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
93 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 Fable 5.1 f9d2cab3bc edge results: codec stream per-step cadence (chain7): client TTFA 152-156 ms, ramp 2/4/8/16/25, stall 0 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:54:03 +08:00
tzhouamandClaude Fable 5.1 565f7475c7 edge results: codec stream cadence check 2 (chain6, server per-step outputs confirmed) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:45:25 +08:00
tzhouamandClaude Fable 5.1 78a30b7562 edge results: codec stream diagnosis run (chain4) + offline parity diag 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:22:01 +08:00
tzhouamandClaude Fable 5.1 0356b22e4c edge results: chunk hop put->get stats, shm vs cuda_ipc (0.6B, chain3) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:08:22 +08:00
tzhouamandClaude Fable 5.1 5104fb6d67 edge results: edge_final2 (parallel init), GPU step-stats (default/edge/cuda_ipc), codec stream parity run 1 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:02:17 +08:00
tzhouamandClaude Fable 5.1 b8d1e44ebe [Edge][WP5] Codec stream diagnostics: per-request output/codes counters, client saves received codes, offline parity diag 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 16:02:02 +08:00
tzhouamandClaude Fable 5.1 15bb6133a3 edge results: init_profile variant YAMLs; recipe GPU memory figure 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 15:35:24 +08:00
tzhouamandClaude Fable 5.1 a786a88919 edge results: cold-cache init profile, GPU hardware emulation, CPU thread sweep, edge_final benches 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 15:23:08 +08:00
tzhouamandClaude Fable 5.1 580a201631 [Edge][WP4] Windowed Code2Wav decoder export (torch.export / ExecuTorch) + client backend; results summarizer 
vllm_omni/edge/decoder_export.py exports a stateless windowed decoder (context
72 frames + chunk, one program per ramp chunk size) via torch.export (.pt2,
verified here on CPU for chunks 2 and 25 with the real 0.6B decoder: exact
match vs eager, 460 MB fp32 per program) or ExecuTorch XNNPACK (.pte; written
against the executorch 1.x API, not executed: package not installed, install
is permission gated; QNN partitioner export-only/unverified). The WP5 client
gains --decoder-backend executorch --pte-manifest for phone/NPU targets.
benchmarks/tts/edge_results/summarize.py renders all committed JSON as tables.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 15:02:19 +08:00
tzhouamandClaude Fable 5.1 37c0ea9a6b [Edge][WP8] Per-step overhead attribution, edge scheduling helpers, CPU thread sweep 
- vllm_omni/edge/step_stats.py: process-local counters enabled by
  VLLM_OMNI_STEP_STATS_DIR (scheduler schedule/update_from_output, engine-core
  step, chunk hop enqueue->put / put->get / get, orchestrator dispatch), dumped
  per process and merged into an attribution table (share of the 80 ms frame).
  Hooks: stage_engine_core_proc (instrument after construction, dump at exit),
  chunk_transfer_adapter (put/get timing, put->get via the shm lock-file mtime),
  orchestrator dispatch loop. Bench: stream_latency_bench.py --step-stats.
- vllm_omni/edge/scheduling.py: playback slack, prefill-throttle decision and
  the no-preemption KV admission rule (exact bytes/token from the talker KV
  geometry); adapt.derive_overrides now sizes max_num_seqs with it and
  auto_profile passes the geometry read from the model config.
- Engine re-reads parallel_stage_init after apply_deploy_profile so the edge
  profile's orchestrator default takes effect (was read before the profile).
- benchmarks/tts/cpu_thread_sweep.py: CPU-only 8/16/32-thread sweep (+inductor)
  with per-step attribution.
- Tests: tests/core/test_edge_scheduling.py, tests/edge/test_step_stats.py,
  tests/benchmarks/test_cpu_thread_sweep.py, profile default test.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 14:55:55 +08:00
tzhouamandClaude Fable 5.1 602e39647e [Edge][WP2] Finalize edge profile init knobs from init_profile; WP4 CPU emulation results 
init_profile 20260909-112200 (1.7B, L20X, warm): default 152.0 s, eager_stage1
149.9 s, eager_both 110.9 s (first TTFA 15 s, no graphs), parallel_stage_init
94.6 s, edge 145.7 s. Edge profile now carries `orchestrator: parallel_stage_init:
true`, applied by the deploy_profile selector (explicit values win); graphs stay on.

hw_emulation 20260909-110229 (CPU venv, 0.6B, deploy_profile=auto):
x86_cpu_avx2 (4c, fp32) TTFA 2.65 s RTF 11.9; x86_cpu_avx512 (8c) TTFA 0.68 s
RTF 2.45; x86_cpu_amx (16c) TTFA 0.51 s RTF 1.72; arm64 logic-only same as avx2.
Emulation rows now record the auto-detected hardware class.

