anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity
2,955 Commits 235 Branches 31 Tags
b36a3431bda13b53dfe74d9d2e5617ec9f86e5b7
Commit Graph
794 Commits
Author SHA1 Message Date
tzhouamandClaude Opus 5 b36a3431bd Package this fork as vllm-embedding: README, docs, roadmap, harness 
Turns a branch of accumulated edge/CPU work into something someone else can
read, reproduce and continue.

README.md states what the fork is for (multimodal models on edge CPUs rather
than servers), the headline memory and speed results with their costs, and why
the numbers are worth reading -- five retracted conclusions, all measurement
errors rather than coding errors. Upstream's README is preserved at
docs/README.vllm-omni.md and NOTICE records the Apache-2.0 derivation and the
files this fork changes.

docs/edge/:
  methodology.md  the measurement discipline, which is the most transferable
                  thing here: what each of the five retractions was, and the
                  rule it left behind. Includes the two traps specific to this
                  stack -- vLLM's AOT compile cache keying on config so two
                  arms sharing an artifact load the wrong graph, and allocated
                  not being resident until mincore(2) says so.
  results.md      every measured figure with its spread and baseline, and an
                  explicit note that the four memory levers overlap and must
                  not be added.
  cpu-4bit.md     the two 4-bit paths, the packed layout as measured rather
                  than assumed, why the path choice is a config field, and the
                  GPTQ output-head trap that invalidated a whole round of W4A8
                  numbers.
  harness.md      how to run a comparison, and the two arguments that matter
                  more than they look (--context-len, --prompt-len), each of
                  which has produced a wrong conclusion by being left at its
                  default.

ROADMAP.md orders by whether the verification method is correct rather than by
expected size, lists what is done with results, what is next with expected
gains and how each would be verified, and -- at equal length -- what was
considered and rejected, including the CPU megakernel with the measurements
that rule it out for now.

benchmarks/edge_harness/ brings the tooling into the repo so the numbers are
reproducible: the grid runner, the single-run benchmarks, the memory ledger,
the fidelity metric, the Android agent suite, the quantizers, and the tests
that pin the harness's judgement. 82 tests pass from the new location.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L1ax5wYLA6XbqEL1i8L9gZ
 2026-09-14 13:25:32 +08:00
tzhouamandClaude Opus 5 c4e61133fb [Edge][CPU] Fused norms off by default, and register the switches this branch added 
Two corrections and the paperwork they imply.

The fused-RMSNorm registration defaulted to *on*, which contradicts the
measurement it was committed with: those kernels are 3.6x and 4.5x faster in
isolation but measured 72.7 tok/s against 74.0 end to end, because inductor
already fuses the native norm into the residual add and an opaque call breaks
that fusion. The implementations stay registered, so the provider remains
selectable through `kernel_config.ir_op_priority` on a machine where the trade
lands differently; only the default ranking is left alone now, and the test
asserts both states rather than just the opt-in one.

The four switches this branch added (VLLM_OMNI_AOT_CACHE_SHIM,
_CPU_FUSED_NORMS, _CPU_RELEASE_MEMORY, _SPARK_GATE_REDUCTION) were not in the
reviewed environment-variable inventory, so
`test_statically_resolvable_environment_accesses_are_classified` failed. They
are classified now, the reviewed count moves 28 -> 32 with a note saying why,
and each has a row in docs/configuration/environment_variables.md carrying the
measurement that justifies its default.

`VLLM_OMNI_CPU_FAST_SAMPLER` and `VLLM_OMNI_STEP_STATS_SYNC` remain
unclassified and that test still fails on them. They arrived in 6ce8de1b, in
files this branch does not touch, so they are left to whoever owns that work
rather than documented from a guess.

