:
+def test_paged_request_without_metadata_reaches_request_forward() -> None:
    runner = make_runner(DiffusionKVCacheMode.PAGED_SCHEDULER)
-    runner._execute_request_list = Mock()
+    expected_output = DiffusionOutput(output=None)
+    runner._execute_request_list = Mock(
+        return_value=SimpleNamespace(
+            runner_outputs=[SimpleNamespace(result=expected_output)],
+        )
+    )
    req = SimpleNamespace(request_id="req-0")

-    with pytest.raises(ValueError, match="requires Diffusion KV metadata"):
-        runner.execute_model(req)
+    assert runner.execute_model(req) is expected_output

-    runner._execute_request_list.assert_not_called()
+    runner._execute_request_list.assert_called_once()


 def test_dense_request_rejects_metadata_before_request_forward() -> None:
@@ -137,18 +141,18 @@ def test_request_rpc_rejects_envelope_request_identity_mismatch() -> None:
    assert calls == []


-def test_batch_path_rejects_missing_metadata_before_forward() -> None:
+def test_batch_path_allows_missing_metadata_before_forward() -> None:
    runner = make_runner(DiffusionKVCacheMode.PAGED_SCHEDULER)
-    runner._execute_request_list = Mock()
+    expected_output = object()
+    runner._execute_request_list = Mock(return_value=expected_output)
    new_req = NewRequestData(
        request_id="req-0",
        req=SimpleNamespace(request_id="req-0"),
    )

-    with pytest.raises(ValueError, match="requires Diffusion KV metadata"):
-        runner.execute_model_batch(make_scheduler_output(new_req), runner.od_config)
+    assert runner.execute_model_batch(make_scheduler_output(new_req), runner.od_config) is expected_output

-    runner._execute_request_list.assert_not_called()
+    runner._execute_request_list.assert_called_once()


 def test_batch_path_rejects_envelope_request_identity_mismatch() -> None:
@@ -166,7 +170,7 @@ def test_batch_path_rejects_envelope_request_identity_mismatch() -> None:
    runner._execute_request_list.assert_not_called()


-def test_step_path_rejects_missing_metadata_before_step_support_check() -> None:
+def test_step_path_allows_missing_metadata_before_step_support_check() -> None:
    runner = make_runner(DiffusionKVCacheMode.PAGED_SCHEDULER)
    runner.pipeline = object()
    runner._supports_step_mode = Mock(return_value=False)
@@ -175,10 +179,10 @@ def test_step_path_rejects_missing_metadata_before_step_support_check() -> None:
        req=SimpleNamespace(request_id="req-0"),
    )

-    with pytest.raises(ValueError, match="requires Diffusion KV metadata"):
+    with pytest.raises(ValueError, match="does not support step execution"):
        runner.execute_stepwise(make_scheduler_output(new_req))

-    runner._supports_step_mode.assert_not_called()
+    runner._supports_step_mode.assert_called_once()


 def test_step_path_rejects_envelope_request_identity_mismatch() -> None:
@@ -284,42 +288,3 @@ def test_dense_worker_call_does_not_extend_model_runner_signature() -> None:
    worker.execute_model(req, SimpleNamespace())

    assert calls == [(req, None)]
-
-
-def test_dlo_worker_selects_request_and_metadata_from_same_envelope(monkeypatch) -> None:
-    calls: list[tuple] = []
-
-    class ModelRunner:
-        def execute_model(self, req, **kwargs):
-            calls.append((req, kwargs))
-            return DiffusionOutput(output=None)
-
-    worker = object.__new__(DiffusionWorker)
-    worker.model_runner = ModelRunner()
-    worker.lora_manager = None
-    worker._get_profiler = lambda: None
-    monkeypatch.setattr(
-        "vllm_omni.diffusion.distributed.parallel_state.get_data_parallel_rank",
-        lambda: 1,
-    )
-    req_0 = SimpleNamespace(request_id="req-0")
-    req_1 = SimpleNamespace(request_id="req-1")
-    metadata_0 = make_metadata("req-0")
-    metadata_1 = make_metadata("req-1")
-    envelopes = [
-        NewRequestData(request_id="req-0", req=req_0, diffusion_kv_metadata=metadata_0),
-        NewRequestData(request_id="req-1", req=req_1, diffusion_kv_metadata=metadata_1),
-    ]
-
-    result = worker.execute_model(envelopes, SimpleNamespace())
-
-    assert calls == [
-        (
-            req_1,
-            {
-                "kv_prefetch_job": None,
-                "diffusion_kv_metadata": metadata_1,
-            },
-        )
-    ]
-    assert result["dp_rank"] == 1
@@ -15,7 +15,7 @@ from vllm.config.load import LoadConfig
 from vllm.model_executor.layers.linear import UnquantizedLinearMethod

 from vllm_omni.diffusion.config import get_current_diffusion_config, get_current_diffusion_config_or_none
-from vllm_omni.diffusion.data import DiffusionParallelConfig, OmniDiffusionConfig
+from vllm_omni.diffusion.data import OmniDiffusionConfig
 from vllm_omni.diffusion.model_loader.diffusers_loader import DiffusersPipelineLoader
 from vllm_omni.diffusion.model_loader.host_weight_plan import (
    HostWeightPlan,
@@ -92,17 +92,6 @@ def _make_loader_with_weights(weight_names: list[str]) -> DiffusersPipelineLoade
    return loader


-def _make_dlo_online_quant_config() -> OmniDiffusionConfig:
-    return OmniDiffusionConfig(
-        model="",
-        dtype=torch.float32,
-        quantization_config="fp8",
-        parallel_config=DiffusionParallelConfig(data_parallel_size=2),
-        enable_distributed_layerwise_offload=True,
-        dlo_use_allgather=True,
-    )
-
-
 def test_strict_check_only_validates_source_prefix_parameters():
    model = _DummyPipelineModel(source_prefix="transformer.")
    loader = _make_loader_with_weights(["transformer.weight"])
@@ -310,7 +299,6 @@ def test_dlo_transfers_loader_plan_and_skips_ordinary_weight_loading(monkeypatch

    assert loader.load_model(load_device="cpu") is model
    assert not loaded_ordinary_weights
-    assert plan.requires_backend
    assert loader.take_host_weight_plan() is plan
    assert loader.take_host_weight_plan() is None

@@ -376,9 +364,8 @@ def test_dlo_plan_loads_component_sources_outside_planned_dit(monkeypatch):
    assert loader.take_host_weight_plan() is plan


-def test_dlo_plan_fallback_retains_ordinary_loader_weights_by_default(monkeypatch):
+def test_dlo_plan_fallback_runs_ordinary_loader(monkeypatch):
    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-    import vllm_omni.diffusion.model_loader.host_weight_cache as cache_mod

    od_config = SimpleNamespace(
        dtype=torch.float32,
@@ -403,19 +390,14 @@ def test_dlo_plan_fallback_retains_ordinary_loader_weights_by_default(monkeypatc
        lambda *_args, **_kwargs: HostWeightPlanResult(None, "not direct-compatible"),
    )

-    def build_cache(*_args, **_kwargs):
-        pytest.fail("host weight cache must require explicit selection")
-
-    monkeypatch.setattr(cache_mod, "build_host_weight_cache_plan", build_cache)
-
    assert loader.load_model(load_device="cpu") is model
    assert calls == ["load", "process"]
    assert loader.take_host_weight_plan() is None


-def test_dlo_host_weight_cache_is_built_after_final_post_load_mutation(monkeypatch, tmp_path):
+def test_dlo_runtime_cache_is_built_after_final_post_load_mutation(monkeypatch, tmp_path):
    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-    import vllm_omni.diffusion.model_loader.host_weight_cache as cache_mod
+    import vllm_omni.diffusion.model_loader.runtime_weight_cache as cache_mod

    od_config = SimpleNamespace(
        dtype=torch.float32,
@@ -434,9 +416,9 @@ def test_dlo_host_weight_cache_is_built_after_final_post_load_mutation(monkeypat
        quantization_config=None,
        enable_distributed_layerwise_offload=True,
        dlo_use_allgather=False,
-        dlo_use_host_weight_cache=True,
-        dlo_host_weight_cache_dir=str(tmp_path),
-        dlo_host_weight_cache_lock_timeout=1.0,
+        dlo_enable_runtime_cache=True,
+        dlo_runtime_cache_dir=str(tmp_path),
+        dlo_runtime_cache_lock_timeout=1.0,
        model="unused",
        revision=None,
        model_config={},
@@ -447,9 +429,10 @@ def test_dlo_host_weight_cache_is_built_after_final_post_load_mutation(monkeypat
    model.transformer.quant_method = UnquantizedLinearMethod()
    events: list[str] = []
    runtime_plan = HostWeightPlan(
-        backing_kind="host_weight_cache",
+        backing_kind="runtime_cache",
        bindings={},
        runtime_layout_key="layout-key",
+        post_load_complete=True,
    )

    def load_weights(_model):
@@ -491,17 +474,16 @@ def test_dlo_host_weight_cache_is_built_after_final_post_load_mutation(monkeypat
        "build_checkpoint_mmap_plan",
        lambda *_args, **_kwargs: HostWeightPlanResult(None, "requires normalization"),
    )
-    monkeypatch.setattr(cache_mod, "build_host_weight_cache_plan", build_cache)
+    monkeypatch.setattr(cache_mod, "build_runtime_weight_cache_plan", build_cache)

    assert loader.load_model(load_device="cpu") is model
    assert events == ["load", "process", "calibrate", "eval", "cache"]
    assert loader.take_host_weight_plan() is runtime_plan


-@pytest.mark.parametrize("pin_cpu_memory", [True, False])
-def test_dlo_host_weight_cache_prefers_final_layout_over_checkpoint_plan(monkeypatch, tmp_path, pin_cpu_memory):
+def test_dlo_registration_prefers_final_runtime_cache_over_checkpoint_plan(monkeypatch, tmp_path):
    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-    import vllm_omni.diffusion.model_loader.host_weight_cache as cache_mod
+    import vllm_omni.diffusion.model_loader.runtime_weight_cache as cache_mod

    od_config = SimpleNamespace(
        dtype=torch.float32,
@@ -516,10 +498,9 @@ def test_dlo_host_weight_cache_prefers_final_layout_over_checkpoint_plan(monkeyp
        quantization_config=None,
        enable_distributed_layerwise_offload=True,
        dlo_use_allgather=False,
-        dlo_use_host_weight_cache=True,
-        pin_cpu_memory=pin_cpu_memory,
-        dlo_host_weight_cache_dir=str(tmp_path),
-        dlo_host_weight_cache_pin_limit_gib=0.0,
+        dlo_enable_runtime_cache=True,
+        dlo_runtime_cache_dir=str(tmp_path),
+        dlo_runtime_cache_pin_limit_gib=1.0,
        model="unused",
    )
    loader = DiffusersPipelineLoader(LoadConfig(), od_config)
@@ -527,9 +508,10 @@ def test_dlo_host_weight_cache_prefers_final_layout_over_checkpoint_plan(monkeyp
    model.transformer = nn.Linear(2, 2, bias=False)
    events: list[str] = []
    runtime_plan = HostWeightPlan(
-        backing_kind="host_weight_cache",
+        backing_kind="runtime_cache",
        bindings={},
        runtime_layout_key="layout-key",
+        post_load_complete=True,
    )

    loader._init_from_load_format = lambda *_args, **_kwargs: model  # type: ignore[method-assign]
@@ -538,23 +520,23 @@ def test_dlo_host_weight_cache_prefers_final_layout_over_checkpoint_plan(monkeyp
    loader._apply_skip_softmax_calibration = lambda _model: None  # type: ignore[method-assign]

    def unexpected_checkpoint_plan(*_args, **_kwargs):
-        pytest.fail("selected host weight cache mode must not select the raw checkpoint mmap plan")
+        pytest.fail("registered runtime-cache mode must not select the raw checkpoint mmap plan")

    def build_cache(_pipeline, **_kwargs):
        events.append("cache")
        return HostWeightPlanResult(runtime_plan)

    monkeypatch.setattr(loader_mod, "build_checkpoint_mmap_plan", unexpected_checkpoint_plan)
-    monkeypatch.setattr(cache_mod, "build_host_weight_cache_plan", build_cache)
+    monkeypatch.setattr(cache_mod, "build_runtime_weight_cache_plan", build_cache)

    assert loader.load_model(load_device="cpu") is model
    assert events == ["load", "process", "cache"]
    assert loader.take_host_weight_plan() is runtime_plan


-def test_dlo_host_weight_cache_failure_retains_ordinary_weights(monkeypatch, tmp_path):
+def test_dlo_runtime_cache_failure_retains_ordinary_weights(monkeypatch, tmp_path):
    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-    import vllm_omni.diffusion.model_loader.host_weight_cache as cache_mod
+    import vllm_omni.diffusion.model_loader.runtime_weight_cache as cache_mod

    od_config = SimpleNamespace(
        dtype=torch.float32,
@@ -569,8 +551,8 @@ def test_dlo_host_weight_cache_failure_retains_ordinary_weights(monkeypatch, tmp
        quantization_config=None,
        enable_distributed_layerwise_offload=True,
        dlo_use_allgather=False,
-        dlo_use_host_weight_cache=True,
-        dlo_host_weight_cache_dir=str(tmp_path),
+        dlo_enable_runtime_cache=True,
+        dlo_runtime_cache_dir=str(tmp_path),
        model="unused",
    )
    loader = DiffusersPipelineLoader(LoadConfig(), od_config)
@@ -591,7 +573,7 @@ def test_dlo_host_weight_cache_failure_retains_ordinary_weights(monkeypatch, tmp
        assert kwargs["quantization_config"] == "module_quant_method"
        return HostWeightPlanResult(None, "writer timed out", "lock_timeout")

-    monkeypatch.setattr(cache_mod, "build_host_weight_cache_plan", build_cache)
+    monkeypatch.setattr(cache_mod, "build_runtime_weight_cache_plan", build_cache)

    original_weight = model.transformer.weight.detach().clone()
    assert loader.load_model(load_device="cpu") is model
@@ -599,78 +581,6 @@ def test_dlo_host_weight_cache_failure_retains_ordinary_weights(monkeypatch, tmp
    assert loader.take_host_weight_plan() is None


-def test_dlo_allgather_online_fp8_uses_ordinary_loader(monkeypatch):
-    from vllm.model_executor.layers.quantization.online.fp8 import (
-        Fp8PerTensorOnlineLinearMethod,
-    )
-
-    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-
-    od_config = _make_dlo_online_quant_config()
-    loader = DiffusersPipelineLoader(LoadConfig(), od_config)
-    model = nn.Module()
-    model.transformer = nn.Linear(2, 2, bias=False)
-    model.transformer.quant_method = object.__new__(Fp8PerTensorOnlineLinearMethod)
-    model.transformer.quant_method.uses_meta_device = True
-    calls: list[object] = []
-    allowlist_models: list[nn.Module] = []
-
-    original_allowlist_check = loader._unsupported_dlo_allgather_online_quant_methods
-
-    def check_allowlist(candidate: nn.Module) -> tuple[str, ...]:
-        allowlist_models.append(candidate)
-        return original_allowlist_check(candidate)
-
-    loader._init_from_load_format = lambda *_args, **_kwargs: model  # type: ignore[method-assign]
-    monkeypatch.setattr(loader, "_unsupported_dlo_allgather_online_quant_methods", check_allowlist)
-    loader._request_offload_after_quant = lambda _model: 1  # type: ignore[method-assign]
-    loader.load_weights = (  # type: ignore[method-assign]
-        lambda _model, *, stream_online_quant_to_cpu=False: calls.append(("load", stream_online_quant_to_cpu))
-    )
-    loader._process_weights_after_loading = lambda *_args: calls.append("process")  # type: ignore[method-assign]
-    loader._apply_skip_softmax_calibration = lambda _model: None  # type: ignore[method-assign]
-    monkeypatch.setattr(
-        loader_mod,
-        "build_checkpoint_mmap_plan",
-        lambda *_args, **_kwargs: HostWeightPlanResult(
-            None,
-            "online quantization requires the ordinary loader",
-        ),
-    )
-
-    assert loader.load_model(load_device="cpu", device=torch.device("cpu")) is model
-    assert allowlist_models == [model]
-    assert calls == [("load", True), "process"]
-    assert loader.take_host_weight_plan() is None
-
-
-def test_dlo_allgather_rejects_unvalidated_online_quant_method(monkeypatch):
-    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
-
-    class UnsupportedOnlineMethod:
-        uses_meta_device = True
-
-    od_config = _make_dlo_online_quant_config()
-    loader = DiffusersPipelineLoader(LoadConfig(), od_config)
-    model = nn.Module()
-    model.transformer = nn.Linear(2, 2, bias=False)
-    model.transformer.quant_method = UnsupportedOnlineMethod()
-
-    loader._init_from_load_format = lambda *_args, **_kwargs: model  # type: ignore[method-assign]
-    loader._apply_skip_softmax_calibration = lambda _model: None  # type: ignore[method-assign]
-    monkeypatch.setattr(
-        loader_mod,
-        "build_checkpoint_mmap_plan",
-        lambda *_args, **_kwargs: HostWeightPlanResult(
-            None,
-            "online quantization requires the ordinary loader",
-        ),
-    )
-
-    with pytest.raises(ValueError, match="per-tensor FP8 linears"):
-        loader.load_model(load_device="cpu")
-
-
 def test_hsdp_processes_quantized_weights_before_sharding(mocker):
    import vllm_omni.diffusion.model_loader.diffusers_loader as loader_mod
    from vllm_omni.diffusion.offloader.module_collector import PipelineModules
@@ -10,15 +10,13 @@ import pytest
 import torch
 from torch import nn

-from vllm_omni.diffusion.model_loader.host_weight_cache import (
-    build_host_weight_cache_plan,
-    default_host_weight_cache_root,
+from vllm_omni.diffusion.model_loader.runtime_weight_cache import (
+    build_runtime_weight_cache_plan,
+    default_runtime_cache_root,
 )

 pytestmark = [pytest.mark.diffusion, pytest.mark.cpu, pytest.mark.core_model]

-_SPAWN_TIMEOUT_SECONDS = 120
-

 class _LoaderMarker:
    pass
@@ -44,7 +42,7 @@ class _Pipeline(nn.Module):
 def test_default_root_is_not_split_by_replica_vllm_cache(monkeypatch):
    monkeypatch.setenv("VLLM_CACHE_ROOT", "/tmp/replica-specific-vllm-cache")

-    assert default_host_weight_cache_root().endswith("/.cache/vllm-omni/dlo-host-weights")
+    assert default_runtime_cache_root().endswith("/.cache/vllm-omni/dlo-runtime-weights")


 def _build(pipeline: nn.Module, cache_root: Path, **overrides):
@@ -69,7 +67,7 @@ def _build(pipeline: nn.Module, cache_root: Path, **overrides):
        "pipeline_parallel_size": 1,
    }
    kwargs.update(overrides)
-    return build_host_weight_cache_plan(pipeline, **kwargs)
+    return build_runtime_weight_cache_plan(pipeline, **kwargs)


 def _multiprocess_builder(cache_root: str, queue) -> None:
@@ -90,7 +88,7 @@ def test_equivalent_process_layouts_reuse_one_entry(tmp_path):
    assert first.plan is not None
    assert second.plan is not None
    assert first.plan.runtime_layout_key == second.plan.runtime_layout_key
-    assert not first.plan.requires_backend
+    assert first.plan.post_load_complete
    assert {binding.file_path for binding in first.plan.bindings.values()} == {
        binding.file_path for binding in second.plan.bindings.values()
    }
@@ -101,32 +99,6 @@ def test_equivalent_process_layouts_reuse_one_entry(tmp_path):
    }