Adds tests/e2e/offline_inference/test_qwen3_tts_edge_profile.py (L3, one GPU).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 14:40:13 +08:00
tzhouamandClaude Fable 5.1 285db2fa3c edge results: WP1 edge-profile stream latency (0.6B/1.7B) + init timelines 
1.7B: TTFA p50 43.9 ms, playback_start 43.9 ms, stall 0 (baseline 32.0 / 116.2 / 84.7).
0.6B: TTFA p50 46.1 ms, playback_start 46.1 ms, stall 0 (baseline 27.8 / 95.4 / 67.3).
RTF unchanged (~0.1). Measured on L20X via gpu run --gpus 1 (canhazgpu).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 11:21:48 +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
tzhouamandClaude Fable 5.1 55192de984 [Edge][WP4] Add hardware adaptation: probe, derived overrides, auto profile, calibration 
* vllm_omni/edge/hardware_probe.py: HardwareProfile from /proc, /sys and torch
  (ISA flags incl. AVX-512/AMX and NEON/dotprod/i8mm/SVE, physical cores,
  big.LITTLE clusters, NUMA, RAM, CUDA discrete vs Tegra unified memory, NPU
  via override), hardware classes jetson / cuda_discrete / arm64_cpu / x86_cpu /
  npu_phone; VLLM_OMNI_HW_PROFILE=<json> overrides the probe (emulation, tests).
* vllm_omni/edge/adapt.py: derive_overrides -> dtype, absolute KV budgets from
  available memory minus weights, max_num_seqs, per-stage disjoint CPU thread
  ranges on the big cluster, eager/graph switches for small or unified memory,
  adaptive-chunk parameters, timeouts from a load cost model.
* vllm_omni/edge/calibrate.py: chunk-schedule calibration from measured chunk
  arrivals with a per-hardware-fingerprint cache under VLLM_CACHE_ROOT.
* vllm_omni/edge/auto_profile.py + deploy_profile="auto": edge/<model>.yaml ->
  edge/hardware/<class>.yaml -> derived overrides -> cached calibration,
  materialized as a temp deploy YAML (platform overlays are superseded).
* benchmarks/tts/hw_emulation.py: emulated classes (ATEN_CPU_CAPABILITY,
  taskset, profile overrides) driving stream_latency_bench with auto profiles.
* CPU-only tests for probe snapshots, derivation rules, calibration and auto.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 11:01:40 +08:00
tzhouamandClaude Fable 5.1 073d31b9cd [Edge][WP2] Add initialization timeline and cold-start profiler 
* vllm_omni/engine/init_timeline.py: InitTimeline phase collector; engine
  phases (stages_init, g3_device_lock, g2_spawn/ready) and worker phases
  (proc_entry, engine_core_init, init_device, load_model/load_weights,
  talker_mtp_init, attn_prewarm, cudagraph_capture, talker_mtp_capture,
  code2wav_cudagraph, predictor_setup_compile/warmup/cudagraph) recorded via
  worker_phase/timed_phase; workers append JSONL under
  VLLM_OMNI_INIT_TIMELINE_DIR (inherited at spawn) and the engine merges them.
  Enabled by VLLM_OMNI_INIT_TIMELINE=<json> or log_stats; table logged at INFO,
  partial table on orchestrator startup timeout, suggested timeouts logged.
* omni_base.py: named deploy profiles default to stage_init_timeout=900 /
  init_timeout=1500 unless given explicitly.
* gpu_model_runner._init_talker_mtp tolerates max_cudagraph_capture_size=None;
  gpu_ar_model_runner.shutdown skips CUDA graph teardown on CPU.
* benchmarks/tts/init_profile.py: runs stream_latency_bench --init-only per
  configuration (default, cold_cache, eager_stage1, eager_both,
  parallel_stage_init, edge, ...) and tabulates phase durations.