Under `-m "core_model and cpu"`: 2593 passed before these fixes with 9
failures, of which two were mine. The rest are CUDA-only tests in a CPU venv
("Torch not compiled with CUDA enabled") plus test_qnn_export.py, which
another session is actively editing.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-09-12 14:55:24 +08:00
tzhouamandClaude Fable 5.1 536866b916 [Edge] Classify and document the edge env vars; fix the client decoder call 
- environment_variable_inventory: VLLM_OMNI_CPU_INDUCTOR / HW_PROFILE /
  INIT_TIMELINE / STEP_STATS_DIR (public), STAGE_ID / INIT_TIMELINE_DIR
  (internal); reference page rows added.
- edge_decoder_client.LocalDecoder: batched_chunked_decode takes a [B, Q, F]
  tensor, not a list (the WP5 e2e run failed with a TypeError); the decoder
  now also carries the ramp schedule attributes the Code2Wav stage sets.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 15:22:52 +08:00
tzhouamandClaude Fable 5.1 fa8b1a878e [Edge] Docs: chunk ramp / edge profile section and an edge recipe for Qwen3-TTS 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 15:03:25 +08:00
tzhouamandClaude Fable 5.1 49a839ecf9 [Edge][WP5] Add codec-token streaming mode with on-device decoding 
* qwen3_tts_talker_only pipeline (single terminal LLM_AR stage,
  final_output_type "latent") + vllm_omni/deploy/qwen3_tts_talker_only.yaml;
  a terminal talker publishes its per-step codes.audio rows to the client
  (omni_client_multimodal_output_keys).
* /v1/audio/speech/stream: session.config.output_mode="codec_tokens" streams
  codec.start (JSON) + binary frames (<IHB seq/n_frames/flags + int16 rows,
  vllm_omni/entrypoints/openai/codec_stream.py) + codec.done; new input.cancel
  message aborts an in-flight generation for both codec and PCM sessions
  (single reader task + session inbox replaces direct socket reads); optional
  session.config.seed; audio.done gains "cancelled" only when it happened.
* serving_speech._generate_codec_chunks aligns the cumulative code tensor to
  output token ids (drops prefill placeholder/EOS rows) and emits deltas.
* TTS adapters: supports_codec_stream / codec_stream_spec (Qwen3-TTS: 16
  codebooks, 2048 ids, 12.5 Hz, 24 kHz, decoder qwen3_tts_tokenizer_v2_12hz).
* examples/online_serving/text_to_speech/qwen3_tts/edge_decoder_client.py
  decodes locally with Qwen3TTSTokenizerV2Decoder (ramp schedule, 72-frame
  context, cancel/reset), codec_parity_check.py and
  benchmarks/tts/codec_stream_e2e.py compare against the two-stage server.
* docs/serving/codec_token_streaming.md; CPU unit tests for the wire format,
  alignment, session config and handler (frames, unsupported adapter, cancel);
  L3 e2e tests/e2e/online_serving/test_qwen3_tts_codec_stream.py.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 11:10:13 +08:00
tzhouamandClaude Fable 5.1 35e43b9983 [Edge][WP3] Add CPU OmniPlatform (AR + generation stages on vLLM's CPU backend) 
* vllm_omni/platforms/cpu/platform.py: CPUOmniPlatform(OmniPlatform, CpuPlatform)
  selected when VLLM_TARGET_DEVICE=cpu or a +cpu vLLM wheel is installed
  (builtin plugin "cpu"; explicit CPU target short-circuits accelerator probes
  like vLLM does). NUMA-based memory queries, one logical device, no graph
  wrapper (sentinel type for isinstance checks), inductor opt-in via
  VLLM_OMNI_CPU_INDUCTOR=1, diffusion stages unsupported.
* vllm_omni/platforms/cpu/worker: CPUARWorker / CPUGenerationWorker compose
  vLLM's CPUWorker with the Omni mixin (XPU pattern); CPUARModelRunner /
  CPUGenerationModelRunner compose the Omni runners with CPUModelRunner
  (no graphs, no async omni output; sleep/wake return error ACKs).
* patch.py skips the CuMem patch on CPU-only torch; Qwen3-TTS speaker encoder
  and prompt-embeds dtype follow the model dtype (float32/float16 on CPU).
* vllm_omni/deploy/qwen3_tts.yaml: `platforms: cpu:` overlay (eager, no async
  scheduling, absolute kv_cache_memory_bytes, disjoint OMP thread ranges).
* requirements/cpu.txt (+psutil, intel-openmp), docs/getting_started/installation/cpu.md.