-def test_equivalent_local_model_paths_reuse_one_entry(tmp_path, monkeypatch):
-    model_dir = tmp_path / "model"
-    model_dir.mkdir()
-    model_link = tmp_path / "model-link"
-    model_link.symlink_to(model_dir, target_is_directory=True)
-    cache_root = tmp_path / "cache"
-    monkeypatch.setenv("HOME", str(tmp_path))
-    monkeypatch.chdir(tmp_path)
-
-    identities = (str(model_dir), "~/model", "model", str(model_link))
-    results = [_build(_Pipeline(), cache_root, model_identity=identity) for identity in identities]
-
-    assert all(result.plan is not None for result in results)
-    assert len({result.plan.runtime_layout_key for result in results if result.plan is not None}) == 1
-    assert len(list((cache_root / "v1").iterdir())) == 1
-
-
-def test_unstable_loader_identity_fails_closed(tmp_path):
-    result = _build(_Pipeline(), tmp_path, loader_inputs={"opaque": object()})
-
-    assert result.plan is None
-    assert result.fallback_code == "unstable_identity"
-    assert result.fallback_reason == "host weight cache identity does not support values of type builtins.object"
-    assert list(tmp_path.rglob("*.safetensors")) == []
-
-
 def test_tp_coordinate_and_sp_implementation_guard_split_entries(tmp_path):
    pipeline = _Pipeline()
    tp0 = _build(pipeline, tmp_path, tensor_parallel_size=2, tensor_parallel_rank=0)
@@ -162,14 +134,6 @@ def test_expert_parallel_ownership_fails_closed(tmp_path):
    assert list(tmp_path.rglob("*.safetensors")) == []


-def test_pipeline_parallel_ownership_fails_closed(tmp_path):
-    result = _build(_Pipeline(), tmp_path, pipeline_parallel_size=2)
-
-    assert result.plan is None
-    assert result.fallback_code == "unsupported_pp"
-    assert list(tmp_path.rglob("*.safetensors")) == []
-
-
 def test_corrupt_published_entry_is_rebuilt_under_the_key_lock(tmp_path):
    first = _build(_Pipeline(), tmp_path)
    assert first.plan is not None
@@ -247,11 +211,9 @@ def test_concurrent_processes_publish_one_immutable_entry(tmp_path):
    processes = [context.Process(target=_multiprocess_builder, args=(str(tmp_path), queue)) for _ in range(2)]
    for process in processes:
        process.start()
-    # Spawn workers import the full vLLM stack before entering the target;
-    # loaded CI hosts can spend more than 30 seconds in that startup path.
-    results = [queue.get(timeout=_SPAWN_TIMEOUT_SECONDS) for _ in processes]
+    results = [queue.get(timeout=30) for _ in processes]
    for process in processes:
-        process.join(timeout=_SPAWN_TIMEOUT_SECONDS)
+        process.join(timeout=30)

    assert all(process.exitcode == 0 for process in processes)
    assert all(code is None for code, _, _ in results)
@@ -118,14 +118,6 @@ def _call_mgr(
    )