* stream_latency_bench: --parallel-stage-init / --deploy-profile flags; the
  resource poller no longer shadows Thread._stop and skips GPU sampling on CPU.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 11:01:39 +08:00
tzhouamandClaude Fable 5.1 f804bbd679 [Edge][WP0] Record Qwen3-TTS streaming baselines on 1x L20X 
Default deploy config (1 -> 25 frame schedule), 12 prompts x 3 repeats,
warm-up excluded, one GPU via `gpu run`:
  1.7B CustomVoice: TTFA p50 32.0 ms, playback-start p50 116 ms (stall at
    TTFA 85 ms), RTF_total 0.086, chunk gap p50 163 ms, init 271 s (cold caches)
  0.6B CustomVoice: TTFA p50 27.8 ms, playback-start p50 95 ms (stall 67 ms),
    RTF_total 0.077, chunk gap p50 146 ms, init 152 s
Per-request chunk arrivals, memory and the full summaries are in the JSON
(force-added: *.json is gitignored repo-wide).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 10:49:37 +08:00
tzhouamandClaude Fable 5.1 b70207f57f [Edge][WP0] Add offline streaming-latency benchmark for chunked TTS 
Adds benchmarks/tts/stream_latency_bench.py (AsyncOmni driver reusing the
Qwen3-TTS example's prompt builder; --model override, warm-up/repeat,
--init-only, per-chunk arrivals, peak RSS and per-process GPU memory) and
benchmarks/tts/stream_latency_metrics.py (TTFA, underrun-free playback start,
stall at TTFA, RTF streamed/total, chunk gaps, percentiles) with CPU unit
tests. Results go under benchmarks/tts/edge_results/stream_latency/.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-09 10:24:53 +08:00
Weiming LiaoandGao Han 1cd92553cf [CI][Bugfix] Use the matching checkpoint for Qwen3-TTS VoiceDesign tests (#7054) 
Signed-off-by: Weiming Liao <liaowm5@gmail.com>
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-09-06 04:27:21 +00:00
0z5aand0z5a 4927fd5412 [Core] Optimize diffusion LoRA expand accumulation (#6268) 
Signed-off-by: 0z5a <0z5a@users.noreply.github.com>
Signed-off-by: 0z5a <dinadegttyarenko6@gmail.com>
Co-authored-by: 0z5a <0z5a@users.noreply.github.com>
 2026-09-03 10:37:35 +08:00
DanaerLee ca6b4e2112 [Perf][Bugfix][OmniVoice] Restore float16 serving, and fuse the generator hot loop (#6317) 
Signed-off-by: MrlixiangWE <mrdanaer@gmail.com>
 2026-09-02 19:54:55 +00:00
Allen WuTRAE CLISy03TRAE CLI
 		f90e992dc1 [Model] Optimize CosyVoice3 TensorRT stream handoff (#5673) 
Signed-off-by: Allen Wu <allenwu2795@gmail.com>
Co-authored-by: TRAE CLI <noreply@bytedance.com>
Co-authored-by: Sy03 <1370724210@qq.com>
Co-authored-by: TRAE CLI <traecli@bytedance.com>
 2026-08-25 22:51:46 -07:00
Yukim1 6f55a06fab [Docs] Remove stale stage config references (#6270) 
Signed-off-by: zwhzzz0821 <2831474076@qq.com>
 2026-08-20 07:09:12 +08:00
Hongsheng Liu 8b8d3f56db [Misc] Consolidate LingBot text-to-video runner (#6076) 
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-08-13 08:42:37 +08:00
Shi Boao Bowie汪志鹏ruirui yangamy-why-3459Gao HanHongsheng Liu
 		fcf1085395 [Refactor][OutputProcessor 2/8]: OmniRequestOutput should inherit RequestOutput and no nested wrap-up (#5146) 
Signed-off-by: Boao Shi <aoibosh@connect.hku.hk>
Signed-off-by: bowieshi <aoibosh@connect.hku.hk>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: R2-Y <ruiruyang2@gmail.com>
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
Signed-off-by: Shi Boao Bowie <aoibosh@connect.hku.hk>
Co-authored-by: 汪志鹏 <wangzhipeng628@gmail.com>
Co-authored-by: ruirui(rein) yang <73573651+R2-Y@users.noreply.github.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-08-12 22:58:23 +08:00
Sy03andYueqian Lin 58007c0084 [New Model] Support IndexTTS 2.5 (#5957) 
Signed-off-by: Sysy <1370724210@qq.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
 2026-08-11 23:00:35 +00:00
WeiQing Chen a874b8e09b [Perf][CI] Add MiniMax-H3 4xH100 diffusion perf config (#5836) 
Signed-off-by: David Chen <530634352@qq.com>
 2026-08-07 00:12:07 +08:00
wtz2333andhsliuustc0106 5a6a93dc5f [diffusion][model] Add LingBot Video dense and MoE support (#5035) 
Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>
Signed-off-by: wtz2333 <2955110911@qq.com>
Co-authored-by: hsliuustc0106 <liuhongsheng4@huawei.com>
 2026-07-21 17:27:29 +08:00
bjf-frz b14ff23041 [PERF][CI]Add Cosmos3 diffusion perf config (#5010) 
Signed-off-by: bjf-frz <frz123db@gmail.com>