Measured (this x86 Xeon 8480C host, vllm 0.28.0+cpu, bf16, 12+4 threads):
Qwen3-TTS 0.6B CustomVoice end-to-end on CPU, 12 prompts: TTFA p50 292 ms,
RTF ~1.7 (25-frame chunk ~4.6 s), init 185 s. Result JSON + init timeline
committed under benchmarks/tts/edge_results/.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 11:02:14 +08:00
Zeyu Huang | 黃澤宇andAlicia 4b96c57730 [skip ci][Doc] Module design doc for diffusion runtime (#6440) 
Signed-off-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com>
Co-authored-by: Alicia <115451386+congw729@users.noreply.github.com>
 2026-09-07 21:52:24 +08:00
Mu GuanLin 3204a0b2ad [Model][Perf] Optimize LTX Ulysses sequence parallelism (#7079) 
Signed-off-by: mglyn <1203789601@qq.com>
 2026-09-07 10:48:26 +00:00
FredHuangNV 85b0bad2d3 [Frontend][Model] Add Cosmos3 control uploads to video API (#7027) 
Signed-off-by: Fred Huang <frhuang@nvidia.com>
 2026-09-07 01:38:56 -07:00
WeiQing Chen b78c31eda6 [Doc] Update vLLM-Omni WeChat QR code (#7178) 
Signed-off-by: David Chen <530634352@qq.com>
 2026-09-07 06:28:30 +00:00
BruceLoveDecimal 5cd925631c [Feature][LingBot World]stepwise execution (#6844) 
Signed-off-by: BruceLoveDecimal <liuqihao970610@gmail.com>
Signed-off-by: liuqihao <liuqihao970610@gmail.com>
 2026-09-07 13:38:23 +08:00
Zeyu Huang | 黃澤宇 2df95de421 [Core][Diffusion] Refactor prompt updates into a modality-ready interaction framework (#6294) 
Signed-off-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com>
 2026-09-07 11:40:04 +08:00
Xiaobaixia 615f97b9c7 [Feature] Define the runner preprocess phase contract and regression coverage (#7136) 
Signed-off-by: xiaobaixia222 <289982752@qq.com>
 2026-09-06 17:20:04 +00:00
汪志鹏andClaude Opus 5 af74a5a155 [Feature][MiniMax-H3] Support the full LightX2V Turbo LoRA matrix (#7062) 
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
 2026-09-06 20:23:36 +08:00
DMAN 06541d1f52 [Model] FP8 Support For Boogu-Image (#6925) 
Signed-off-by: Dmaner <2663515256@qq.com>
 2026-09-06 04:48:38 +00:00
Max Freedom Pollard 946b843e61 [Doc] Fix markdownlint findings in the serving API reference (#7111) 
Signed-off-by: Max Freedom Pollard <272618364+MaxFreedomPollard@users.noreply.github.com>
 2026-09-06 08:40:10 +08:00
SYLAR ea3cf99894 [diffusion][feature] Add component-selective offload policies (#5929) 
Signed-off-by: lishunyang12 <lishunyang12@163.com>
 2026-09-06 05:40:09 +08:00
chickeytonClaude Fable 5Gao Han
 		5d7fdf9350 [Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196) 
Signed-off-by: chickeyton <ngton2014@gmail.com>
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-09-05 08:20:21 +00:00
Hongsheng Liu d4096e01fe [Model] Add native MAGI-2 Preview diffusion support (#5918) 
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-09-05 15:32:30 +08:00
vOv 13c617a140 [Model] Add sequence parallelism to SANA-Video 2B (#5940) 
Signed-off-by: cr-gao <gaochenrui@sjtu.edu.cn>
 2026-09-05 15:22:53 +08:00
boatman 348fd0fc07 [Feat][dots.tts] Add Online Serving Support (#6235) 
Signed-off-by: boatman <109857087+sphinxkkkbc@users.noreply.github.com>
 2026-09-04 14:44:10 -07:00
TriggerandGao Han ea14b04e9f [Bugfix][BAGEL] Restore original denoise step count (#7049) 