-def test_cache_manager_registers_attention_without_adding_dense_state() -> None:
-    mgr = _make_cache_mgr()
-
-    assert isinstance(mgr, nn.Module)
-    assert dict(mgr.named_modules())["attn"] is mgr.attn
-    assert mgr.state_dict() == {}
-
-
 # ============================================================
 # Test 1: No AR KV — basic cache → reuse
 # ============================================================
@@ -1,2 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
@@ -1,366 +0,0 @@
-#!/usr/bin/env python
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""π0 LeRobot parity (in-process): the bit-for-bit correctness oracle.
-
-Verifies that vllm-omni's ``Pi0ForActionPrediction`` produces bit-for-bit
-matching action chunks with LeRobot's ``PI0Policy`` when fed:
-  * the same weights (``lerobot/pi0_base``)
-  * the same pre-processed inputs (images, masks, tokens, state)
-  * the same initial noise tensor
-
-π0 is a flow-matching model (Euler-integrated ODE from t=1 → t=0 with a fixed
-``num_steps``), so the output is deterministic once the noise is fixed and
-``torch.allclose`` on the final action chunk is a valid oracle. This is the
-authoritative correctness oracle for the vllm-omni π0 port (max|Δ| < 1e-4).
-
-The OpenPI websocket online-serving e2e lives separately in
-``tests/e2e/online_serving/test_pi0_expansion.py``.
-
-Run in a SEPARATE ``lerobot[pi]`` venv (avoids dep conflict with the vllm-omni
-env), with the vllm-omni ``pi0`` package importable::
-
-    python -m pytest tests/diffusion/models/pi0/test_pi0_parity.py -v -s
-
-Skipped automatically when LeRobot is not installed (e.g. the vllm-omni env).
-The run is CPU/float32 with fixed defaults; the only override is
-``PI0_PARITY_MODEL_PATH`` (a local pi0_base dir in LeRobot format, to skip the
-HF download; defaults to ``lerobot/pi0_base``).
-"""
-
-from __future__ import annotations
-
-import copy
-import importlib.util
-import os
-
-import pytest
-import torch
-
-# local_model: needs real weights + a lerobot venv (transformers 5.3.0), so it
-# runs locally rather than in the ready-CI (see docs/contributing/ci/CI_5levels.md).
-# Additionally gated on lerobot being importable.
-_HAS_LEROBOT = importlib.util.find_spec("lerobot") is not None
-
-pytestmark = [pytest.mark.local_model, pytest.mark.diffusion]
-
-
-# ─── Config (fixed; matches LeRobot defaults for ``lerobot/pi0_base``) ──
-DEVICE = "cpu"
-DTYPE_STR = "float32"
-ATOL = 1e-4
-NUM_STEPS = 10
-BATCH_SIZE = 2
-ACTION_DIM = 32
-STATE_DIM = 32
-ACTION_HORIZON = 50
-MAX_TOKEN_LEN = 48
-
-# The only knob: point at a local pi0_base dir (LeRobot format) to skip the HF
-# download; defaults to the HF repo id.
-MODEL_PATH = os.environ.get("PI0_PARITY_MODEL_PATH", "lerobot/pi0_base")
-
-
-def _resolve_checkpoint_dir() -> str:
-    """Return a local dir containing the pi0_base checkpoint (download if needed)."""
-    if os.path.isdir(MODEL_PATH):
-        return MODEL_PATH
-    from huggingface_hub import snapshot_download
-
-    return snapshot_download(repo_id=MODEL_PATH, repo_type="model")
-
-
-# ─── Dummy dataset stats (identity) ──────────────────────────────────
-def _dummy_dataset_stats() -> dict:
-    return {
-        "observation.state": {
-            "mean": torch.zeros(STATE_DIM),
-            "std": torch.ones(STATE_DIM),
-            "q01": torch.zeros(STATE_DIM),
-            "q99": torch.ones(STATE_DIM),
-        },
-        "action": {
-            "mean": torch.zeros(ACTION_DIM),
-            "std": torch.ones(ACTION_DIM),
-            "q01": torch.zeros(ACTION_DIM),
-            "q99": torch.ones(ACTION_DIM),
-        },
-        "images": {
-            cam: {
-                "mean": torch.zeros(3, 224, 224),
-                "std": torch.ones(3, 224, 224),
-                "q01": torch.zeros(3, 224, 224),
-                "q99": torch.ones(3, 224, 224),
-            }
-            for cam in ("base_0_rgb", "left_wrist_0_rgb", "right_wrist_0_rgb")
-        },
-    }
-
-
-def _create_dummy_batch(batch_size: int = BATCH_SIZE, device: str = DEVICE) -> dict:
-    """Reproducible dummy inputs — identical across both implementations."""
-    g = torch.Generator(device="cpu").manual_seed(0)
-    prompt = "Pick up the red block and place it in the bin"
-    return {
-        "observation.state": torch.randn(batch_size, STATE_DIM, generator=g, dtype=torch.float32).to(device),
-        "action": torch.randn(batch_size, ACTION_HORIZON, ACTION_DIM, generator=g, dtype=torch.float32).to(device),
-        "observation.images.base_0_rgb": torch.rand(batch_size, 3, 224, 224, generator=g, dtype=torch.float32).to(
-            device
-        ),
-        "observation.images.left_wrist_0_rgb": torch.rand(batch_size, 3, 224, 224, generator=g, dtype=torch.float32).to(
-            device
-        ),
-        "observation.images.right_wrist_0_rgb": torch.rand(
-            batch_size, 3, 224, 224, generator=g, dtype=torch.float32
-        ).to(device),
-        "task": [prompt for _ in range(batch_size)],
-    }
-
-
-# ─── LeRobot instantiation ────────────────────────────────────────────
-def _instantiate_lerobot():
-    from lerobot.policies.pi0 import PI0Policy
-    from lerobot.policies.pi0.processor_pi0 import make_pi0_pre_post_processors
-
-    policy = PI0Policy.from_pretrained(MODEL_PATH, strict=True)
-    policy.to(DEVICE)
-    policy.config.device = DEVICE
-    policy.eval()
-
-    pre, post = make_pi0_pre_post_processors(config=policy.config, dataset_stats=_dummy_dataset_stats())
-    return policy, pre, post
-
-
-# ─── vllm-omni instantiation ──────────────────────────────────────────
-def _instantiate_vllm_omni():
-    """Build the vllm-omni π0 model in isolation (no pipeline, no engine)."""
-    from vllm_omni.diffusion.models.pi0 import Pi0Config, Pi0ForActionPrediction
-
-    cfg = Pi0Config(
-        max_action_dim=ACTION_DIM,
-        max_state_dim=STATE_DIM,
-        chunk_size=ACTION_HORIZON,
-        num_inference_steps=NUM_STEPS,
-        dtype=DTYPE_STR,
-    )
-    model = Pi0ForActionPrediction(cfg)
-    model.to(DEVICE).eval()
-    _load_lerobot_weights(model)
-    return model
-
-
-def _load_lerobot_weights(model):
-    """Feed the ``lerobot/pi0_base`` safetensors into the vllm-omni model.
-
-    ``Pi0ForActionPrediction.load_weights`` handles the leading ``model.``
-    prefix and the flat→nested / lm_head→embed_tokens remaps, so we can pass
-    the raw checkpoint dict.
-    """
-    import safetensors.torch
-
-    cache_dir = _resolve_checkpoint_dir()
-    path = os.path.join(cache_dir, "model.safetensors")
-    state = safetensors.torch.load_file(path)
-    model.load_weights(list(state.items()))
-
-
-# ─── Helpers to extract LeRobot's pre-processed inputs ────────────────
-def _extract_lerobot_model_inputs(lerobot_policy, processed_batch):
-    """Mimic what ``PI0Policy.predict_action_chunk`` feeds into
-    ``self.model.sample_actions``. We use these *exact* tensors for vllm-omni
-    so any divergence must come from the core model, not preprocessing.
-    """
-    images, img_masks = lerobot_policy._preprocess_images(processed_batch)
-    from lerobot.utils.constants import (
-        OBS_LANGUAGE_ATTENTION_MASK,
-        OBS_LANGUAGE_TOKENS,
-    )
-
-    lang_tokens = processed_batch[OBS_LANGUAGE_TOKENS]
-    lang_masks = processed_batch[OBS_LANGUAGE_ATTENTION_MASK]
-    state = lerobot_policy.prepare_state(processed_batch)
-    return images, img_masks, lang_tokens, lang_masks, state
-
-
-# ─── Shared fixed-noise sampler ───────────────────────────────────────
-def _make_fixed_noise(batch_size: int, device: str) -> torch.Tensor:
-    g = torch.Generator(device="cpu").manual_seed(42)
-    return torch.randn(batch_size, ACTION_HORIZON, ACTION_DIM, generator=g, dtype=torch.float32).to(device)
-
-
-# ─── Main test ────────────────────────────────────────────────────────
-@pytest.mark.skipif(not _HAS_LEROBOT, reason="lerobot not installed (run in a lerobot venv).")
-def test_pi0_vllm_omni_vs_lerobot():
-    print("\n[parity] Instantiating LeRobot…")
-    lerobot_policy, lerobot_pre, _ = _instantiate_lerobot()
-
-    print("[parity] Instantiating vllm-omni…")
-    omni_model = _instantiate_vllm_omni()
-
-    print("[parity] Preparing shared inputs…")
-    raw_batch = _create_dummy_batch()
-    processed_batch = lerobot_pre(copy.deepcopy(raw_batch))
-    images, img_masks, lang_tokens, lang_masks, state = _extract_lerobot_model_inputs(lerobot_policy, processed_batch)
-    noise = _make_fixed_noise(raw_batch["observation.state"].shape[0], DEVICE)
-
-    print(f"[parity] state.shape={state.shape}  lang_tokens.shape={lang_tokens.shape}")
-    print(f"[parity] images[0].shape={images[0].shape} (num_cams={len(images)})")
-    print(f"[parity] noise.shape={noise.shape}  dtype={noise.dtype}")
-
-    # ── LeRobot forward ──
-    print("[parity] Running LeRobot sample_actions…")
-    with torch.no_grad():
-        lerobot_actions = lerobot_policy.model.sample_actions(
-            images,
-            img_masks,
-            lang_tokens,
-            lang_masks,
-            state,
-            noise=noise,
-            num_steps=NUM_STEPS,
-        )
-    print(
-        f"[parity] LeRobot actions: shape={lerobot_actions.shape} "
-        f"mean={lerobot_actions.mean().item():.6f} std={lerobot_actions.std().item():.6f}"
-    )
-
-    # ── vllm-omni forward ──
-    print("[parity] Running vllm-omni sample_actions…")
-    with torch.no_grad():
-        omni_actions = omni_model.sample_actions(
-            images=images,
-            image_masks=img_masks,
-            lang_tokens=lang_tokens,
-            lang_masks=lang_masks,
-            state=state,
-            noise=noise,
-            num_steps=NUM_STEPS,
-        )
-    print(
-        f"[parity] vllm-omni actions: shape={omni_actions.shape} "
-        f"mean={omni_actions.mean().item():.6f} std={omni_actions.std().item():.6f}"
-    )
-
-    # ── Compare ──
-    diff = (lerobot_actions.float() - omni_actions.float()).abs()
-    max_diff = diff.max().item()
-    mean_diff = diff.mean().item()
-    print(f"[parity] |Δ| max={max_diff:.2e}  mean={mean_diff:.2e}  atol={ATOL:.1e}")
-    close = torch.allclose(lerobot_actions.float(), omni_actions.float(), atol=ATOL)
-    print(f"[parity] torch.allclose(atol={ATOL}): {close}")
-
-    if not close:
-        print("\n[parity] ⚠️  Outputs diverge — running per-stage diagnostics…")
-        _diagnose_divergence(
-            lerobot_policy.model,
-            omni_model,
-            images,
-            img_masks,
-            lang_tokens,
-            lang_masks,
-            state,
-            noise,
-        )
-
-    assert close, (
-        f"vllm-omni vs LeRobot actions differ beyond atol={ATOL}. max_diff={max_diff:.2e}  mean_diff={mean_diff:.2e}"
-    )
-
-
-# ─── Per-stage divergence diagnostics ─────────────────────────────────
-@torch.no_grad()
-def _diagnose_divergence(
-    lerobot_flow_model,
-    omni_model,
-    images,
-    img_masks,
-    lang_tokens,
-    lang_masks,
-    state,
-    noise,
-):
-    """Localize a numerical mismatch to a specific pipeline stage:
-    1. Prefix embeddings (SigLIP / embed_tokens / projector) — image vs lang.
-    2. Prefix KV cache layer 0 — PaliGemma LM attention.
-    3. A single denoise_step velocity at t=1.0 — action expert forward.
-    """
-    from vllm_omni.diffusion.models.pi0.modeling_pi0 import (
-        make_att_2d_masks,
-        prepare_attention_masks_4d,
-    )
-
-    # ── Stage 1: prefix embeddings ──
-    lr_prefix_embs, lr_prefix_pad, lr_prefix_att = lerobot_flow_model.embed_prefix(
-        images, img_masks, lang_tokens, lang_masks
-    )
-    sg_prefix_embs, sg_prefix_pad, sg_prefix_att = omni_model.embed_prefix(images, img_masks, lang_tokens, lang_masks)
-    total_diff = (lr_prefix_embs.float() - sg_prefix_embs.float()).abs().max().item()
-    print(f"[diag] prefix_embs max |Δ| = {total_diff:.2e}   (shape={tuple(sg_prefix_embs.shape)})")
-    print(f"[diag] prefix_pad_masks equal: {torch.equal(lr_prefix_pad, sg_prefix_pad)}")
-    print(f"[diag] prefix_att_masks equal: {torch.equal(lr_prefix_att.bool(), sg_prefix_att.bool())}")
-
-    num_cams = len(images)
-    img_len = 256 * num_cams
-    lang_len = lr_prefix_embs.shape[1] - img_len
-    img_diff = (lr_prefix_embs[:, :img_len].float() - sg_prefix_embs[:, :img_len].float()).abs().max().item()
-    lang_diff = (lr_prefix_embs[:, img_len:].float() - sg_prefix_embs[:, img_len:].float()).abs().max().item()
-    print(f"[diag]   image slice [:{img_len}] max|Δ| = {img_diff:.2e}   (num_cams={num_cams})")
-    print(f"[diag]   lang slice  [{img_len}:] max|Δ| = {lang_diff:.2e}   (len={lang_len})")
-
-    def _stats(name, t):
-        t = t.float()
-        print(
-            f"[diag] {name} prefix_embs: mean={t.mean().item():+.4f} "
-            f"std={t.std().item():.4f} min={t.min().item():+.2f} "
-            f"max={t.max().item():+.2f}"
-        )
-
-    _stats("LeRobot ", lr_prefix_embs)
-    _stats("vllm-omni", sg_prefix_embs)
-
-    # ── Stage 2: prefix KV cache ──
-    prefix_att_2d = make_att_2d_masks(sg_prefix_pad, sg_prefix_att)
-    prefix_pos = torch.cumsum(sg_prefix_pad, dim=1) - 1
-    prefix_att_4d = prepare_attention_masks_4d(prefix_att_2d)
-
-    lerobot_flow_model.paligemma_with_expert.paligemma.language_model.config._attn_implementation = "eager"
-    _, lr_kv = lerobot_flow_model.paligemma_with_expert.forward(
-        attention_mask=prefix_att_4d,
-        position_ids=prefix_pos,
-        past_key_values=None,
-        inputs_embeds=[lr_prefix_embs, None],
-        use_cache=True,
-    )
-    _, sg_kv = omni_model.paligemma_with_expert.forward(
-        attention_mask=prefix_att_4d,
-        position_ids=prefix_pos,
-        past_key_values=None,
-        inputs_embeds=[sg_prefix_embs, None],
-        use_cache=True,
-    )
-
-    def _layer0_kv(cache):
-        if isinstance(cache, list):
-            return cache[0]
-        layer0 = cache.layers[0]
-        return layer0.keys, layer0.values
-
-    try:
-        lr_k, lr_v = _layer0_kv(lr_kv)
-        sg_k, sg_v = _layer0_kv(sg_kv)
-        dk = (lr_k.float() - sg_k.float()).abs().max().item()
-        dv = (lr_v.float() - sg_v.float()).abs().max().item()
-        print(f"[diag] prefix KV layer0  K max|Δ|={dk:.2e}  V max|Δ|={dv:.2e}")
-    except Exception as e:  # noqa: BLE001
-        print(f"[diag] could not extract prefix KV for comparison: {e}")
-
-    # ── Stage 3: a single denoise_step at t=1.0 ──
-    bsize = state.shape[0]
-    t = torch.ones(bsize, dtype=torch.float32, device=state.device)
-    lr_vt = lerobot_flow_model.denoise_step(state, sg_prefix_pad, lr_kv, noise, t)
-    sg_vt = omni_model.denoise_step(state, sg_prefix_pad, sg_kv, noise, t)
-    print(f"[diag] denoise_step(t=1) v_t max|Δ| = {(lr_vt.float() - sg_vt.float()).abs().max().item():.2e}")
-
-
-if __name__ == "__main__":
-    test_pi0_vllm_omni_vs_lerobot()
@@ -1,585 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""CPU unit tests for the π0 (Pi-Zero) VLA model.
-
-Deliberately lightweight: pure utility code (config plumbing, mask building,
-sinusoidal embeddings, normalization, image preprocessing, weight-load remap)
-that runs on CPU with no weights and no lerobot. The full LeRobot parity check
-lives in ``tests/diffusion/models/pi0/test_pi0_parity.py`` and requires a model download.
-
-    pytest tests/diffusion/models/pi0/test_pi0_units.py -v
-"""
-
-from __future__ import annotations
-
-import logging
-import math
-
-import pytest
-import torch
-
-from vllm_omni.diffusion.models.pi0.config import Pi0Config
-from vllm_omni.diffusion.models.pi0.modeling_pi0 import (
-    OPENPI_ATTENTION_MASK_VALUE,
-    Pi0ForActionPrediction,
-    _apply_norm,
-    _build_norm_buffers,
-    create_sinusoidal_pos_embedding,
-    get_gemma_config,
-    make_att_2d_masks,
-    prepare_attention_masks_4d,
-)
-from vllm_omni.diffusion.models.pi0.processor_pi0 import (
-    Pi0ImageProcessor,
-    build_model_inputs,
-    pil_image_to_tensor,
-    resize_with_pad,
-)
-
-pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
-
-
-# ----------------------------------------------------------------------------
-# Pi0Config dataclass + checkpoint resolver
-# ----------------------------------------------------------------------------
-_LEROBOT_CFG = {
-    "type": "pi0",
-    "paligemma_variant": "gemma_2b",
-    "action_expert_variant": "gemma_300m",
-    "chunk_size": 50,
-    "max_action_dim": 32,
-    "max_state_dim": 32,
-    "num_inference_steps": 10,
-    "image_resolution": [224, 224],
-    "tokenizer_max_length": 48,
-    "dtype": "float32",
-    # Training-only keys that must be filtered out of the dataclass.
-    "optimizer_lr": 2.5e-5,
-    "scheduler_warmup_steps": 1000,
-    "input_features": {
-        "observation.images.base_0_rgb": {"type": "VISUAL", "shape": [3, 224, 224]},
-        "observation.images.left_wrist_0_rgb": {"type": "VISUAL", "shape": [3, 224, 224]},
-        "observation.images.right_wrist_0_rgb": {"type": "VISUAL", "shape": [3, 224, 224]},
-        "observation.state": {"type": "STATE", "shape": [32]},
-    },
-}
-
-_EXPECTED_CAMERA_ORDER = [
-    "observation.images.base_0_rgb",
-    "observation.images.left_wrist_0_rgb",
-    "observation.images.right_wrist_0_rgb",
-]
-
-
-def test_config_camera_order_from_input_features():
-    c = Pi0Config.from_model_config(_LEROBOT_CFG)
-    assert c.image_feature_keys == _EXPECTED_CAMERA_ORDER
-
-
-def test_config_filters_training_only_keys():
-    c = Pi0Config.from_model_config(_LEROBOT_CFG)
-    assert not hasattr(c, "optimizer_lr")
-    assert not hasattr(c, "scheduler_warmup_steps")
-
-
-def test_config_coerces_image_resolution_to_tuple():
-    c = Pi0Config.from_model_config(_LEROBOT_CFG)
-    assert c.image_resolution == (224, 224)
-    assert isinstance(c.image_resolution, tuple)
-
-
-def test_config_non_square_resolution_raises():
-    with pytest.raises(ValueError):
-        Pi0Config(image_resolution=(224, 256))
-
-
-@pytest.mark.parametrize("bad", [(224,), (224, 224, 3), 224])
-def test_config_malformed_resolution_raises(bad):
-    with pytest.raises(ValueError):
-        Pi0Config(image_resolution=bad)
-
-
-def test_config_defaults_and_empty():
-    c = Pi0Config.from_model_config(None)
-    assert c.chunk_size == 50
-    assert c.max_action_dim == 32
-    assert c.max_state_dim == 32
-    assert c.num_inference_steps == 10
-    assert c.image_resolution == (224, 224)
-    assert c.tokenizer_max_length == 48
-    assert c.max_cameras == 3
-    assert c.norm_stats is None
-    assert c.image_key_map == {}
-
-
-def test_config_deploy_yaml_override():
-    """Deploy-yaml model_config is authoritative (e.g. force dtype/variant)."""
-    c = Pi0Config.from_model_config({**_LEROBOT_CFG, "dtype": "bfloat16", "num_inference_steps": 4})
-    assert c.dtype == "bfloat16"
-    assert c.num_inference_steps == 4
-
-
-# ----------------------------------------------------------------------------
-# Kernel utilities: sinusoidal timestep embedding
-# ----------------------------------------------------------------------------
-def test_sinusoidal_embedding_shape():
-    emb = create_sinusoidal_pos_embedding(torch.tensor([0.0, 0.5, 1.0]), dimension=64)
-    assert emb.shape == (3, 64)
-
-
-def test_sinusoidal_embedding_values_at_zero():
-    emb = create_sinusoidal_pos_embedding(torch.tensor([0.0]), dimension=4)
-    assert torch.allclose(emb[0, :2].float(), torch.zeros(2), atol=1e-6)
-    assert torch.allclose(emb[0, 2:].float(), torch.ones(2), atol=1e-6)
-
-
-def test_sinusoidal_embedding_odd_dim_raises():
-    with pytest.raises(ValueError):
-        create_sinusoidal_pos_embedding(torch.tensor([1.0]), dimension=3)
-
-
-def test_sinusoidal_embedding_non_1d_raises():
-    with pytest.raises(ValueError):
-        create_sinusoidal_pos_embedding(torch.tensor([[0.0, 0.5]]), dimension=4)
-
-
-def test_sinusoidal_embedding_distinct_timesteps_differ():
-    emb = create_sinusoidal_pos_embedding(torch.tensor([0.1, 0.2, 0.5, 0.9]), dimension=16)
-    for i in range(emb.shape[0]):
-        for j in range(i + 1, emb.shape[0]):
-            assert not torch.allclose(emb[i], emb[j])
-
-
-# ----------------------------------------------------------------------------
-# Kernel utilities: attention mask construction
-# ----------------------------------------------------------------------------
-def test_make_att_2d_masks_bidirectional():
-    result = make_att_2d_masks(torch.ones(1, 4, dtype=torch.bool), torch.zeros(1, 4, dtype=torch.long))
-    assert result.all()
-
-
-def test_make_att_2d_masks_causal():
-    result = make_att_2d_masks(torch.ones(1, 4, dtype=torch.bool), torch.ones(1, 4, dtype=torch.long))
-    expected = torch.tril(torch.ones(4, 4, dtype=torch.bool))
-    assert torch.equal(result[0], expected)
-
-
-def test_make_att_2d_masks_prefix_lm():
-    result = make_att_2d_masks(torch.ones(1, 5, dtype=torch.bool), torch.tensor([[0, 0, 0, 1, 1]], dtype=torch.long))
-    assert result[0, 0, 0]
-    assert not result[0, 0, 3]
-    assert result[0, 3, 0]
-    assert not result[0, 3, 4]
-
-
-def test_make_att_2d_masks_respects_padding():
-    pad = torch.tensor([[1, 1, 1, 0]], dtype=torch.bool)  # last token is padding
-    att = torch.zeros(1, 4, dtype=torch.long)  # full bidirectional
-    result = make_att_2d_masks(pad, att)
-    assert not result[0, :, 3].any(), "Nothing should attend to padded key"
-    assert not result[0, 3, :].any(), "Padded query should attend to nothing"
-
-
-def test_prepare_attention_masks_4d():
-    mask_2d = torch.tensor([[[True, False], [True, True]]])
-    mask_4d = prepare_attention_masks_4d(mask_2d)
-    assert mask_4d.shape == (1, 1, 2, 2)
-    assert mask_4d[0, 0, 0, 0].item() == 0.0
-    assert mask_4d[0, 0, 1, 1].item() == 0.0
-    expected = torch.tensor(OPENPI_ATTENTION_MASK_VALUE, dtype=mask_4d.dtype)
-    assert torch.equal(mask_4d[0, 0, 0, 1], expected)
-
-
-def test_attention_mask_value_finite_and_negative():
-    assert math.isfinite(OPENPI_ATTENTION_MASK_VALUE)
-    assert OPENPI_ATTENTION_MASK_VALUE < -1e30
-
-
-# ----------------------------------------------------------------------------
-# Normalization round-trips
-# ----------------------------------------------------------------------------
-def test_build_norm_buffers_missing():
-    assert _build_norm_buffers(None, "state") is None
-    assert _build_norm_buffers({}, "state") is None
-    assert _build_norm_buffers({"state": None}, "state") is None
-
-
-def test_build_norm_buffers_mean_std():
-    stats = {"state": {"mode": "mean_std", "mean": [1.0, 2.0], "std": [0.5, 0.5]}}
-    out = _build_norm_buffers(stats, "state")
-    assert out["mode"] == "mean_std"
-    assert torch.equal(out["mean"], torch.tensor([1.0, 2.0]))
-    assert torch.equal(out["std"], torch.tensor([0.5, 0.5]))
-
-
-def test_build_norm_buffers_min_max():
-    stats = {"action": {"mode": "min_max", "min": [-1.0, 0.0], "max": [1.0, 2.0]}}
-    out = _build_norm_buffers(stats, "action")
-    assert out["mode"] == "min_max"
-    assert torch.equal(out["min"], torch.tensor([-1.0, 0.0]))
-    assert torch.equal(out["max"], torch.tensor([1.0, 2.0]))
-
-
-def test_apply_norm_mean_std_forward_inverse():
-    stats = _build_norm_buffers({"s": {"mode": "mean_std", "mean": [0.5, -0.2], "std": [2.0, 0.1]}}, "s")
-    x = torch.tensor([[1.0, 0.3]])
-    z = _apply_norm(x, stats, inverse=False)
-    assert torch.allclose(z, torch.tensor([[0.25, 5.0]]), atol=1e-5)
-    x_back = _apply_norm(z, stats, inverse=True)
-    assert torch.allclose(x_back, x, atol=1e-5)
-
-
-def test_apply_norm_min_max_forward_inverse():
-    stats = _build_norm_buffers({"s": {"mode": "min_max", "min": [0.0, -1.0], "max": [1.0, 1.0]}}, "s")
-    x = torch.tensor([[0.25, 0.5]])
-    z = _apply_norm(x, stats, inverse=False)
-    assert torch.allclose(z, torch.tensor([[-0.5, 0.5]]), atol=1e-5)
-    x_back = _apply_norm(z, stats, inverse=True)
-    assert torch.allclose(x_back, x, atol=1e-5)
-
-
-def test_apply_norm_preserves_padded_tail():
-    stats = _build_norm_buffers({"s": {"mode": "mean_std", "mean": [0.0, 0.0], "std": [1.0, 1.0]}}, "s")
-    x = torch.tensor([[1.0, 2.0, 99.0, -99.0]])
-    z = _apply_norm(x, stats, inverse=False)
-    assert torch.allclose(z[:, :2], x[:, :2])
-    assert torch.allclose(z[:, 2:], x[:, 2:])
-
-
-def test_apply_norm_noop_when_stats_none():
-    x = torch.randn(2, 5)
-    assert torch.equal(_apply_norm(x, None, inverse=False), x)
-    assert torch.equal(_apply_norm(x, None, inverse=True), x)
-
-
-# ----------------------------------------------------------------------------
-# Gemma variant configs
-# ----------------------------------------------------------------------------
-def test_gemma_2b_dims():
-    c = get_gemma_config("gemma_2b")
-    assert c.width == 2048
-    assert c.depth == 18
-
-
-def test_gemma_300m_dims():
-    c = get_gemma_config("gemma_300m")
-    assert c.width == 1024
-    assert c.depth == 18
-
-
-def test_gemma_unknown_variant_raises():
-    with pytest.raises(ValueError):
-        get_gemma_config("gemma_7b")
-
-
-# ----------------------------------------------------------------------------
-# load_weights remap rules (tiny model; needs torch + transformers)
-# ----------------------------------------------------------------------------
-def _tiny_pi0_model():
-    """Construct a tiny π0 model for remap tests (full Gemma depth but small
-    action/state dims — the PaliGemma backbone is always full-size)."""
-    cfg = Pi0Config(max_action_dim=8, max_state_dim=8, chunk_size=2)
-    return Pi0ForActionPrediction(cfg)
-
-
-@pytest.fixture(autouse=True)
-def _silence_pi0_loader(caplog):
-    # The remap tests feed a single synthetic weight; the loader's
-    # "missing params" warning is expected noise here.
-    logging.getLogger("vllm_omni.diffusion.models.pi0.modeling_pi0").setLevel(logging.ERROR)
-    yield
-
-
-@pytest.mark.slow
-def test_remap_paligemma_submodules_to_nested_layout():
-    model = _tiny_pi0_model()
-    params = dict(model.named_parameters())
-    target = next(k for k in params if k.startswith("paligemma_with_expert.paligemma.model.vision_tower."))
-    flat_key = "model." + target.replace(
-        "paligemma_with_expert.paligemma.model.vision_tower.",
-        "paligemma_with_expert.paligemma.vision_tower.",
-    )
-    payload = torch.full(params[target].shape, 0.12345)
-    before = params[target].detach().clone()
-    model.load_weights([(flat_key, payload)])
-    after = dict(model.named_parameters())[target]
-    assert not torch.equal(before, after), "flat→nested remap broken"
-    assert torch.allclose(after, payload)
-
-
-@pytest.mark.slow
-def test_lm_head_remap_to_embed_tokens():
-    model = _tiny_pi0_model()
-    params = dict(model.named_parameters())
-    target = "paligemma_with_expert.paligemma.model.language_model.embed_tokens.weight"
-    assert target in params
-    payload = torch.full(params[target].shape, 0.5)
-    before = params[target].detach().clone()
-    model.load_weights([("model.paligemma_with_expert.paligemma.lm_head.weight", payload)])
-    after = dict(model.named_parameters())[target]
-    assert not torch.equal(before, after), "lm_head→embed_tokens remap broken"
-    assert torch.allclose(after, payload)
-
-
-@pytest.mark.slow
-def test_vision_tower_version_robust_remap():
-    """transformers >=5.4 flattens SigLIP from vision_tower.vision_model.* to
-    vision_tower.*; the lerobot/pi0_base checkpoint uses the vision_model infix.
-    load_weights must adapt so vision weights aren't silently dropped (which
-    would leave the vision encoder at random init)."""
-    model = _tiny_pi0_model()
-    params = dict(model.named_parameters())
-    # A representative vision-tower param as it exists on the model.
-    target = next(
-        k
-        for k in params
-        if k.startswith("paligemma_with_expert.paligemma.model.vision_tower.") and "patch_embedding.weight" in k
-    )
-    vt = "paligemma_with_expert.paligemma.model.vision_tower."
-    rest = target[len(vt) :]
-    # Build the checkpoint key in whichever layout differs from the model's.
-    if rest.startswith("vision_model."):
-        ckpt_key = vt + rest[len("vision_model.") :]
-    else:
-        ckpt_key = vt + "vision_model." + rest
-    payload = torch.full(params[target].shape, 0.073)
-    before = params[target].detach().clone()
-    model.load_weights([("model." + ckpt_key, payload)])
-    after = dict(model.named_parameters())[target]
-    assert not torch.equal(before, after), "vision_tower version-robust remap broken"
-    assert torch.allclose(after, payload)
-
-
-# ----------------------------------------------------------------------------
-# Image processor
-# ----------------------------------------------------------------------------
-def test_resize_with_pad_square():
-    out = resize_with_pad(torch.ones(1, 3, 224, 224), 224, 224)
-    assert out.shape == (1, 3, 224, 224)
-
-
-def test_resize_with_pad_wide():
-    """Wide source (H < W): padding applied to top/bottom (rows)."""
-    out = resize_with_pad(torch.zeros(1, 3, 100, 200), 224, 224)
-    assert out.shape == (1, 3, 224, 224)
-    assert out[0, 0, 0, 112].item() == -1.0  # top row is padded
-
-
-def test_resize_with_pad_tall():
-    """Tall source (H > W): padding applied to left/right (cols)."""
-    out = resize_with_pad(torch.zeros(1, 3, 400, 100), 224, 224)
-    assert out.shape == (1, 3, 224, 224)
-    assert out[0, 0, 112, 0].item() == -1.0  # left-most column padded
-
-
-def test_resize_with_pad_rejects_non_4d():
-    with pytest.raises(ValueError):
-        resize_with_pad(torch.zeros(3, 100, 200), 224, 224)
-
-
-def test_pil_to_tensor():
-    import numpy as np
-    from PIL import Image
-
-    img = Image.fromarray(np.full((10, 10, 3), 128, dtype=np.uint8))
-    t = pil_image_to_tensor(img)
-    assert t.shape == (1, 3, 10, 10)
-    expected = 128.0 / 255.0 * 2.0 - 1.0
-    assert abs(t[0, 0, 5, 5].item() - expected) < 1e-4
-
-
-def test_pil_to_tensor_rgba_to_rgb():
-    import numpy as np
-    from PIL import Image
-
-    rgba = np.full((8, 8, 4), 200, dtype=np.uint8)
-    img = Image.fromarray(rgba, mode="RGBA")
-    t = pil_image_to_tensor(img)
-    assert t.shape == (1, 3, 8, 8)
-
-
-def test_make_empty_image():
-    proc = Pi0ImageProcessor(image_size=224)
-    empty = proc.make_empty_image()
-    assert empty.shape == (1, 3, 224, 224)
-    assert (empty == -1.0).all()
-
-
-def test_preprocess_single_resizes_ndarray():
-    import numpy as np
-
-    proc = Pi0ImageProcessor(image_size=224)
-    arr = np.full((120, 120, 3), 100, dtype=np.uint8)
-    t = proc.preprocess_single(arr)
-    assert t.shape == (1, 3, 224, 224)
-
-
-# ----------------------------------------------------------------------------
-# build_model_inputs: camera ordering, padding, masks, state
-# ----------------------------------------------------------------------------
-class _FakeTokenizer:
-    """Minimal stand-in for the PaliGemma tokenizer (no download)."""
-
-    def __call__(
-        self, text, padding=None, max_length=None, truncation=None, add_special_tokens=None, return_tensors=None
-    ):
-        ids = [2] + [ord(c) % 100 for c in text][: max_length - 1]
-        ids = ids[:max_length]
-        attn = [1] * len(ids)
-        while len(ids) < max_length:
-            ids.append(0)
-            attn.append(0)
-        return {"input_ids": ids, "attention_mask": attn}
-
-
-def test_build_model_inputs_camera_order_and_padding():
-    import numpy as np
-
-    config = Pi0Config.from_model_config(_LEROBOT_CFG)  # 3-camera config
-    device = torch.device("cpu")
-    # Provide only 2 of the 3 declared cameras.
-    robot_obs = {
-        "images": {
-            "observation.images.base_0_rgb": np.full((224, 224, 3), 50, dtype=np.uint8),
-            "observation.images.left_wrist_0_rgb": np.full((224, 224, 3), 80, dtype=np.uint8),
-        },
-        "state": [0.1] * 14,  # raw state shorter than max_state_dim=32
-        "prompt": "pick up the block",
-    }
-    images, image_masks, lang_tokens, lang_masks, state = build_model_inputs(
-        robot_obs, config, _FakeTokenizer(), device
-    )
-    assert len(images) == 3
-    assert all(img.shape == (1, 3, 224, 224) for img in images)
-    masks = [bool(m.item()) for m in image_masks]
-    assert masks == [True, True, False]
-    # The padded (3rd) camera must be all -1.
-    assert (images[2] == -1.0).all()
-    assert lang_tokens.shape == (1, 48)
-    assert lang_masks.shape == (1, 48)
-    assert state.shape == (1, 32)
-    # State padded with zeros beyond the 14 real entries.
-    assert torch.allclose(state[0, 14:], torch.zeros(18))
-
-
-def test_build_model_inputs_state_truncated():
-    config = Pi0Config.from_model_config(_LEROBOT_CFG)
-    robot_obs = {"images": {}, "state": [0.5] * 40, "prompt": "x"}
-    _, _, _, _, state = build_model_inputs(robot_obs, config, _FakeTokenizer(), torch.device("cpu"))
-    assert state.shape == (1, 32)
-
-
-def test_build_model_inputs_image_key_map():
-    import numpy as np
-
-    cfg = Pi0Config.from_model_config({**_LEROBOT_CFG, "image_key_map": {"cam_high": "observation.images.base_0_rgb"}})
-    robot_obs = {
-        "images": {"cam_high": np.full((224, 224, 3), 60, dtype=np.uint8)},
-        "prompt": "go",
-    }
-    _, image_masks, _, _, _ = build_model_inputs(robot_obs, cfg, _FakeTokenizer(), torch.device("cpu"))
-    masks = [bool(m.item()) for m in image_masks]
-    assert masks == [True, False, False]  # only base_0_rgb (mapped) is present
-
-
-# ----------------------------------------------------------------------------
-# Version-stability / self-consistency (gated on a real pi0_base checkpoint)
-#
-# The LeRobot parity gate (test_pi0_parity.py) only validates correctness at
-# transformers 5.3.0 (where LeRobot loads pi0_base cleanly). Production vllm-omni
-# ships transformers 5.6.2, and LeRobot is not a usable reference there (it
-# incorrectly loads the SigLIP vision tower at >=5.4). So we pin the kernel's deterministic
-# output to golden values established at 5.3.0 and re-check them here: running on
-# the production transformers proves the port is version-stable and guards the
-# vision_tower / embed_scale version-robust paths against regressions.
-#
-#   PI0_SELFCONSIST_CKPT=/path/to/pi0_base \
-#       pytest tests/diffusion/models/pi0/test_pi0_units.py -k version_stable -v -s
-# ----------------------------------------------------------------------------
-import os  # noqa: E402
-
-_SELFCONSIST_CKPT = os.environ.get("PI0_SELFCONSIST_CKPT")
-
-# Golden values established at transformers 5.3.0 (where parity = 7.15e-07).
-_GOLDEN_SUM = -67.808113
-_GOLDEN_STD = 0.252789
-_GOLDEN_MEAN = -0.021190
-_GOLDEN_EMBED_IMAGE_STD = 2.17603
-
-
-def _selfconsist_inputs(device):
-    # NB: draw order is load-bearing — the golden values were established with
-    # state drawn first, then the 3 images, then lang (shared RNG, seed 0).
-    g = torch.Generator(device="cpu").manual_seed(0)
-    b = 2
-    state = torch.randn(b, 32, generator=g)
-    images = [torch.rand(b, 3, 224, 224, generator=g) for _ in range(3)]
-    masks = [torch.ones(b, dtype=torch.bool) for _ in range(3)]
-    lang = torch.randint(0, 1000, (b, 48), generator=g)
-    lmask = torch.ones(b, 48, dtype=torch.bool)
-    ng = torch.Generator(device="cpu").manual_seed(42)
-    noise = torch.randn(b, 50, 32, generator=ng)
-    return (
-        [im.to(device) for im in images],
-        [m.to(device) for m in masks],
-        lang.to(device),
-        lmask.to(device),
-        state.to(device),
-        noise.to(device),
-    )
-
-
-def _selfconsist_model(device):
-    import safetensors.torch
-
-    model = Pi0ForActionPrediction(
-        Pi0Config(max_action_dim=32, max_state_dim=32, chunk_size=50, num_inference_steps=10, dtype="float32")
-    )
-    state = safetensors.torch.load_file(os.path.join(_SELFCONSIST_CKPT, "model.safetensors"))
-    filled = model.load_weights(list(state.items()))
-    expected = {n for n, _ in model.named_parameters() if "rotary_emb" not in n and not n.endswith(".inv_freq")}
-    missing = expected - (filled or set())
-    assert not missing, f"{len(missing)} params received no weight (e.g. {sorted(missing)[:3]})"
-    return model.to(device).eval()
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(not _SELFCONSIST_CKPT, reason="Set PI0_SELFCONSIST_CKPT to a pi0_base dir.")
-@torch.no_grad()
-def test_version_stable_actions():
-    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-    model = _selfconsist_model(device)
-    images, masks, lang, lmask, state, noise = _selfconsist_inputs(device)
-    actions = (
-        model.sample_actions(
-            images=images,
-            image_masks=masks,
-            lang_tokens=lang,
-            lang_masks=lmask,
-            state=state,
-            noise=noise,
-            num_steps=10,
-        )
-        .float()
-        .cpu()
-    )
-    s, m, std = actions.sum().item(), actions.mean().item(), actions.std().item()
-    assert abs(s - _GOLDEN_SUM) < 0.5, f"action sum drifted: {s} vs {_GOLDEN_SUM}"
-    assert abs(std - _GOLDEN_STD) < 0.01, f"action std drifted: {std} vs {_GOLDEN_STD}"
-    assert abs(m - _GOLDEN_MEAN) < 0.01, f"action mean drifted: {m} vs {_GOLDEN_MEAN}"
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(not _SELFCONSIST_CKPT, reason="Set PI0_SELFCONSIST_CKPT to a pi0_base dir.")
-@torch.no_grad()
-def test_version_stable_embed_image():
-    """Directly guards the SigLIP vision-tower load across the >=5.4 flattening."""
-    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-    model = _selfconsist_model(device)
-    g = torch.Generator(device="cpu").manual_seed(7)
-    img = torch.rand(1, 3, 224, 224, generator=g, dtype=torch.float32).to(device)
-    std = model.paligemma_with_expert.embed_image(img).float().std().item()
-    assert abs(std - _GOLDEN_EMBED_IMAGE_STD) < 0.05, (
-        f"embed_image std drifted: {std} vs {_GOLDEN_EMBED_IMAGE_STD} — "
-        "likely a vision_tower weight-load regression under this transformers version."
-    )
@@ -1,72 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""Regression test: SenseNovaU1ForCausalLM instantiation under the diffusion
-config shim must not crash on missing ``head_dtype`` and the resulting
-LogitsProcessor must have ``head_dtype is None`` (= use model dtype)."""
-
-import os
-
-import pytest
-import torch
-from vllm.config import DeviceConfig, VllmConfig, set_current_vllm_config
-from vllm.distributed.parallel_state import (
-    cleanup_dist_env_and_memory,
-    init_distributed_environment,
-    initialize_model_parallel,
-)
-
-from vllm_omni.diffusion.models.sensenova_u1.sensenova_u1_transformer import (
-    SenseNovaU1ForCausalLM,
-)
-from vllm_omni.diffusion.vllm_config import _DiffusionVllmModelConfig
-from vllm_omni.transformers_utils.configs.sensenova_u1 import (
-    SenseNovaU1Config,
-)
-
-pytestmark = [pytest.mark.core_model, pytest.mark.diffusion, pytest.mark.cpu]
-
-
-def _build_tiny_sensenova_u1(quant_config=None, prefix="model"):
-    """Build a minimal SenseNovaU1ForCausalLM for unit tests."""
-    config = SenseNovaU1Config(
-        llm_config={
-            "hidden_size": 64,
-            "num_attention_heads": 2,
-            "num_hidden_layers": 1,
-            "intermediate_size": 128,
-            "vocab_size": 32,
-            "max_position_embeddings": 128,
-            "max_position_embeddings_hw": 128,
-        },
-    )
-    return SenseNovaU1ForCausalLM(config.llm_config, quant_config=quant_config, prefix=prefix)
-
-
-def test_logits_processor_head_dtype_under_diffusion_shim():
-    # Initialize the distributed environment
-    os.environ.setdefault("MASTER_ADDR", "localhost")
-    os.environ.setdefault("MASTER_PORT", "29543")
-    init_distributed_environment(world_size=1, rank=0, local_rank=0, distributed_init_method="env://")
-    initialize_model_parallel()
-
-    # Create a vLLM config with a wrapped diffusion config
-    fake_diff_config = _DiffusionVllmModelConfig(
-        model="sensenova-u1-test",
-        dtype=torch.bfloat16,
-        max_model_len=8192,
-        original_max_model_len=8192,
-    )
-    vllm_config = VllmConfig(
-        device_config=DeviceConfig(device="cpu"),
-    )
-    vllm_config.modelconfig = fake_diff_config  # type: ignore[assignment]
-
-    # Build a tiny version of the Causal LM component
-    with set_current_vllm_config(vllm_config):
-        model = _build_tiny_sensenova_u1()
-
-    # Ensure that we have a set head_dtype attribute. Currently, the head_dtype
-    # is set to None
-    head_dtype = model.logits_processor.head_dtype
-    assert head_dtype is None
-    cleanup_dist_env_and_memory()
@@ -3,11 +3,9 @@