Signed-off-by: Trigger <Meng.Bo.Wang@outlook.com>
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-09-04 17:05:05 +00:00
SYLARandhsliuustc0106 22cefe04b4 [Bugfix] Update Hub kernel dependency floor (#7020) 
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
Co-authored-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-09-04 22:15:55 +08:00
SYLARandhsliuustc0106 7493cbd3c2 [Bugfix] Fix Qwen3-Omni MoE backend selection (#7019) 
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
Co-authored-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-09-04 21:16:58 +08:00
NATURERuirui Yang | Reinamy-why-3459
 		8d98cc0cb7 [Benchmark] Add local OmniInteract performance cases (#6817) 
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: Ruirui Yang | Rein <73573651+R2-Y@users.noreply.github.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-09-04 10:36:02 +08:00
Trigger e43c787544 [Model] Add BAGEL step execution and continuous batching (#6359) 2026-09-04 01:36:11 +00:00
hurukawa 8f284b343a [Feature][Diffusion]: CFG parallelism support for Boogu-Image (#6786) 
Signed-off-by: nagisa <1434936049@qq.com>
Signed-off-by: nagisa-kun <1434936049@qq.com>
 2026-09-03 10:37:21 +08:00
GXINandHongsheng Liu 3ab582c993 [Bugfix][Qwen3-TTS] Reject task mismatches before EngineCore dispatch (#6113) 
Signed-off-by: XIN GAO <1037396230@qq.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-09-02 11:16:55 -07:00
wangyulishunyang12CursorBo Li
 		95a9311696 [CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543) 
Signed-off-by: wangyu <410167048@qq.com>
Signed-off-by: lishunyang12 <lishunyang12@163.com>
Signed-off-by: [Your Name] <your.email@example.com>
Signed-off-by: wangyu <53896905+yenuo26@users.noreply.github.com>
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
Co-authored-by: lishunyang12 <lishunyang12@163.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Bo Li <22713281+bobboli@users.noreply.github.com>
 2026-09-02 15:39:49 +00:00
Bo Li 5d20f6b900 [Doc] Refine TRTLLM attention guidance (#6724) 
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
 2026-09-02 19:42:37 +08:00
汪志鹏 c59b48b47e [Feature][MiniMax-H3] Complete VSA and Ulysses support (#6909) 
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
 2026-09-02 07:19:21 -04:00
vOv 57c08d34c3 [Model] Add TP and CFG parallelism to SANA-Video 2B (#5861) 
Signed-off-by: cr-gao <gaochenrui@sjtu.edu.cn>
 2026-09-02 17:44:41 +08:00
wangyuandamy-why-3459 b6e72f3879 [CI] Migrate L4 jobs to l4-k8s and split by GPU count (#6890) 
Signed-off-by: wangyu <410167048@qq.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-09-02 17:09:59 +08:00
DanaerLeeandHongsheng Liu 0288c3f56e [Model] Support SenseNova-U1.5-8B-MoT and its distilled 8-step LoRA (#6516) 
Signed-off-by: MrlixiangWE <mrdanaer@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-09-01 22:53:53 +08:00
Hongsheng Liuandhsliu_ustc 44c8d1c449 docs: add concise API server endpoint guide (#6114) 
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: hsliu_ustc <hsliu_ustc@noreply.gitcode.com>
 2026-09-01 20:34:09 +08:00