 from __future__ import annotations

-import ctypes
 from types import SimpleNamespace

 import pytest
-import torch

 import vllm_omni.diffusion.offloader.cuda_host_registration as registration_module
 from vllm_omni.diffusion.offloader.cuda_host_registration import (
@@ -17,10 +15,6 @@ from vllm_omni.diffusion.offloader.cuda_host_registration import (
    CudaHostRegistrationError,
    _coalesce_ranges,
 )
-from vllm_omni.diffusion.offloader.host_registration import (
-    HostRegistrationError,
-    register_host_mappings,
-)

 pytestmark = [pytest.mark.diffusion, pytest.mark.cpu, pytest.mark.core_model]

@@ -59,23 +53,12 @@ class _FakeRuntime:
        self,
        register_results: list[int | Exception],
        unregister_results: list[int | Exception] | None = None,
-        *,
-        read_only_registration_supported: bool = True,
-        attribute_result: int = 0,
    ) -> None:
        self._register_results = iter(register_results)
        self._unregister_results = iter(unregister_results or [])
-        self._read_only_registration_supported = read_only_registration_supported
-        self._attribute_result = attribute_result
-        self.attribute_queries: list[tuple[int, int]] = []
        self.registered: list[tuple[int, int, int]] = []
        self.unregistered: list[int] = []

-    def cudaDeviceGetAttribute(self, value, attribute: int, device: int) -> int:
-        self.attribute_queries.append((attribute, device))
-        ctypes.cast(value, ctypes.POINTER(ctypes.c_int)).contents.value = int(self._read_only_registration_supported)
-        return self._attribute_result
-
    def cudaHostRegister(self, pointer: int, size: int, flags: int) -> int:
        self.registered.append((pointer, size, flags))
        result = next(self._register_results)
@@ -95,44 +78,6 @@ class _FakeRuntime:
        return f"error-{error}"


-@pytest.fixture(autouse=True)
-def _current_cuda_device(monkeypatch: pytest.MonkeyPatch) -> None:
-    monkeypatch.setattr(registration_module.torch.accelerator, "current_device_index", lambda: 0)
-
-
-def test_platform_factory_reports_unsupported_registration() -> None:
-    with pytest.raises(HostRegistrationError, match="not supported on cpu"):
-        register_host_mappings({}, device=torch.device("cpu"), max_bytes=4096)
-
-
-def test_platform_factory_dispatches_cuda(monkeypatch: pytest.MonkeyPatch) -> None:
-    sentinel = object()
-    sources = {"weights": []}
-
-    def create(actual_sources, *, max_bytes):
-        assert actual_sources is sources
-        assert max_bytes == 4096
-        return sentinel
-
-    monkeypatch.setattr(CudaHostRegistration, "create", staticmethod(create))
-
-    assert register_host_mappings(sources, device=torch.device("cuda"), max_bytes=4096) is sentinel
-
-
-def test_platform_factory_allows_registration_without_additional_budget(monkeypatch: pytest.MonkeyPatch) -> None:
-    sentinel = object()
-    sources = {"weights": []}
-
-    def create(actual_sources, *, max_bytes):
-        assert actual_sources is sources
-        assert max_bytes is None
-        return sentinel
-
-    monkeypatch.setattr(CudaHostRegistration, "create", staticmethod(create))
-
-    assert register_host_mappings(sources, device=torch.device("cuda"), max_bytes=None) is sentinel
-
-
 def test_coalesce_ranges_aligns_and_merges_overlapping_pages() -> None:
    assert _coalesce_ranges(
        [(0x1003, 4096), (0x2800, 1024), (0x9001, 1)],
@@ -157,61 +102,10 @@ def test_registration_rejects_over_budget_before_calling_cuda(monkeypatch: pytes
    assert runtime.registered == []


-def test_registration_without_additional_budget_registers_complete_mapping(monkeypatch: pytest.MonkeyPatch) -> None:
-    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
-    runtime = _FakeRuntime([0])
-    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-
-    registration = CudaHostRegistration.create(
-        {"weights": [_FakeTensor(0x1003, 4096)]},  # type: ignore[list-item]
-        max_bytes=None,
-    )
-
-    assert registration.total_bytes == 8192
-    assert len(runtime.registered) == 1
-    assert registration.close() == []
-
-
-def test_registration_rejects_unsupported_read_only_mapping(monkeypatch: pytest.MonkeyPatch) -> None:
-    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
-    runtime = _FakeRuntime([0], read_only_registration_supported=False)
-    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-
-    with pytest.raises(CudaHostRegistrationError, match="does not support read-only host registration"):
-        CudaHostRegistration.create(
-            {"weights": [_FakeTensor(0x1003, 1)]},  # type: ignore[list-item]
-            max_bytes=4096,
-        )
-
-    assert runtime.attribute_queries == [
-        (registration_module._CUDA_DEVICE_ATTRIBUTE_HOST_REGISTER_READ_ONLY_SUPPORTED, 0)
-    ]
-    assert runtime.registered == []
-
-
-def test_registration_consumes_capability_query_error(monkeypatch: pytest.MonkeyPatch) -> None:
-    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
-    runtime = _FakeRuntime([0], attribute_result=7)
-    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-    consumed_errors: list[int] = []
-    monkeypatch.setattr(registration_module, "_consume_last_cuda_error", consumed_errors.append)
-
-    with pytest.raises(CudaHostRegistrationError, match="error-7"):
-        CudaHostRegistration.create(
-            {"weights": [_FakeTensor(0x1003, 1)]},  # type: ignore[list-item]
-            max_bytes=4096,
-        )
-
-    assert consumed_errors == [7]
-    assert runtime.registered == []
-
-
 def test_registration_rolls_back_partial_success(monkeypatch: pytest.MonkeyPatch) -> None:
    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
    runtime = _FakeRuntime([0, 7])
    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-    consumed_errors: list[int] = []
-    monkeypatch.setattr(registration_module, "_consume_last_cuda_error", consumed_errors.append)

    with pytest.raises(CudaHostRegistrationError, match="error-7"):
        CudaHostRegistration.create(
@@ -223,7 +117,6 @@ def test_registration_rolls_back_partial_success(monkeypatch: pytest.MonkeyPatch
        )

    assert runtime.unregistered == [0x1000]
-    assert consumed_errors == [7]


 def test_registration_wraps_runtime_exception_and_rolls_back(monkeypatch: pytest.MonkeyPatch) -> None:
@@ -247,8 +140,6 @@ def test_registration_fails_closed_when_rollback_fails(monkeypatch: pytest.Monke
    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
    runtime = _FakeRuntime([0, 7], unregister_results=[9])
    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-    consumed_errors: list[int] = []
-    monkeypatch.setattr(registration_module, "_consume_last_cuda_error", consumed_errors.append)

    with pytest.raises(CudaHostRegistrationCleanupError, match="rollback errors"):
        CudaHostRegistration.create(
@@ -259,35 +150,6 @@ def test_registration_fails_closed_when_rollback_fails(monkeypatch: pytest.Monke
            max_bytes=8192,
        )

-    assert consumed_errors == [7, 9]
-
-
-@pytest.mark.parametrize("pending_error", [0, 801])
-def test_consume_last_cuda_error_accepts_cleared_or_matching_state(
-    monkeypatch: pytest.MonkeyPatch,
-    pending_error: int,
-) -> None:
-    class _GetLastError:
-        def __call__(self) -> int:
-            return pending_error
-
-    runtime = SimpleNamespace(cudaGetLastError=_GetLastError())
-    monkeypatch.setattr(registration_module.ctypes, "CDLL", lambda _name: runtime)
-
-    registration_module._consume_last_cuda_error(801)
-
-
-def test_consume_last_cuda_error_rejects_unrelated_pending_error(monkeypatch: pytest.MonkeyPatch) -> None:
-    class _GetLastError:
-        def __call__(self) -> int:
-            return 700
-
-    runtime = SimpleNamespace(cudaGetLastError=_GetLastError())
-    monkeypatch.setattr(registration_module.ctypes, "CDLL", lambda _name: runtime)
-
-    with pytest.raises(CudaHostRegistrationCleanupError, match="cudaGetLastError reported 700"):
-        registration_module._consume_last_cuda_error(801)
-

 def test_registration_closes_all_coalesced_ranges(monkeypatch: pytest.MonkeyPatch) -> None:
    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
@@ -333,8 +195,6 @@ def test_registration_close_retries_failed_ranges(monkeypatch: pytest.MonkeyPatc
    monkeypatch.setattr(registration_module.torch.cuda, "is_available", lambda: True)
    runtime = _FakeRuntime([0], unregister_results=[9, 0])
    monkeypatch.setattr(registration_module.torch.cuda, "cudart", lambda: runtime)
-    consumed_errors: list[int] = []
-    monkeypatch.setattr(registration_module, "_consume_last_cuda_error", consumed_errors.append)
    registration = CudaHostRegistration.create(
        {"weights": [_FakeTensor(0x1003, 1)]},  # type: ignore[list-item]
        max_bytes=4096,
@@ -343,4 +203,3 @@ def test_registration_close_retries_failed_ranges(monkeypatch: pytest.MonkeyPatc
    assert registration.close() == ["cudaHostUnregister(0x1000) failed: error-9"]
    assert registration.close() == []
    assert runtime.unregistered == [0x1000, 0x1000]
-    assert consumed_errors == [9]
@@ -18,7 +18,6 @@ from torch import nn
 from torch.distributed.tensor import DeviceMesh, DTensor, Replicate

 import vllm_omni.diffusion.offloader.distributed_layerwise_backend as dist_backend_module
-from vllm_omni.diffusion.model_loader.host_weight_cache_utils import file_digest
 from vllm_omni.diffusion.model_loader.host_weight_plan import (
    HostWeightPlan,
    TensorBinding,
@@ -186,79 +185,6 @@ class TestDistributedLayerwiseOffloadHook:
        assert next_block.weight.stride() == expected_stride
        assert torch.equal(next_block.weight, expected)

-    def test_allgather_reconstructs_online_fp8_weight_and_scale(
-        self,
-        patched_offload_runtime,
-        monkeypatch,
-    ):
-        """DLO must reconstruct finalized online-FP8 tensors byte-for-byte."""
-
-        class OnlineFp8Block(nn.Module):
-            def __init__(self):
-                super().__init__()
-                logical_weight = torch.tensor(
-                    [
-                        [1.0, 2.0, 3.0, 4.0],
-                        [5.0, 6.0, 7.0, 8.0],
-                        [9.0, 10.0, 11.0, 12.0],
-                    ],
-                    dtype=torch.float32,
-                ).to(torch.float8_e4m3fn)
-                self.weight = nn.Parameter(logical_weight.t(), requires_grad=False)
-                self.weight_scale = nn.Parameter(torch.tensor([0.25]), requires_grad=False)
-
-        current_block = OnlineFp8Block()
-        rank0_block = OnlineFp8Block()
-        rank1_block = OnlineFp8Block()
-        expected_weight = rank0_block.weight.detach().clone()
-        expected_scale = rank0_block.weight_scale.detach().clone()
-        expected_stride = rank0_block.weight.stride()
-
-        rank0_hook = DistributedLayerwiseOffloadHook(
-            next_block=rank0_block,
-            device=torch.device("cpu"),
-            dp_group=object(),
-            dp_size=2,
-            rank=0,
-            copy_stream=DummyStream(),
-            comm_stream=DummyStream(),
-            pin_memory=False,
-        )
-        rank1_hook = DistributedLayerwiseOffloadHook(
-            next_block=rank1_block,
-            device=torch.device("cpu"),
-            dp_group=object(),
-            dp_size=2,
-            rank=1,
-            copy_stream=DummyStream(),
-            comm_stream=DummyStream(),
-            pin_memory=False,
-        )
-        rank0_hook.initialize_hook(current_block)
-        rank1_hook.initialize_hook(OnlineFp8Block())
-        rank0_hook.gpu_shard_buffers = [
-            {dtype: torch.empty_like(shard) for dtype, shard in rank0_hook.cpu_shards.items()} for _ in range(2)
-        ]
-
-        def fake_allgather(output, local_shard, *, group):
-            del group
-            remote_shard = rank1_hook.cpu_shards[local_shard.dtype]
-            shard_size = local_shard.numel()
-            output[:shard_size].copy_(local_shard)
-            output[shard_size : 2 * shard_size].copy_(remote_shard)
-
-        monkeypatch.setattr(torch.distributed, "all_gather_into_tensor", fake_allgather)
-        rank0_hook.prefetch_layer(slot=0, non_blocking=False)
-
-        assert rank0_block.weight.stride() == expected_stride
-        torch.testing.assert_close(
-            rank0_block.weight.float(),
-            expected_weight.float(),
-            rtol=0,
-            atol=0,
-        )
-        torch.testing.assert_close(rank0_block.weight_scale, expected_scale)
-
    def test_dtensor_wrapper_preserved_across_prefetch_and_offload(self, dist_group, patched_offload_runtime):
        """DTensor wrapper should be preserved through prefetch/offload cycle."""
        current_block = TinyBlock(_make_values(1.0))
@@ -809,7 +735,7 @@ class _MmapPostLoadPipeline(nn.Module):


 class TestMmapWeightLoading:
-    def test_host_weight_cache_maps_final_weights_without_rerunning_post_load(
+    def test_runtime_cache_maps_final_weights_without_rerunning_post_load(
        self,
        tmp_path,
        patched_offload_runtime,
@@ -834,7 +760,7 @@ class TestMmapWeightLoading:
        weight_file = tmp_path / "runtime.safetensors"
        save_file({"transformer.proj.weight": torch.full((2, 2), 9.0)}, str(weight_file))
        plan = HostWeightPlan(
-            backing_kind="host_weight_cache",
+            backing_kind="runtime_cache",
            bindings={
                "transformer.proj.weight": TensorBinding(
                    storage_key="transformer.proj.weight",
@@ -842,7 +768,7 @@ class TestMmapWeightLoading:
                )
            },
            runtime_layout_key="runtime-layout",
-            expected_file_digests={str(weight_file): file_digest(weight_file)},
+            post_load_complete=True,
        )
        backend = DistributedLayerwiseOffloadBackend(
            OffloadConfig(
@@ -855,7 +781,7 @@ class TestMmapWeightLoading:
        )
        modules = SimpleNamespace(dits=[pipeline.transformer], dit_names=["transformer"])

-        backend._load_weights_from_host_weight_cache(pipeline, modules, plan)
+        backend._load_weights_from_runtime_cache(pipeline, modules, plan)

        assert pipeline.transformer.post_load_calls == 0
        assert pipeline.transformer.proj.weight is old_parameter
@@ -863,7 +789,7 @@ class TestMmapWeightLoading:
        assert torch.equal(pipeline.transformer.proj.weight, torch.full((2, 2), 9.0))
        backend._release_mmap_handles()

-    def test_host_weight_cache_mapping_failure_leaves_ordinary_tensors_intact(
+    def test_runtime_cache_mapping_failure_leaves_ordinary_tensors_intact(
        self,
        tmp_path,
        patched_offload_runtime,
@@ -872,7 +798,7 @@ class TestMmapWeightLoading:
        pipeline.transformer = nn.Linear(2, 2, bias=False)
        old_parameter = pipeline.transformer.weight
        plan = HostWeightPlan(
-            backing_kind="host_weight_cache",
+            backing_kind="runtime_cache",
            bindings={
                "transformer.weight": TensorBinding(
                    storage_key="transformer.weight",
@@ -880,7 +806,7 @@ class TestMmapWeightLoading:
                )
            },
            runtime_layout_key="runtime-layout",
-            expected_file_digests={str(tmp_path / "missing.safetensors"): "unused"},
+            post_load_complete=True,
        )
        backend = DistributedLayerwiseOffloadBackend(
            OffloadConfig(
@@ -894,55 +820,10 @@ class TestMmapWeightLoading:
        modules = SimpleNamespace(dits=[pipeline.transformer], dit_names=["transformer"])

        with pytest.raises(Exception, match="missing.safetensors"):
-            backend._load_weights_from_host_weight_cache(pipeline, modules, plan)
+            backend._load_weights_from_runtime_cache(pipeline, modules, plan)

        assert pipeline.transformer.weight is old_parameter

-    def test_host_weight_cache_digest_change_leaves_ordinary_tensors_intact(
-        self,
-        tmp_path,
-        patched_offload_runtime,
-    ):
-        pipeline = nn.Module()
-        pipeline.transformer = nn.Linear(2, 2, bias=False)
-        old_parameter = pipeline.transformer.weight
-        old_value = old_parameter.detach().clone()
-        weight_file = tmp_path / "runtime.safetensors"
-        save_file({"transformer.weight": torch.full((2, 2), 9.0)}, str(weight_file))
-        expected_digest = file_digest(weight_file)
-        with weight_file.open("r+b") as handle:
-            handle.seek(-1, os.SEEK_END)
-            original = handle.read(1)
-            handle.seek(-1, os.SEEK_END)
-            handle.write(bytes([original[0] ^ 0xFF]))
-        plan = HostWeightPlan(
-            backing_kind="host_weight_cache",
-            bindings={
-                "transformer.weight": TensorBinding(
-                    storage_key="transformer.weight",
-                    file_path=str(weight_file),
-                )
-            },
-            runtime_layout_key="runtime-layout",
-            expected_file_digests={str(weight_file): expected_digest},
-        )
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=False,
-                dlo_use_allgather=False,
-            ),
-            torch.device("cpu"),
-            host_weight_plan=plan,
-        )
-        modules = SimpleNamespace(dits=[pipeline.transformer], dit_names=["transformer"])
-
-        with pytest.raises(RuntimeError, match="changed after loader validation"):
-            backend._load_weights_from_host_weight_cache(pipeline, modules, plan)
-
-        assert pipeline.transformer.weight is old_parameter
-        assert torch.equal(pipeline.transformer.weight, old_value)
-
    def test_runs_model_post_load_hook(self, tmp_path, patched_offload_runtime):
        pipeline = _MmapPostLoadPipeline()
        weights = {name: torch.ones(param.shape, dtype=torch.bfloat16) for name, param in pipeline.named_parameters()}
@@ -1037,7 +918,7 @@ class TestMmapWeightLoading:
        assert all(hook.rank_local_mmap for group in backend._all_hook_groups for hook in group)
        backend.disable()

-    def test_host_weight_cache_enable_reclaims_private_weights_before_hook_setup(
+    def test_runtime_cache_enable_reclaims_private_weights_before_hook_setup(
        self,
        tmp_path,
        patched_offload_runtime,
@@ -1059,10 +940,10 @@ class TestMmapWeightLoading:
        weight_file = tmp_path / "runtime.safetensors"
        save_file(weights, str(weight_file))
        plan = HostWeightPlan(
-            backing_kind="host_weight_cache",
+            backing_kind="runtime_cache",
            bindings={name: TensorBinding(storage_key=name, file_path=str(weight_file)) for name in weights},
            runtime_layout_key="runtime-layout",
-            expected_file_digests={str(weight_file): file_digest(weight_file)},
+            post_load_complete=True,
        )
        backend = DistributedLayerwiseOffloadBackend(
            OffloadConfig(
@@ -1084,78 +965,7 @@ class TestMmapWeightLoading:
        assert len(cleanup_calls) == 2
        backend.disable()

-    def test_host_weight_cache_enable_falls_back_to_ordinary_tensors(
-        self,
-        tmp_path,
-        patched_offload_runtime,
-    ):
-        class Transformer(nn.Module):
-            _layerwise_offload_blocks_attrs = ["blocks"]
-
-            def __init__(self):
-                super().__init__()
-                self.blocks = nn.ModuleList([nn.Linear(2, 2, bias=False) for _ in range(2)])
-
-        class Pipeline(nn.Module):
-            def __init__(self):
-                super().__init__()
-                self.transformer = Transformer()
-
-        pipeline = Pipeline()
-        weight_file = tmp_path / "runtime.safetensors"
-        weights = {name: torch.full_like(param, 4) for name, param in pipeline.named_parameters()}
-        save_file(weights, str(weight_file))
-        plan = HostWeightPlan(
-            backing_kind="host_weight_cache",
-            bindings={name: TensorBinding(storage_key=name, file_path=str(weight_file)) for name in weights},
-            runtime_layout_key="runtime-layout",
-            expected_file_digests={str(weight_file): "stale-digest"},
-        )
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=False,
-                dlo_use_allgather=False,
-            ),
-            torch.device("cpu"),
-            host_weight_plan=plan,
-        )
-
-        backend.enable(pipeline)
-
-        assert backend.enabled
-        assert backend.host_weight_plan is None
-        assert not backend._using_mmap
-        assert not backend._using_rank_local_mmap
-        backend.disable()
-
-    def test_disable_synchronizes_allgather_device_work(
-        self,
-        monkeypatch: pytest.MonkeyPatch,
-        patched_offload_runtime,
-    ):
-        synchronize_calls: list[None] = []
-        monkeypatch.setattr(
-            dist_backend_module.current_omni_platform,
-            "synchronize",
-            lambda: synchronize_calls.append(None),
-        )
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=False,
-                dlo_use_allgather=True,
-            ),
-            torch.device("cpu"),
-        )
-        backend.enabled = True
-        backend._using_rank_local_mmap = False
-
-        backend.disable()
-
-        assert synchronize_calls == [None]
-
-    def test_host_weight_cache_registration_uses_budget_and_releases_before_mmap(
+    def test_runtime_cache_registration_uses_budget_and_releases_before_mmap(
        self,
        monkeypatch: pytest.MonkeyPatch,
        patched_offload_runtime,
@@ -1173,25 +983,24 @@ class TestMmapWeightLoading:

        create_calls: list[tuple[dict[str, list[torch.Tensor]], int]] = []

-        def register(sources, *, device, max_bytes):
-            assert device == torch.device("cuda")
+        def create(sources, *, max_bytes):
            create_calls.append((sources, max_bytes))
            return Registration()

-        monkeypatch.setattr(dist_backend_module, "register_host_mappings", register)
+        monkeypatch.setattr(dist_backend_module.CudaHostRegistration, "create", staticmethod(create))
        backend = DistributedLayerwiseOffloadBackend(
            OffloadConfig(
                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
                pin_cpu_memory=True,
                dlo_use_allgather=False,
-                dlo_host_weight_cache_pin_limit_gib=1.5,
+                dlo_runtime_cache_pin_limit_gib=1.5,
            ),
            torch.device("cuda"),
        )
        sources = {"runtime.safetensors": [torch.ones(1)]}
-        backend._host_weight_cache_mapped_sources = sources
+        backend._runtime_cache_mapped_sources = sources

-        assert backend._try_register_host_weight_cache_mmap()
+        assert backend._try_register_runtime_cache_mmap()
        assert create_calls == [(sources, int(1.5 * 1024**3))]

        backend.enabled = True
@@ -1202,127 +1011,6 @@ class TestMmapWeightLoading:

        assert events == ["unregister", "mmap"]

-    def test_host_weight_cache_registration_defaults_to_pinned_memory_policy(
-        self,
-        monkeypatch: pytest.MonkeyPatch,
-        patched_offload_runtime,
-    ):
-        class Registration:
-            total_bytes = 4096
-            region_count = 1
-
-            @staticmethod
-            def close() -> list[str]:
-                return []
-
-        registration = Registration()
-        create_calls: list[tuple[dict[str, list[torch.Tensor]], int | None]] = []
-
-        def register(sources, *, device, max_bytes):
-            assert device == torch.device("cuda")
-            create_calls.append((sources, max_bytes))
-            return registration
-
-        monkeypatch.setattr(dist_backend_module, "register_host_mappings", register)
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=True,
-                dlo_use_allgather=False,
-                dlo_host_weight_cache_pin_limit_gib=0.0,
-            ),
-            torch.device("cuda"),
-        )
-        sources = {"runtime.safetensors": [torch.ones(1)]}
-        backend._host_weight_cache_mapped_sources = sources
-
-        assert backend._try_register_host_weight_cache_mmap()
-        assert create_calls == [(sources, None)]
-        assert backend._host_registration is registration
-
-    def test_host_weight_cache_registration_respects_disabled_pinned_memory(
-        self,
-        monkeypatch: pytest.MonkeyPatch,
-        patched_offload_runtime,
-    ):
-        monkeypatch.setattr(
-            dist_backend_module,
-            "register_host_mappings",
-            lambda *_args, **_kwargs: pytest.fail("registration must follow pin_cpu_memory=False"),
-        )
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=False,
-                dlo_use_allgather=False,
-                dlo_host_weight_cache_pin_limit_gib=0.0,
-            ),
-            torch.device("cuda"),
-        )
-        backend._host_weight_cache_mapped_sources = {"runtime.safetensors": [torch.ones(1)]}
-
-        assert not backend._try_register_host_weight_cache_mmap()
-        assert backend._host_registration is None
-
-    def test_host_weight_cache_transfer_is_configured_once_for_all_consumers(
-        self,
-        monkeypatch: pytest.MonkeyPatch,
-        patched_offload_runtime,
-    ):
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=True,
-                dlo_use_allgather=False,
-            ),
-            torch.device("cuda"),
-            host_weight_plan=HostWeightPlan(
-                backing_kind="host_weight_cache",
-                bindings={},
-            ),
-        )
-        backend._using_rank_local_mmap = True
-        backend._host_weight_cache_mapped_sources = {"runtime.safetensors": [torch.ones(1)]}
-        streamed_hook = SimpleNamespace(registered_mmap=False)
-        resident_group = SimpleNamespace(
-            registered_mmap=False,
-            _cpu_staging_buffers=[{"buffer": torch.ones(1)}],
-        )
-        backend._resident_layer_group = resident_group
-        register_calls: list[None] = []
-
-        def register() -> bool:
-            register_calls.append(None)
-            return True
-
-        monkeypatch.setattr(backend, "_try_register_host_weight_cache_mmap", register)
-
-        backend._configure_host_weight_cache_transfer([streamed_hook])  # type: ignore[list-item]
-
-        assert register_calls == [None]
-        assert streamed_hook.registered_mmap
-        assert resident_group.registered_mmap
-        assert resident_group._cpu_staging_buffers == []
-        assert backend._host_weight_cache_mapped_sources == {}
-
-    def test_host_weight_cache_registration_falls_back_on_unsupported_platform(
-        self,
-        patched_offload_runtime,
-    ):
-        backend = DistributedLayerwiseOffloadBackend(
-            OffloadConfig(
-                strategy=OffloadStrategy.DISTRIBUTED_LAYER_WISE,
-                pin_cpu_memory=True,
-                dlo_use_allgather=False,
-                dlo_host_weight_cache_pin_limit_gib=1.0,
-            ),
-            torch.device("cpu"),
-        )
-        backend._host_weight_cache_mapped_sources = {"runtime.safetensors": [torch.ones(1)]}
-
-        assert not backend._try_register_host_weight_cache_mmap()
-        assert backend._host_registration is None
-

 class TestGetBlocksFromDit:
    def test_get_blocks_from_dit_single_block_attr(self):
@@ -130,21 +130,6 @@ class TestCreateDefaultDiffusion:

        assert od.diffusion_kv_mode is DiffusionKVCacheMode.PAGED_SCHEDULER

-    def test_diffusion_kv_sizing_fields_roundtrip(self, monkeypatch):
-        monkeypatch.setattr(OmniDiffusionConfig, "_resolve_master_port", lambda self: 29500)
-        od = _roundtrip_diffusion_config(
-            model="x",
-            kv_cache_memory_bytes=4096,
-            gpu_memory_utilization=0.75,
-            max_num_batched_tokens=2048,
-            max_model_len=4096,
-        )
-
-        assert od.kv_cache_memory_bytes == 4096
-        assert od.gpu_memory_utilization == 0.75
-        assert od.max_num_batched_tokens == 2048
-        assert od.max_model_len == 4096
-

 def test_qwen_image_edit_plus_sets_generic_multimodal_limit():
    od_config = OmniDiffusionConfig(model="Qwen/Qwen-Image-Edit-2511", model_class_name="QwenImageEditPlusPipeline")
@@ -161,83 +161,9 @@ def test_scheduler_initialization_failure_closes_scheduler_and_executor(monkeypa
    assert exc_info.value is initialization_error
    custom_scheduler.initialize.assert_called_once_with(od_config)
    custom_scheduler.close.assert_called_once_with()
-
-
-def test_init_shuts_down_executor_when_kv_control_plane_initialization_fails(
-    monkeypatch: pytest.MonkeyPatch,
-) -> None:
-    fake_executor = SimpleNamespace(shutdown=Mock())
-    monkeypatch.setattr(DiffusionEngine, "_init_process_hooks", lambda self, config: None)
-    monkeypatch.setattr(
-        DiffusionEngine,
-        "_resolve_execution_mode",
-        lambda self, config: DiffusionExecutionMode.REQUEST_BATCH,
-    )
-    monkeypatch.setattr(
-        DiffusionEngine, "_init_executor", lambda self, config: setattr(self, "executor", fake_executor)
-    )
-    monkeypatch.setattr(
-        "vllm_omni.diffusion.diffusion_engine.initialize_diffusion_kv_control_plane",
-        Mock(side_effect=RuntimeError("KV initialization failed")),
-    )
-
-    with pytest.raises(RuntimeError, match="KV initialization failed"):
-        DiffusionEngine(SimpleNamespace())
-
    fake_executor.shutdown.assert_called_once_with()