Hongsheng Liu d1c8beeb18 [Doc] Update README and docs for v0.28.0 (#6858) 
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-09-01 19:50:26 +08:00
AuFlowandHongsheng Liu 49c6fc1540 [Frontend] Add configurable output sample rate to Speech API (#6553) 
Signed-off-by: 郑倩倩 <2458983186@qq.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-09-01 17:25:09 +08:00
XRay2016 8e26dce1f4 [Diffusion] Add Boogu Edit-Turbo support docs and tests (#6701) 
Signed-off-by: xRay2016 <1150722393@qq.com>
 2026-09-01 08:41:23 +08:00
WeiQing Chen 4cca28e9bd [Doc] Update vLLM-Omni WeChat QR code (#6832) 
Signed-off-by: David Chen <530634352@qq.com>
 2026-08-31 09:44:33 +08:00
wangyu e338f0ece3 [CI]Split GPU jobs by cards_* and reject hand-written SKU marks (#6650) 
Signed-off-by: wangyu <410167048@qq.com>
Signed-off-by: [Your Name] <your.email@example.com>
 2026-08-29 18:42:33 +08:00
HAAZZZEEEE c50b7de167 [NPU][Diffusion] MiniMax-H3 RainFusion end_step tail fallback (#6037) 
Signed-off-by: HAAZZZEEEE <wang-taicheng@qq.com>
 2026-08-29 18:36:05 +08:00
ooooooye 5a5bc55c25 [Feature][Diffusion] Allow online INT8 quantization with DLO AllGather (#6573) 
Signed-off-by: brandneway <gyuan4892@gmail.com>
 2026-08-29 09:50:30 +00:00
Nguyen Kha Nhat LongCursorSamit
 		bb5bf8fd0a [Bugfix] Fix colocate-async sleep admission race and deliver AR abort tokens (#6367) 
Signed-off-by: knlnguyen1802 <knlnguyen1802@gmail.com>
Signed-off-by: Cursor <cursoragent@cursor.com>
Signed-off-by: Samit <285365963@qq.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Samit <285365963@qq.com>
 2026-08-29 09:11:39 +00:00
Zijun HuangandWeiQing Chen 08cdee8dc4 [Feature] Support MiniMax-H3 FlashGen native LoRA with the legacy manager (#6666) 
Signed-off-by: Zijun Huang <zijunhua@usc.edu>
Co-authored-by: WeiQing Chen <40507679+david6666666@users.noreply.github.com>
 2026-08-29 06:52:19 +00:00
Chendi.Xue 034af1f657 [CI][XPU]switch to use vLLM base docker for XPU (#6727) 
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
 2026-08-29 09:03:00 +08:00
Hongsheng Liu 6ec5ab11bd [Bugfix] Make HWR filesystem lifecycle transitions fail-closed (#6692) 
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-08-28 09:30:17 +08:00
Hongsheng Liuandhsliuustc0106 4245fd6955 [Fix] Restore MammothModa2 support (#6694) 
Signed-off-by: hsliuustc0106 <hsliuustc0106@users.noreply.github.com>
Co-authored-by: hsliuustc0106 <hsliuustc0106@users.noreply.github.com>
 2026-08-28 08:32:32 +08:00
Zheng Wengangandamy-why-3459 94b1546ce3 [Model] Qwen3-Omni: register thinker-only pipeline for Instruct serve (#6284) 
Signed-off-by: ZhengWG <zwg0606@gmail.com>
Signed-off-by: Zheng Wengang <zwg0606@gmail.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-08-27 08:21:03 +00:00
Zhou TaichangandClaude Opus 5 651b03225b [Rebase] Rebase to vllm 0.28.0 (#6606) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-27 06:12:55 +00:00
Hongsheng Liuandhsliuustc0106 da4a08b65b [Diffusion] Add FLUX.2-klein BF16 HWR contract (#6651) 
Signed-off-by: hsliuustc0106 <hsliuustc0106@users.noreply.github.com>
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: hsliuustc0106 <hsliuustc0106@users.noreply.github.com>
 2026-08-27 12:51:15 +08:00