-def test_paged_init_profiles_before_scheduler_initialization(monkeypatch: pytest.MonkeyPatch) -> None:
-    events = []
-    expected_profile_requests = [object()]
-    fake_executor = SimpleNamespace(shutdown=Mock())
-    kv_cache_config = object()
-    kv_vllm_config = object()
-    od_config = SimpleNamespace(diffusion_kv_mode="paged_scheduler")
-
-    monkeypatch.setattr(DiffusionEngine, "_init_process_hooks", lambda self, config: None)
-    monkeypatch.setattr(
-        DiffusionEngine,
-        "_resolve_execution_mode",
-        lambda self, config: DiffusionExecutionMode.REQUEST_BATCH,
-    )
-    monkeypatch.setattr(
-        DiffusionEngine,
-        "_init_executor",
-        lambda self, config: setattr(self, "executor", fake_executor),
-    )
-    monkeypatch.setattr(
-        DiffusionEngine,
-        "_prepare_diffusion_kv_profile_requests",
-        lambda self: events.append("prepare-profile") or expected_profile_requests,
-    )
-
-    def initialize(executor, config, *, profile_requests):
-        events.append("profile-workers")
-        assert executor is fake_executor
-        assert config is od_config
-        assert profile_requests is expected_profile_requests
-        return kv_cache_config, 16, 16, kv_vllm_config
-
-    monkeypatch.setattr(
-        "vllm_omni.diffusion.diffusion_engine.initialize_diffusion_kv_control_plane",
-        initialize,
-    )
-
-    def init_scheduler(self, config, scheduler, *args, **kwargs):
-        events.append("initialize-scheduler")
-        self.scheduler = SimpleNamespace(close=Mock())
-
-    monkeypatch.setattr(DiffusionEngine, "_init_scheduler", init_scheduler)
-    monkeypatch.setattr(DiffusionEngine, "_init_runtime_state", lambda self: None)
-    monkeypatch.setattr(DiffusionEngine, "_init_execute_fn", lambda self: None)
-    monkeypatch.setattr(DiffusionEngine, "_log_execution_mode", lambda self, config: None)
-
-    DiffusionEngine(od_config)
-
-    assert events == ["prepare-profile", "profile-workers", "initialize-scheduler"]
-
-
@pytest.mark.asyncio
 async def test_step_compatibility_wrapper_returns_final_batch() -> None:
    engine = _make_engine()
@@ -1,17 +1,9 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project

-from types import SimpleNamespace
-from unittest.mock import Mock
-
 import pytest

 from vllm_omni.diffusion import io_support
-from vllm_omni.diffusion.diffusion_engine import DiffusionEngine, DiffusionExecutionMode
-from vllm_omni.diffusion.diffusion_kv.config import DiffusionKVCacheMode
-from vllm_omni.diffusion.diffusion_kv.request import DiffusionKVRequest
-from vllm_omni.diffusion.request import OmniDiffusionRequest
-from vllm_omni.inputs.data import OmniDiffusionSamplingParams

 pytestmark = [pytest.mark.core_model, pytest.mark.cpu, pytest.mark.diffusion]

@@ -63,151 +55,3 @@ def test_dummy_run_num_frames_uses_audio_input_fallback(monkeypatch: pytest.Monk
    )

    assert io_support.get_dummy_run_num_frames("unknown", supports_audio_input=True) == 2
-
-
-def test_dummy_run_image_count_resolves_hunyuan_architecture_alias() -> None:
-    assert io_support.get_dummy_run_num_image_inputs("HunyuanImage3ForCausalMM") == 3
-    assert io_support.get_dummy_run_num_image_inputs("unknown") == 1
-
-
-def test_dense_mode_does_not_build_kv_profile_request() -> None:
-    engine = object.__new__(DiffusionEngine)
-    engine.od_config = SimpleNamespace(diffusion_kv_mode=DiffusionKVCacheMode.DENSE_LEGACY)
-    engine._make_dummy_request = Mock(side_effect=AssertionError("dense mode must not profile"))
-
-    assert engine._prepare_diffusion_kv_profile_requests() is None
-    engine._make_dummy_request.assert_not_called()
-
-
-@pytest.mark.parametrize(
-    ("execution_mode", "uses_dlo_dp", "max_num_seqs", "expected_profile_requests"),
-    [
-        (DiffusionExecutionMode.STEP_BATCH, False, 2, 2),
-        (DiffusionExecutionMode.REQUEST_BATCH, True, 4, 1),
-        (DiffusionExecutionMode.STEP_BATCH, True, 3, 3),
-    ],
-)
-def test_paged_kv_profile_requests_match_per_rank_batch(
-    execution_mode: DiffusionExecutionMode,
-    uses_dlo_dp: bool,
-    max_num_seqs: int,
-    expected_profile_requests: int,
-) -> None:
-    engine = object.__new__(DiffusionEngine)
-    engine.execution_mode = execution_mode
-    engine.od_config = SimpleNamespace(
-        diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER,
-        model_class_name="HunyuanImage3ForCausalMM",
-        max_num_seqs=max_num_seqs,
-        parallel_config=SimpleNamespace(data_parallel_size=max_num_seqs if uses_dlo_dp else 1),
-        enable_distributed_layerwise_offload=uses_dlo_dp,
-        dlo_use_allgather=True,
-    )
-    request = OmniDiffusionRequest(
-        prompt="profile",
-        request_id="profile-request",
-        sampling_params=OmniDiffusionSamplingParams(num_inference_steps=1),
-    )
-    engine._make_dummy_request = Mock(return_value=request)
-    prepared_layout = object()
-    scheduler_kv_state = (
-        DiffusionKVRequest(
-            "profile-request/diffusion-kv/0",
-            sequence_id=0,
-            prefix_len=3,
-            target_len=8,
-            seq_len=12,
-        ),
-        DiffusionKVRequest(
-            "profile-request/diffusion-kv/1",
-            sequence_id=1,
-            prefix_len=4,
-            target_len=8,
-            seq_len=13,
-        ),
-    )
-
-    def preprocess(req: OmniDiffusionRequest) -> OmniDiffusionRequest:
-        req.prepared_layout = prepared_layout
-        req.diffusion_kv_requests = scheduler_kv_state
-        return req
-
-    engine._prepare_request_for_admission = Mock(side_effect=preprocess)
-
-    result = engine._prepare_diffusion_kv_profile_requests()
-
-    assert result is not None
-    assert len(result) == expected_profile_requests
-    assert len({profile_request.request_id for profile_request in result}) == expected_profile_requests
-    assert all(profile_request.is_dummy_run() for profile_request in result)
-    assert all(profile_request.prepared_layout is prepared_layout for profile_request in result)
-    assert all(profile_request.diffusion_kv_requests is None for profile_request in result)
-    assert len({id(profile_request.sampling_params) for profile_request in result}) == expected_profile_requests
-    assert engine._diffusion_kv_profile_limits == (2, 13, 8)
-    assert len(result) * engine._diffusion_kv_profile_limits[0] == expected_profile_requests * 2
-    engine._make_dummy_request.assert_called_once_with(
-        height=1024,
-        width=1024,
-        guidance_scale=5.0,
-        num_image_inputs=3,
-    )
-    engine._prepare_request_for_admission.assert_called_once_with(request)
-
-
-@pytest.mark.parametrize(
-    ("num_sequences", "seq_len", "target_len"),
-    [
-        (3, 16, 8),
-        (1, 17, 8),
-        (1, 16, 9),
-    ],
-)
-def test_paged_kv_admission_rejects_shape_beyond_profile_envelope(
-    num_sequences: int,
-    seq_len: int,
-    target_len: int,
-) -> None:
-    engine = object.__new__(DiffusionEngine)
-    engine._diffusion_kv_profile_limits = (2, 16, 8)
-    engine.pre_process_func = lambda request: request
-    request = OmniDiffusionRequest(
-        prompt="too large",
-        request_id="too-large",
-        sampling_params=OmniDiffusionSamplingParams(num_inference_steps=1),
-        diffusion_kv_requests=tuple(
-            DiffusionKVRequest(
-                f"too-large/diffusion-kv/{sequence_id}",
-                sequence_id=sequence_id,
-                prefix_len=0,
-                target_len=target_len,
-                seq_len=seq_len,
-            )
-            for sequence_id in range(num_sequences)
-        ),
-    )
-
-    with pytest.raises(ValueError, match="exceeds the startup memory-profile envelope"):
-        engine._prepare_request_for_admission(request)
-
-
-def test_paged_kv_admission_accepts_shape_at_profile_envelope() -> None:
-    engine = object.__new__(DiffusionEngine)
-    engine._diffusion_kv_profile_limits = (2, 16, 8)
-    engine.pre_process_func = lambda request: request
-    request = OmniDiffusionRequest(
-        prompt="fits",
-        request_id="fits",
-        sampling_params=OmniDiffusionSamplingParams(num_inference_steps=1),
-        diffusion_kv_requests=tuple(
-            DiffusionKVRequest(
-                f"fits/diffusion-kv/{sequence_id}",
-                sequence_id=sequence_id,
-                prefix_len=8,
-                target_len=8,
-                seq_len=16,
-            )
-            for sequence_id in range(2)
-        ),
-    )
-
-    assert engine._prepare_request_for_admission(request) is request
@@ -473,38 +473,6 @@ async def test_cancelled_future_does_not_kill_busy_loop():
        _stop_engine(engine)


-@pytest.mark.asyncio
-async def test_native_kv_reservation_error_wakes_stream_without_killing_busy_loop():
-    loop = asyncio.get_running_loop()
-    engine = _make_engine_with_loop(loop)
-
-    def fail_reservation(*args, **kwargs):
-        raise RuntimeError("native allocation bug")
-
-    engine.scheduler._diffusion_kv_manager = SimpleNamespace(
-        has_request=lambda request_id: False,
-        reserve_request=fail_reservation,
-        free_request=lambda request_id: None,
-    )
-    request = _make_request("kv-error")
-    request.diffusion_kv_requests = (object(),)
-    try:
-        response_stream = engine.async_add_req_and_stream_response(request)
-        output_queue = engine._out_streams[request.request_id]
-        for _ in range(300):
-            if not output_queue.empty():
-                break
-            await asyncio.sleep(0.01)
-        assert not output_queue.empty(), "terminal KV allocation error was not delivered to the request stream"
-
-        output = await anext(response_stream)
-        assert output.error == "native allocation bug"
-        assert output.finished
-        assert engine.worker_thread.is_alive()
-    finally:
-        _stop_engine(engine)
-
-
 # ─────────────────────────── _RpcTask basics ───────────────────────────────


@@ -3,7 +3,6 @@

 from contextlib import contextmanager
 from types import SimpleNamespace
-from unittest.mock import Mock

 import pytest
 import torch
@@ -513,84 +512,6 @@ def test_execute_model_accepts_bare_diffusion_output_from_single_request_pipelin
    assert runner.pipeline.last_req.num_reqs == 1


-@pytest.mark.core_model
-@pytest.mark.cpu
-def test_profile_run_executes_forward_without_scheduler_kv_validation(monkeypatch):
-    runner = _make_runner(cache_backend=None, cache_backend_name="none")
-    request = _make_request()
-    runner._validate_diffusion_kv_metadata = Mock(side_effect=AssertionError("profile must bypass admission"))
-    record_names = []
-    original_execute = runner._execute_request_list
-
-    def execute(*args, **kwargs):
-        record_names.append(kwargs["record_name"])
-        return original_execute(*args, **kwargs)
-
-    runner._execute_request_list = execute
-    monkeypatch.setattr(model_runner_module, "set_forward_context", _noop_forward_context)
-    reset_peak_memory_stats = Mock()
-    monkeypatch.setattr(
-        model_runner_module.current_omni_platform,
-        "reset_peak_memory_stats",
-        reset_peak_memory_stats,
-    )
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "max_memory_reserved", lambda: 0)
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "max_memory_allocated", lambda: 0)
-    synchronize = Mock()
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "synchronize", synchronize)
-
-    runner.profile_run([request])
-
-    assert runner.pipeline.forward_calls == 1
-    assert record_names == ["pipeline_memory_profile"]
-    runner._validate_diffusion_kv_metadata.assert_not_called()
-    reset_peak_memory_stats.assert_not_called()
-    synchronize.assert_called_once_with()
-
-
-@pytest.mark.core_model
-@pytest.mark.cpu
-def test_profile_run_executes_maximum_step_batch_without_resetting_peak(monkeypatch):
-    runner = _make_runner(cache_backend=None, cache_backend_name=None)
-    runner.pipeline = _FinalOnlyStepPipeline()
-    runner.od_config.step_execution = True
-    requests = [_make_request(), _make_request()]
-    requests[0].request_id = "profile-0"
-    requests[1].request_id = "profile-1"
-    for request in requests:
-        request.sampling_params.num_inference_steps = 1
-
-    observed_batch_rows = []
-    original_denoise_step = runner.pipeline.denoise_step
-
-    def denoise_step(input_batch, states):
-        observed_batch_rows.append((len(states), int(input_batch.latents.shape[0])))
-        return original_denoise_step(input_batch, states)
-
-    runner.pipeline.denoise_step = denoise_step
-    runner._validate_diffusion_kv_metadata = Mock(side_effect=AssertionError("profile must bypass admission"))
-    monkeypatch.setattr(model_runner_module, "set_forward_context", _noop_forward_context)
-    reset_peak_memory_stats = Mock()
-    monkeypatch.setattr(
-        model_runner_module.current_omni_platform,
-        "reset_peak_memory_stats",
-        reset_peak_memory_stats,
-    )
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "max_memory_reserved", lambda: 0)
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "max_memory_allocated", lambda: 0)
-    synchronize = Mock()
-    monkeypatch.setattr(model_runner_module.current_omni_platform, "synchronize", synchronize)
-
-    runner.profile_run(requests)
-
-    assert observed_batch_rows == [(2, 2)]
-    assert runner.state_cache == {}
-    assert runner.input_batch is None
-    runner._validate_diffusion_kv_metadata.assert_not_called()
-    reset_peak_memory_stats.assert_not_called()
-    synchronize.assert_called_once_with()
-
-
 class _BatchPipeline:
    """Pipeline returning a configurable list of outputs from forward()."""

@@ -52,7 +52,6 @@ def test_dummy_run_request_is_identified_by_reserved_request_id():

    assert req.is_dummy_run()
    assert OmniDiffusionRequest.is_dummy_run_request_id(DUMMY_DIFFUSION_REQUEST_ID)
-    assert OmniDiffusionRequest.is_dummy_run_request_id(f"{DUMMY_DIFFUSION_REQUEST_ID}/kv-profile-1")


 def test_non_dummy_request_is_not_identified_as_dummy_run():
@@ -8,14 +8,12 @@ from types import SimpleNamespace

 import pytest
 import torch
-import vllm.v1.core.single_type_kv_cache_manager as native_kv_managers
 from pytest_mock import MockerFixture
-from vllm.v1.kv_cache_interface import FullAttentionSpec, KVCacheConfig, KVCacheGroupSpec, KVCacheTensor

 from vllm_omni.diffusion.data import DiffusionOutput, DiffusionRequestAbortedError
 from vllm_omni.diffusion.diffusion_engine import DiffusionEngine, DiffusionExecutionMode
 from vllm_omni.diffusion.diffusion_kv.config import DiffusionKVCacheMode
-from vllm_omni.diffusion.diffusion_kv.request import DiffusionKVRequest
+from vllm_omni.diffusion.diffusion_kv.request import DiffusionKVContext, DiffusionKVRequest
 from vllm_omni.diffusion.request import OmniDiffusionRequest
 from vllm_omni.diffusion.sched import (
    BaseScheduler,
@@ -89,52 +87,6 @@ def _cached_ids(sched_output) -> list[str]:
    return list(sched_output.scheduled_cached_reqs.request_ids)


-def _initialize_paged_scheduler(
-    scheduler: BaseScheduler,
-    *,
-    num_blocks: int = 64,
-    max_num_seqs: int = 1,
-) -> None:
-    native_kv_managers.register_all_kvcache_specs(None)
-    spec = FullAttentionSpec(
-        block_size=4,
-        num_kv_heads=2,
-        head_size=8,
-        dtype=torch.bfloat16,
-    )
-    config = KVCacheConfig(
-        num_blocks=num_blocks,
-        kv_cache_tensors=[KVCacheTensor(size=spec.page_size_bytes * num_blocks, shared_by=["layer0"])],
-        kv_cache_groups=[KVCacheGroupSpec(layer_names=["layer0"], kv_cache_spec=spec)],
-    )
-    scheduler.initialize(
-        SimpleNamespace(
-            diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER,
-            max_model_len=64,
-            max_num_seqs=max_num_seqs,
-        ),
-        kv_cache_config=config,
-        scheduler_block_size=4,
-        hash_block_size=4,
-        kv_vllm_config=SimpleNamespace(
-            model_config=SimpleNamespace(max_model_len=64),
-            max_in_flight_tokens=64,
-        ),
-    )
-
-
-def _attach_diffusion_kv(request: OmniDiffusionRequest, *, seq_len: int = 8) -> None:
-    request.diffusion_kv_requests = (
-        DiffusionKVRequest(
-            f"{request.request_id}/diffusion-kv/0",
-            sequence_id=0,
-            prefix_len=4,
-            target_len=4,
-            seq_len=seq_len,
-        ),
-    )
-
-
 class _StubScheduler:
    def __init__(self, request: OmniDiffusionRequest, output) -> None:
        self._request = request
@@ -448,16 +400,18 @@ class TestRequestScheduler:
        assert self.scheduler.has_requests() is False

    def test_diffusion_kv_request_moves_to_scheduler_state(self) -> None:
-        _initialize_paged_scheduler(self.scheduler)
+        self.scheduler.initialize(SimpleNamespace(diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER))
        request = _make_request("diffusion-kv")
        prepared_layout = object()
        request.prepared_layout = prepared_layout
+        text_context = DiffusionKVContext(context_id="text", cache_role="cross.text", num_tokens=8)
        kv_request = DiffusionKVRequest(
            "diffusion-kv/diffusion-kv/0",
            sequence_id=0,
            prefix_len=4,
            target_len=8,
            seq_len=16,
+            kv_contexts=(text_context,),
        )
        request.diffusion_kv_requests = (kv_request,)

@@ -467,179 +421,12 @@ class TestRequestScheduler:

        assert state is not None
        assert state.diffusion_kv_requests == (kv_request,)
+        assert state.diffusion_kv_requests[0].kv_contexts == (text_context,)
        assert request.diffusion_kv_requests is None
        assert scheduler_output.scheduled_new_reqs[0].req.prepared_layout is prepared_layout
-        metadata = scheduler_output.scheduled_new_reqs[0].diffusion_kv_metadata
-        assert metadata is not None
-        assert metadata.request_id == request.request_id
-        assert len(metadata.sequences[0].block_ids[0]) == 4
-
-    def test_diffusion_kv_capacity_backpressures_fifo_until_blocks_are_freed(self) -> None:
-        _initialize_paged_scheduler(self.scheduler, num_blocks=3, max_num_seqs=2)
-        first_request = _make_request("first")
-        second_request = _make_request("second")
-        _attach_diffusion_kv(first_request)
-        _attach_diffusion_kv(second_request)
-        self.scheduler.add_request(first_request)
-        self.scheduler.add_request(second_request)
-
-        first_output = self.scheduler.schedule()
-
-        assert _new_ids(first_output) == ["first"]
-        assert first_output.num_waiting_reqs == 1
-
-        self.scheduler.update_from_output(first_output, _make_request_output("first"))
-        second_output = self.scheduler.schedule()
-
-        assert _new_ids(second_output) == ["second"]
-        assert second_output.num_waiting_reqs == 0
-        # The same output carries both the release and the newly admitted block
-        # table; the future Worker data plane must process finished ids first.
-        assert second_output.finished_req_ids == {"first"}
-        assert second_output.scheduled_new_reqs[0].diffusion_kv_metadata is not None
-
-    def test_impossible_diffusion_kv_capacity_finishes_only_that_request(self) -> None:
-        _initialize_paged_scheduler(self.scheduler, num_blocks=2, max_num_seqs=2)
-        impossible = _make_request("impossible")
-        schedulable = _make_request("schedulable")
-        _attach_diffusion_kv(impossible, seq_len=8)
-        schedulable.diffusion_kv_requests = (
-            DiffusionKVRequest(
-                "schedulable/diffusion-kv/0",
-                sequence_id=0,
-                prefix_len=0,
-                target_len=4,
-                seq_len=4,
-            ),
-        )
-        self.scheduler.add_request(impossible)
-        self.scheduler.add_request(schedulable)
-
-        sched_output = self.scheduler.schedule()
-
-        failed_state = self.scheduler.get_request_state("impossible")
-        assert failed_state is not None
-        assert failed_state.status == DiffusionRequestStatus.FINISHED_ERROR
-        assert failed_state.error is not None
-        assert "cannot fit even when the block pool is empty" in failed_state.error
-        assert sched_output.finished_req_ids == {"impossible"}
-        assert _new_ids(sched_output) == ["schedulable"]
-
-        finished = self.scheduler.update_from_output(
-            sched_output,
-            _make_request_output("schedulable"),
-        )
-
-        assert finished == {"impossible", "schedulable"}
-
-    def test_impossible_diffusion_kv_capacity_does_not_block_waiters_under_load(self) -> None:
-        _initialize_paged_scheduler(self.scheduler, num_blocks=4, max_num_seqs=2)
-        running = _make_request("running")
-        _attach_diffusion_kv(running)
-        self.scheduler.add_request(running)
-        assert _new_ids(self.scheduler.schedule()) == ["running"]
-
-        impossible = _make_request("impossible")
-        impossible.diffusion_kv_requests = tuple(
-            DiffusionKVRequest(
-                f"impossible/diffusion-kv/{sequence_id}",
-                sequence_id=sequence_id,
-                prefix_len=0,
-                target_len=4,
-                seq_len=4,
-            )
-            for sequence_id in range(4)
-        )
-        schedulable = _make_request("schedulable")
-        schedulable.diffusion_kv_requests = (
-            DiffusionKVRequest(
-                "schedulable/diffusion-kv/0",
-                sequence_id=0,
-                prefix_len=0,
-                target_len=4,
-                seq_len=4,
-            ),
-        )
-        self.scheduler.add_request(impossible)
-        self.scheduler.add_request(schedulable)
-
-        sched_output = self.scheduler.schedule()
-
-        failed_state = self.scheduler.get_request_state("impossible")
-        assert failed_state is not None
-        assert failed_state.status == DiffusionRequestStatus.FINISHED_ERROR
-        assert failed_state.error is not None
-        assert "required_blocks=4, available_blocks=3" in failed_state.error
-        assert sched_output.finished_req_ids == {"impossible"}
-        assert _new_ids(sched_output) == ["schedulable"]
-
-    def test_diffusion_kv_internal_allocation_error_finishes_request(self, monkeypatch) -> None:
-        _initialize_paged_scheduler(self.scheduler)
-        request = _make_request("native-error")
-        _attach_diffusion_kv(request)
-        self.scheduler.add_request(request)
-        manager = self.scheduler._diffusion_kv_manager
-        assert manager is not None
-
-        def raise_native_error(*args, **kwargs):
-            raise ValueError("native allocation bug")
-
-        monkeypatch.setattr(
-            manager,
-            "reserve_request",
-            raise_native_error,
-        )
-
-        sched_output = self.scheduler.schedule()
-
-        state = self.scheduler.get_request_state("native-error")
-        assert state is not None
-        assert state.status == DiffusionRequestStatus.FINISHED_ERROR
-        assert state.error == "native allocation bug"
-        assert sched_output.finished_req_ids == {"native-error"}
-        assert sched_output.scheduled_request_ids == []
-
-    def test_diffusion_kv_preemption_retains_allocation(self) -> None:
-        _initialize_paged_scheduler(self.scheduler, num_blocks=3)
-        request = _make_request("preempted")
-        _attach_diffusion_kv(request)
-        self.scheduler.add_request(request)
-        self.scheduler.schedule()
-        manager = self.scheduler._diffusion_kv_manager
-        assert manager is not None
-        metadata = manager.get_metadata("preempted")
-
-        assert self.scheduler.preempt_request("preempted") is True
-        resumed_output = self.scheduler.schedule()
-
-        assert resumed_output.scheduled_new_reqs == []
-        assert _cached_ids(resumed_output) == ["preempted"]
-        assert manager.get_metadata("preempted") == metadata
-
-    @pytest.mark.parametrize(
-        "status",
-        [
-            DiffusionRequestStatus.FINISHED_COMPLETED,
-            DiffusionRequestStatus.FINISHED_ABORTED,
-            DiffusionRequestStatus.FINISHED_ERROR,
-        ],
-    )
-    def test_diffusion_kv_terminal_status_releases_allocation(self, status: DiffusionRequestStatus) -> None:
-        _initialize_paged_scheduler(self.scheduler, num_blocks=3)
-        request = _make_request("terminal")
-        _attach_diffusion_kv(request)
-        self.scheduler.add_request(request)
-        self.scheduler.schedule()
-        manager = self.scheduler._diffusion_kv_manager
-        assert manager is not None
-        assert manager.has_request("terminal") is True
-
-        self.scheduler.finish_requests("terminal", status)
-
-        assert manager.has_request("terminal") is False

    def test_paged_scheduler_rejects_missing_kv_request(self) -> None:
-        _initialize_paged_scheduler(self.scheduler)
+        self.scheduler.initialize(SimpleNamespace(diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER))
        request = _make_request("diffusion-kv")

        with pytest.raises(ValueError, match="did not produce DiffusionKVRequest"):
@@ -656,7 +443,7 @@ class TestRequestScheduler:
        ],
    )
    def test_diffusion_kv_request_rejects_legacy_dense_kv(self, owner_name: str, field_name: str) -> None:
-        _initialize_paged_scheduler(self.scheduler)
+        self.scheduler.initialize(SimpleNamespace(diffusion_kv_mode=DiffusionKVCacheMode.PAGED_SCHEDULER))
        request = _make_request("diffusion-kv")
        owner = request if owner_name == "request" else request.sampling_params
        setattr(owner, field_name, object())
@@ -1134,45 +921,6 @@ class TestDiffusionEngine:
        assert engine.scheduler.max_num_running_reqs == 1
        fake_executor_cls.assert_called_once_with(od_config)

-    def test_initializes_paged_scheduler_from_native_control_plane(
-        self,
-        monkeypatch: pytest.MonkeyPatch,
-        mocker: MockerFixture,
-    ) -> None:
-        od_config = SimpleNamespace(model_class_name="mock_model", streaming_output=False)
-        fake_executor_cls = mocker.Mock(return_value=mocker.Mock())
-        scheduler = mocker.Mock()
-        kv_cache_config = object()
-        kv_vllm_config = object()
-
-        monkeypatch.setattr(
-            "vllm_omni.diffusion.diffusion_engine.get_diffusion_post_process_func",
-            lambda *args, **kwargs: None,
-        )
-        monkeypatch.setattr(
-            "vllm_omni.diffusion.diffusion_engine.get_diffusion_pre_process_func",
-            lambda *args, **kwargs: None,
-        )
-        monkeypatch.setattr(
-            "vllm_omni.diffusion.diffusion_engine.DiffusionExecutor.get_class",
-            lambda *args, **kwargs: fake_executor_cls,
-        )
-        monkeypatch.setattr(
-            "vllm_omni.diffusion.diffusion_engine.initialize_diffusion_kv_control_plane",
-            lambda *args, **kwargs: (kv_cache_config, 16, 16, kv_vllm_config),
-        )
-
-        engine = DiffusionEngine(od_config, scheduler=scheduler)
-
-        assert engine.scheduler is scheduler
-        scheduler.initialize.assert_called_once_with(
-            od_config,
-            kv_cache_config=kv_cache_config,
-            scheduler_block_size=16,
-            hash_block_size=16,
-            kv_vllm_config=kv_vllm_config,
-        )
-
    def test_scheduler_alias_keeps_default_request_scheduler(self) -> None:
        scheduler = Scheduler()
        scheduler.initialize(SimpleNamespace())
@@ -1239,21 +987,6 @@ class TestDiffusionEngine:
        assert output.aborted is True
        assert output.abort_message == "Request req-finalize aborted."

-    def test_finalize_finished_request_returns_scheduler_admission_error(self) -> None:
-        engine = DiffusionEngine.__new__(DiffusionEngine)
-        engine.scheduler = StepScheduler()
-        engine.scheduler.initialize(SimpleNamespace())
-
-        req_id = engine.scheduler.add_request(_make_request("req-error"))
-        engine.scheduler._finish_requests(
-            {req_id: DiffusionRequestStatus.FINISHED_ERROR},
-            {req_id: "KV request cannot fit"},
-        )
-
-        output = engine._finalize_finished_request(req_id)
-
-        assert output.error == "KV request cannot fit"
-
    @pytest.mark.asyncio
    async def test_streaming_runner_output_notifies_each_chunk(self) -> None:
        engine = DiffusionEngine.__new__(DiffusionEngine)
@@ -13,7 +13,11 @@ import pytest
 import torch
 from pytest_mock import MockerFixture

-from vllm_omni.diffusion.worker.diffusion_worker import DiffusionWorker
+from vllm_omni.diffusion.worker.diffusion_worker import (
+    DiffusionWorker,
+    _create_diffusion_worker_vllm_config,
+    _make_diffusion_vllm_model_config,
+)

 pytestmark = [pytest.mark.core_model, pytest.mark.diffusion, pytest.mark.gpu]

@@ -54,6 +58,34 @@ def mock_gpu_worker(mocker: MockerFixture, mock_od_config):
    return worker


+def test_diffusion_vllm_model_config_supplies_dtype_for_quant_methods():
+    from types import SimpleNamespace
+
+    from vllm_omni.quantization import build_quant_config
+
+    od_config = SimpleNamespace(
+        model="dummy",
+        dtype=torch.bfloat16,
+        quantization_config=build_quant_config(
+            {
+                "quant_method": "modelopt",
+                "quant_algo": "FP8",
+                "ignore": [],
+            }
+        ),
+        tf_model_config=SimpleNamespace(),
+        enforce_eager=True,
+        is_moe=False,
+    )
+
+    model_config = _make_diffusion_vllm_model_config(od_config)
+
+    assert model_config.dtype is torch.bfloat16
+    assert model_config.quantization == "modelopt"
+    assert model_config.quantization_config is od_config.quantization_config
+    assert model_config.is_quantized()
+
+
 class TestDiffusionWorkerSleep:
    """Test DiffusionWorker.sleep method."""

@@ -297,3 +329,49 @@ class TestDiffusionWorkerWakeUp:
        mock_buffer2.data.copy_.assert_not_called()

        assert result is True
+
+
+class TestWorkerVllmConfigAdditionalConfig:
+    """Test worker-side VllmConfig construction with additional_config."""
+
+    def test_create_diffusion_worker_vllm_config_passes_additional_config(self, mocker: MockerFixture):
+        mock_vllm_config = mocker.Mock()
+        mock_vllm_cls = mocker.patch(
+            "vllm_omni.diffusion.worker.diffusion_worker.VllmConfig",
+            return_value=mock_vllm_config,
+        )
+        od_config = mocker.Mock(additional_config={"torchair_graph_config": {"enabled": True}})
+
+        result = _create_diffusion_worker_vllm_config(torch.device("cpu"), od_config)
+
+        assert result is mock_vllm_config
+        assert mock_vllm_cls.call_args.kwargs["additional_config"] == od_config.additional_config
+
+    def test_create_diffusion_worker_vllm_config_falls_back_when_constructor_rejects_additional_config(
+        self, mocker: MockerFixture
+    ):
+        mock_vllm_config = mocker.Mock()
+        mock_vllm_cls = mocker.patch(
+            "vllm_omni.diffusion.worker.diffusion_worker.VllmConfig",
+            side_effect=[
+                TypeError("VllmConfig.__init__() got an unexpected keyword argument 'additional_config'"),
+                mock_vllm_config,
+            ],
+        )
+        od_config = mocker.Mock(additional_config={"ascend_scheduler_config": {"foo": "bar"}})
+
+        result = _create_diffusion_worker_vllm_config(torch.device("cpu"), od_config)
+
+        assert result is mock_vllm_config
+        assert mock_vllm_cls.call_count == 2
+        assert getattr(mock_vllm_config, "additional_config") == od_config.additional_config
+
+    def test_create_diffusion_worker_vllm_config_reraises_other_type_errors(self, mocker: MockerFixture):
+        mocker.patch(
+            "vllm_omni.diffusion.worker.diffusion_worker.VllmConfig",
+            side_effect=TypeError("additional_config values must be hashable"),
+        )
+        od_config = mocker.Mock(additional_config={"ascend_scheduler_config": {"foo": "bar"}})
+
+        with pytest.raises(TypeError, match="hashable"):
+            _create_diffusion_worker_vllm_config(torch.device("cpu"), od_config)
@@ -413,26 +413,6 @@ class TestRequestModeDispatch:
        assert [output.result.error for output in result.runner_outputs] == ["A", "B"]
        executor.collective_rpc.assert_called_once()

-    def test_dlo_dp_forwards_request_metadata_envelopes_as_one_wave(self):
-        executor, _, _ = _make_executor(num_gpus=2)
-        executor.od_config = SimpleNamespace(
-            step_execution=False,
-            parallel_config=SimpleNamespace(data_parallel_size=2),
-            enable_distributed_layerwise_offload=True,
-            dlo_use_allgather=True,
-        )
-        executor.collective_rpc = Mock(return_value=[_tagged_output("A"), _tagged_output("B")])
-        scheduler_output = _make_sched_output("A", "B")
-        for new_req in scheduler_output.scheduled_new_reqs:
-            new_req.diffusion_kv_metadata = SimpleNamespace(request_id=new_req.request_id)
-
-        result = executor.execute_request(scheduler_output)
-
-        assert [output.result.error for output in result.runner_outputs] == ["A", "B"]
-        forwarded_envelopes = executor.collective_rpc.call_args.kwargs["args"][0]
-        assert forwarded_envelopes is scheduler_output.scheduled_new_reqs
-        assert [envelope.diffusion_kv_metadata.request_id for envelope in forwarded_envelopes] == ["A", "B"]
-
    @pytest.mark.parametrize(
        ("field", "value"),
        [
@@ -1728,112 +1728,3 @@ def test_purge_is_noop_on_empty_deques(build_adapter):
    adapter.restore_queues(waiting_queue, running_queue, scheduler_requests={})
    assert running_queue == []
    assert waiting_queue == []
-
-
-# --------------------------------------------------------------------------- #
-#  Chunk-wait deadline (RFC #4855 R1.1, issue #3833)
-#
-#  Before this, the async-chunk path had no deadline of any kind, so a dropped
-#  terminal chunk or an upstream stage that died mid-stream parked the request
-#  in WAITING_FOR_CHUNK forever. None of these scenarios were covered.
-# --------------------------------------------------------------------------- #
-
-
-def _park_in_chunk_wait(adapter, request, *, waiting=True):
-    """Drive a request through one process_pending_chunks round into the wait."""
-    queue = DummyWaitingQueue([request]) if waiting else [request]
-    if waiting:
-        adapter.process_pending_chunks(queue, [])
-    else:
-        adapter.process_pending_chunks(DummyWaitingQueue(), queue)
-    return queue
-
-
-def test_chunk_wait_clock_starts_when_a_request_parks(build_adapter):
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    request = _req("r1", RequestStatus.WAITING)
-
-    _park_in_chunk_wait(adapter, request)
-
-    assert request.status == RequestStatus.WAITING_FOR_CHUNK
-    assert "r1" in adapter._waiting_since
-
-
-def test_chunk_wait_does_not_expire_before_the_deadline(build_adapter):
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    _park_in_chunk_wait(adapter, _req("r1", RequestStatus.WAITING))
-
-    assert adapter.collect_timed_out_request_ids(timeout_s=600.0) == set()
-    assert "r1" in adapter._waiting_since
-
-
-def test_dropped_terminal_chunk_expires_the_request(build_adapter):
-    """The #3833 scenario: upstream stops sending and never marks the stream done."""
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    _park_in_chunk_wait(adapter, _req("r1", RequestStatus.WAITING))
-
-    adapter._waiting_since["r1"] -= 601.0
-
-    assert adapter.collect_timed_out_request_ids(timeout_s=600.0) == {"r1"}
-    # Cleared, so a second sweep does not re-report the same request.
-    assert adapter.collect_timed_out_request_ids(timeout_s=600.0) == set()
-
-
-def test_arriving_chunk_resets_the_clock(build_adapter):
-    """A slow but healthy stream must never expire: the deadline measures stall
-    time between chunks, not the lifetime of the stream."""
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    request = _req("r1", RequestStatus.WAITING)
-    queue = _park_in_chunk_wait(adapter, request)
-    adapter.restore_queues(queue, [], scheduler_requests={"r1": request})
-
-    # Age the wait to just under the deadline, then deliver a chunk.
-    adapter._waiting_since["r1"] -= 599.0
-    adapter._finished_load_reqs.add("r1")
-    adapter.process_pending_chunks(queue, [])
-
-    assert "r1" not in adapter._waiting_since
-    assert adapter.collect_timed_out_request_ids(timeout_s=600.0) == set()
-
-
-def test_a_disabled_deadline_never_expires(build_adapter):
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    _park_in_chunk_wait(adapter, _req("r1", RequestStatus.WAITING))
-    adapter._waiting_since["r1"] -= 10_000.0
-
-    assert adapter.collect_timed_out_request_ids(timeout_s=0.0) == set()
-    assert adapter.collect_timed_out_request_ids(timeout_s=-1.0) == set()
-
-
-def test_finished_request_leaves_no_stale_timestamp(build_adapter):
-    """Otherwise a completed request would be reported as timed out later."""
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    request = _req("r1", RequestStatus.WAITING)
-    _park_in_chunk_wait(adapter, request)
-
-    adapter.finish_requests(["r1"], RequestStatus.FINISHED_STOPPED, {"r1": request})
-
-    assert "r1" not in adapter._waiting_since
-    assert adapter.collect_timed_out_request_ids(timeout_s=0.001) == set()
-
-
-def test_aborted_request_leaves_no_stale_timestamp(build_adapter):
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar")
-    _park_in_chunk_wait(adapter, _req("r1", RequestStatus.WAITING))
-
-    adapter.cleanup_receiver("r1")
-
-    assert "r1" not in adapter._waiting_since
-
-
-def test_expiry_is_per_request(build_adapter):
-    """One stalled stream must not take down its healthy neighbours."""
-    adapter, _ = build_adapter(stage_id=1, model_mode="ar", max_num_seqs=4)
-    stalled = _req("stalled", RequestStatus.WAITING)
-    healthy = _req("healthy", RequestStatus.WAITING)
-    adapter.process_pending_chunks(DummyWaitingQueue([stalled, healthy]), [])
-
-    adapter._waiting_since["stalled"] -= 601.0
-
-    assert adapter.collect_timed_out_request_ids(timeout_s=600.0) == {"stalled"}
-    assert "healthy" in adapter._waiting_since
@@ -463,7 +463,7 @@ def _audio_metrics(reference: np.ndarray, prediction: np.ndarray) -> dict[str, f
    }


-@pytest.mark.slow
+@pytest.mark.full_model
@pytest.mark.benchmark
@pytest.mark.diffusion
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@@ -591,7 +591,7 @@ def test_ltx25_distilled_two_stage_matches_official(accuracy_artifact_root: Path
    assert audio_metrics["relative_l2"] <= LTX25_AUDIO_RELATIVE_L2_THRESHOLD


-@pytest.mark.slow
+@pytest.mark.full_model
@pytest.mark.benchmark
@pytest.mark.diffusion
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@@ -544,7 +544,7 @@ def _audio_metrics(reference: np.ndarray, prediction: np.ndarray) -> dict[str, f


@pytest.mark.parametrize("case", CASES, ids=lambda case: case.name)
-@pytest.mark.slow
+@pytest.mark.full_model
@pytest.mark.benchmark
@pytest.mark.diffusion
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@@ -17,7 +17,7 @@ from tests.helpers.mark import hardware_test
 from tests.helpers.media import get_asset_path
 from tests.helpers.runtime import OmniServer

-pytestmark = [pytest.mark.diffusion]
+pytestmark = [pytest.mark.full_model, pytest.mark.diffusion]


 SINGLE_MODEL = "Qwen/Qwen-Image-Edit"
@@ -253,7 +253,6 @@ def _diffusers_output_single_image(accuracy_artifact_root: Path, qwen_bear_image
    )


-@pytest.mark.slow
@pytest.mark.benchmark
@hardware_test(res={"cuda": "H100"}, num_cards=1)
 def test_qwen_image_edit_single_matches_diffusers(
@@ -279,7 +278,6 @@ def test_qwen_image_edit_single_matches_diffusers(
    )


-@pytest.mark.slow
@pytest.mark.benchmark
@hardware_test(res={"cuda": "H100"}, num_cards=1)
 def test_qwen_image_edit_2511_matches_diffusers_pixelwise(accuracy_artifact_root: Path) -> None:
@@ -17,7 +17,7 @@ from tests.helpers.env import run_post_test_cleanup, run_pre_test_cleanup
 from tests.helpers.mark import hardware_test
 from tests.helpers.runtime import OmniServer

-pytestmark = [pytest.mark.slow, pytest.mark.diffusion]
+pytestmark = [pytest.mark.full_model, pytest.mark.diffusion]


 MODEL_ID = "Qwen/Qwen-Image-Layered"
@@ -11,7 +11,7 @@ from tests.e2e.accuracy.helpers import assert_video_similarity_metrics, probe_vi
 from tests.helpers.mark import hardware_marks
 from tests.helpers.runtime import DiffusionResponse, OmniServer, OpenAIClientHandler

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 MODEL = "BestWishYsh/Helios-Distilled"
 PROMPT = "A serene lakeside sunrise with mist over the water."
@@ -138,7 +138,7 @@ def _make_file_lora_request(adapter_dir: Path) -> LoRARequest:


 pytestmark = [
-    pytest.mark.slow,
+    pytest.mark.full_model,
    pytest.mark.diffusion,
    pytest.mark.parametrize("omni_runner", [_OMNI_RUNNER_PARAM], indirect=True),
 ]
@@ -109,7 +109,7 @@ def _diffusion_sampling_params(


 pytestmark = [
-    pytest.mark.slow,
+    pytest.mark.full_model,
    pytest.mark.diffusion,
 ]

@@ -96,7 +96,7 @@ def _get_test_config() -> str:

 pytestmark = [
    pytest.mark.skip(reason="https://github.com/vllm-project/vllm-omni/issues/4700"),
-    pytest.mark.slow,
+    pytest.mark.full_model,
    pytest.mark.tts,
    pytest.mark.parametrize(
        "omni_runner",
@@ -271,6 +271,7 @@ def _collect_audio(omni_runner: OmniRunner, request: dict) -> tuple[torch.Tensor
 # ---------------------------------------------------------------------------


+@pytest.mark.advanced_model
@hardware_test(res={"cuda": "L4"}, num_cards=1)
 def test_moss_tts_realtime_english(omni_runner: OmniRunner, ref_audio_path: str) -> None:
    """MossTTSRealtime: English voice_clone produces non-empty 24 kHz audio."""
@@ -282,6 +283,7 @@ def test_moss_tts_realtime_english(omni_runner: OmniRunner, ref_audio_path: str)
    assert not torch.all(audio == 0), "Audio is silence"


+@pytest.mark.advanced_model
@hardware_test(res={"cuda": "L4"}, num_cards=1)
 def test_moss_tts_realtime_chinese(omni_runner: OmniRunner, ref_audio_path: str) -> None:
    """MossTTSRealtime: Chinese input produces non-empty audio."""
@@ -22,7 +22,7 @@ from tests.helpers.mark import hardware_marks
 from tests.helpers.runtime import OmniServer, OmniServerParams, OpenAIClientHandler, dummy_messages_from_mix_data
 from tests.helpers.stage_config import get_deploy_config_path, modify_stage_config

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 PROMPT = "A futuristic city skyline at twilight, cyberpunk style, ultra-detailed, high resolution."
 NEGATIVE_PROMPT = "low quality, blurry, distorted, deformed, watermark"
@@ -1,8 +1,6 @@
-"""Weekly e2e coverage for BAGEL diffusion multi-replica serving.
+"""Nightly-only e2e coverage for BAGEL diffusion multi-replica serving.

 This test needs 4 H100 GPUs; keep it out of generic test_bagel_* jobs.
-It is covered by the weekly Diffusion · H100 · Multi-GPU lane
-(``.buildkite/cuda/test-weekly.yml``, ``mirror_hardwares: h100_4``).
 """

 import os
@@ -19,7 +17,7 @@ MODEL = "ByteDance-Seed/BAGEL-7B-MoT"
 MULTI_REPLICA_DEPLOY = get_deploy_config_path("ci/bagel_multi_replicas_4gpu.yaml")
 ROUTE_STRESS_REQUESTS = 6

-pytestmark = [pytest.mark.slow, pytest.mark.diffusion]
+pytestmark = [pytest.mark.full_model, pytest.mark.diffusion]

 test_params = [
    OmniServerParams(
@@ -16,7 +16,7 @@ parity vs. the Diffusers baseline is covered separately (see the support plan, s
 From ``tests/``::

    pytest -s -v e2e/online_serving/test_boogu_image.py \
-        -m "slow and diffusion and H100"
+        -m "full_model and diffusion and H100" --run-level=full_model
 """

 import os
@@ -28,7 +28,7 @@ from tests.helpers.runtime import OmniServer, OmniServerParams, OpenAIClientHand

 os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 MODEL = "Boogu/Boogu-Image-0.1-Base"
 T2I_PROMPT = "A mountain lake at sunset, photorealistic, cinematic lighting"
@@ -19,7 +19,7 @@ reference image is supported.
 From ``tests/``::

    pytest -s -v e2e/online_serving/test_boogu_image_edit.py \
-        -m "slow and diffusion and H100"
+        -m "full_model and diffusion and H100" --run-level=full_model
 """

 import base64
@@ -35,7 +35,7 @@ from tests.helpers.runtime import (
    dummy_messages_from_mix_data,
 )

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 EDIT_PROMPT = "Change the style to a colored pencil drawing."
 SINGLE_CARD_FEATURE_MARKS = hardware_marks(res={"cuda": "H100"})
@@ -25,7 +25,7 @@ model and are intentionally omitted.
 From ``tests/``::

    pytest -s -v e2e/online_serving/test_boogu_image_edit_expansion.py \
-        -m "slow and diffusion and H100"
+        -m "full_model and diffusion and H100" --run-level=full_model
 """

 import os
@@ -38,7 +38,7 @@ from tests.helpers.runtime import OmniServer, OmniServerParams, OpenAIClientHand

 os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 MODEL = "Boogu/Boogu-Image-0.1-Edit"
 EDIT_PROMPT = "Change the style to a colored pencil drawing."
@@ -30,7 +30,7 @@ omitted (add rows here when those features land).
 From ``tests/``::

    pytest -s -v e2e/online_serving/test_boogu_image_expansion.py \
-        -m "slow and diffusion and H100"
+        -m "full_model and diffusion and H100" --run-level=full_model
 """

 import os
@@ -42,7 +42,7 @@ from tests.helpers.runtime import OmniServer, OmniServerParams, OpenAIClientHand

 os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 MODEL = "Boogu/Boogu-Image-0.1-Base"
 T2I_PROMPT = "A mountain lake at sunset, photorealistic, cinematic lighting"
@@ -30,7 +30,7 @@ from tests.helpers.runtime import (
 )
 from tests.helpers.stage_config import get_deploy_config_path, modify_stage_config

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 MODEL = os.environ.get("GLM_IMAGE_MODEL_PATH", "zai-org/GLM-Image")

@@ -63,7 +63,7 @@ def test_text_to_image_moe(omni_server: OmniServer, openai_client: OpenAIClientH
    )


-@pytest.mark.slow
+@pytest.mark.full_model
@pytest.mark.diffusion
@pytest.mark.parametrize("omni_server", _get_server_cases(MODEL), indirect=True)
 def test_video_generation_modes_moe(omni_server: OmniServer, openai_client: OpenAIClientHandler) -> None:
@@ -10,7 +10,7 @@ from tests.helpers.mark import hardware_marks
 from tests.helpers.media import generate_synthetic_image
 from tests.helpers.runtime import OmniServer, OmniServerParams, OpenAIClientHandler

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]

 DISTILLED_MODEL = os.getenv("VLLM_TEST_LTX23_DISTILLED_MODEL", "diffusers/LTX-2.3-Distilled-Diffusers")
 BASE_MODEL = os.getenv("VLLM_TEST_LTX23_MODEL", "diffusers/LTX-2.3-Diffusers")
@@ -20,7 +20,7 @@ MODEL = os.getenv("VLLM_TEST_LTX25_MODEL", DEFAULT_MODEL)
 MODEL_REVISION = os.getenv("VLLM_TEST_LTX25_MODEL_REVISION", DEFAULT_REVISION if MODEL == DEFAULT_MODEL else "")
 PROMPT = "A red fox walks through a snowy forest while the camera remains fixed."

-pytestmark = [pytest.mark.diffusion, pytest.mark.slow]
+pytestmark = [pytest.mark.diffusion, pytest.mark.full_model]
 SINGLE_CARD_MARKS = hardware_marks(res={"cuda": "H100"})


@@ -20,7 +20,7 @@ from tests.helpers.stage_config import get_deploy_config_path

 os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"

-pytestmark = [pytest.mark.omni, pytest.mark.slow]
+pytestmark = [pytest.mark.omni, pytest.mark.full_model]

 _SKIP_NEED_4_H100_NOT_CI = pytest.mark.skip(
    reason="Requires 4x H100 GPUs; skipped in CI for now.",
@@ -1,112 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-"""E2E online tests for MiniMax Music 3 text-to-music.
-
-This is a music model on the speech endpoint, so the request shape differs
-from every other TTS model here: ``input`` carries the lyrics, ``instructions``
-carries the caption that decides genre and arrangement, and both are required.
-
-Weekly rather than per-PR: the checkpoint is 27 GB across five components and a
-single request occupies two decode rows, because classifier-free guidance is
-mandatory for this model. Rejection cases live in
-``tests/dfx/reliability/invalid_param_test/test_invalid_audio_speech.py``.
-"""
-
-import os
-
-os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"
-
-import pytest
-
-from tests.helpers.mark import hardware_test
-from tests.helpers.runtime import OmniServerParams
-from tests.helpers.stage_config import get_deploy_config_path
-
-MODEL = "MiniMaxAI/MiniMax-Music3"
-DEFAULT_AUDIO_SPEECH_TIMEOUT_S = 600.0
-
-# 25 frames per second, so 250 frames is ten seconds and 750 is thirty. The
-# byte floors sit below the true payload (32 kHz stereo, 16-bit) to leave room
-# for the model ending a song before the cap.
-_FRAMES_10S = 250
-_FRAMES_30S = 750
-_MIN_BYTES_10S = 500_000
-_MIN_BYTES_30S = 1_500_000
-
-LYRICS = (
-    "[Verse]\nWalking down the empty street at midnight\n"
-    "Streetlights flicker like a broken dream\n"
-    "[Chorus]\nAnd I keep on walking\nTill the morning finds me"
-)
-
-CAPTION = (
-    "A melancholic lo-fi hip-hop track at 85 BPM in F minor: mellow Rhodes piano riff, "
-    "soft vinyl crackle, dusty boom-bap drums with a laid-back swing, warm upright bass. "
-    "Intimate bedroom production, gentle tape saturation, no bright cymbals."
-)
-
-
-tts_server_params = [
-    pytest.param(
-        OmniServerParams(
-            model=MODEL,
-            stage_config_path=get_deploy_config_path("minimax_music3.yaml"),
-            server_args=["--trust-remote-code", "--disable-log-stats"],
-        ),
-        id="minimax_music3",
-    )
-]
-
-
-@pytest.mark.slow
-@pytest.mark.tts
-@hardware_test(res={"cuda": "H100"}, num_cards={"cuda": 1})
-@pytest.mark.parametrize("omni_server", tts_server_params, indirect=True)
-def test_text_to_music_001(omni_server, openai_client) -> None:
-    """Baseline: lyrics plus caption in, one ten-second song out.
-
-    Deploy Setting: minimax_music3.yaml
-    Input Modal: text (lyrics) + text (caption)
-    Output Modal: audio, 32 kHz stereo
-    Input Setting: stream=False, seed pinned, 250 frames
-    """
-    request_config = {
-        "model": omni_server.model,
-        "input": LYRICS,
-        "instructions": CAPTION,
-        "seed": 1,
-        "max_new_tokens": _FRAMES_10S,
-        "stream": False,
-        "response_format": "wav",
-        "timeout": DEFAULT_AUDIO_SPEECH_TIMEOUT_S,
-        "min_audio_bytes": _MIN_BYTES_10S,
-    }
-    openai_client.send_audio_speech_request(request_config)
-
-
-@pytest.mark.slow
-@pytest.mark.tts
-@hardware_test(res={"cuda": "H100"}, num_cards={"cuda": 1})
-@pytest.mark.parametrize("omni_server", tts_server_params, indirect=True)
-def test_text_to_music_multi_window_002(omni_server, openai_client) -> None:
-    """A song long enough to span several acoustic windows.
-
-    The acoustic stage decodes overlapping 200-frame windows on a 100-frame
-    hop and crops the overlap back off, so a request that fits in one window
-    never exercises the stitching. 750 frames covers seven of them.
-
-    Deploy Setting: minimax_music3.yaml
-    Input Setting: stream=False, seed pinned, 750 frames
-    """
-    request_config = {
-        "model": omni_server.model,
-        "input": LYRICS,
-        "instructions": CAPTION,
-        "seed": 1,
-        "max_new_tokens": _FRAMES_30S,
-        "stream": False,
-        "response_format": "wav",
-        "timeout": DEFAULT_AUDIO_SPEECH_TIMEOUT_S,
-        "min_audio_bytes": _MIN_BYTES_30S,
-    }
-    openai_client.send_audio_speech_request(request_config)
Author	SHA1	Message	Date
Hongsheng Liu	f1abb13958	[Diffusion] Register runtime cache for direct DLO H2D Avoid recurrent mmap-to-pinned packing in no-AllGather DLO by registering transform-complete runtime-cache mappings under an explicit per-worker budget. Copy mapped tensors directly into the rotating device buffers while preserving bounded-staging fallback and ordered shutdown cleanup. Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>	2026-08-17 11:50:32 +08:00
Hongsheng Liu	b732ad2cea	[Diffusion] Document runtime cache hardware tradeoffs Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>	2026-08-17 00:43:06 +08:00
Hongsheng Liu	99347e1ba7	[Diffusion] Tighten runtime cache compatibility guards Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>	2026-08-16 23:26:25 +08:00
Hongsheng Liu	db86a6a078	[Diffusion] Add node-local runtime weight cache for DLO Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>	2026-08-16 23:15:48 +08:00