 2026-07-14 10:25:17 +00:00
faceless void 11f6a0b72e Make speech streaming default to SSE (#4679) 
Signed-off-by: syd520zy <529477025@qq.com>
 2026-06-30 02:17:24 +00:00
Canlin Guo 2e6c83c89c [Model] Support MOSS-TTS-local-v1.5 (#4664) 
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
 2026-06-25 15:43:03 +08:00
bjf-frz 2447740308 [Perf]perf: skip unused Cosmos3 I2V conditioning latents (#4614) 
Signed-off-by: bjf-frz <frz123db@gmail.com>
 2026-06-22 09:51:47 +00:00
NATUREandAlicia 83e3b129f0 [Bugfix] Fix missing memory metrics info for bagel (ar->dit diffusion) (#4477) 
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: Alicia <115451386+congw729@users.noreply.github.com>
 2026-06-17 14:06:14 +08:00
汪志鹏Yufeng ZhuZhengyuan SuClaude Opus 4.6Cursor
 		46e0b71423 [Enhancement] Introduce High-Performance MoT (Mixture-of-Tokens) Kernels: Triton Implementation (#3960) 
Signed-off-by: Yufeng Zhu <yzhu802@gatech.edu>
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: Zhengyuan Su <su.zhengyuan@u.nus.edu>
Co-authored-by: Yufeng Zhu <yzhu802@gatech.edu>
Co-authored-by: Zhengyuan Su <su.zhengyuan@u.nus.edu>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-06-16 03:08:02 +00:00
Sy03Yueqian LinHongsheng Liu
 		e27fd46ab9 [Perf] Optimize Higgs Audio v3 serving (#4204) 
Signed-off-by: Sy03 <1370724210@qq.com>
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-06-11 13:29:31 -07:00
Jianbao Huang 6cefe91370 [Perf] Enable fused RMSNorm for HunyuanImage3 (#3959) 2026-06-05 07:41:56 +00:00
Zhang JianandCanlin Guo b550709bba [TTS][Model] support MOSS-TTS series (#3420) 
Signed-off-by: Zhang <jianmusings@gmail.com>
Signed-off-by: Zhang Jian <jianmusings@gmail.com>
Signed-off-by: jian <jianmusings@gmail.com>
Signed-off-by: Canlin Guo <961750412@qq.com>
Co-authored-by: Canlin Guo <961750412@qq.com>
 2026-06-02 15:46:01 +08:00
+1
Zhou TaichangClaude Opus 4.7skfZheng WengangHongsheng LiuWeiQing ChenSYLARYueqian Linnainiu258JasonYueqian Lin
 		f92d84fcf9 [Rebase] Rebase to vllm releases/v0.22.0 (#3891) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Signed-off-by: ZhengWG <zwg0606@gmail.com>
Signed-off-by: david6666666 <530634352@qq.com>
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Signed-off-by: nainiu258 <cperfect02@163.com>
Signed-off-by: nainiu258 <101917677+nainiu258@users.noreply.github.com>
Signed-off-by: JasonJ2021 <jasonj@zju.edu.cn>
Signed-off-by: Zhou Taichang <tzhouam@connect.ust.hk>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Co-authored-by: skf <54565339+skf-1999@users.noreply.github.com>
Co-authored-by: Zheng Wengang <zwg0606@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: WeiQing Chen <40507679+david6666666@users.noreply.github.com>
Co-authored-by: SYLAR <125541396+lishunyang12@users.noreply.github.com>
Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: nainiu258 <101917677+nainiu258@users.noreply.github.com>
Co-authored-by: Jason <72191212+JasonJ2021@users.noreply.github.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
 2026-06-01 11:08:12 +08:00
ChenWenjingandGao Han 834abdaf8b [ci][nightly] Voxcpm2 performance benchmark (#3864) 
Signed-off-by: CHEN <116010019@link.cuhk.edu.cn>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-05-29 09:44:29 +08:00
WeiQing ChenlishunyangGao Han
 		126fb4b4a9 [Blackwell][1/N] Add SageAttention3 diffusion backend on blackwell(GB200/B200/RTX5090/PRO6000/DGX Spark available) (#3015) 
Signed-off-by: lishunyang <lishunyang12@163.com>
Signed-off-by: WeiQing Chen <40507679+david6666666@users.noreply.github.com>
Co-authored-by: lishunyang <lishunyang12@163.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-05-28 14:08:45 +00:00
SYLARfrauttauteffasuZhang Jian
 		373d3c7476 [Blackwell] Add CUDNN_ATTN and FLASHINFER_ATTN backends for diffusion (auto-route) (#3079) 
Signed-off-by: lishunyang <lishunyang12@163.com>
Signed-off-by: Zhang <jianmusings@gmail.com>
Signed-off-by: Zhang Jian <jianmusings@gmail.com>
Co-authored-by: frauttauteffasu <247219357+frauttauteffasu@users.noreply.github.com>
Co-authored-by: Zhang Jian <jianmusings@gmail.com>
 2026-05-28 11:09:25 +08:00
Dnoob a3802d1e5c [Doc][Benchmark] Rewrite benchmarks/README.md as repo-wide index (#1939) 
Signed-off-by: root <Dnoob>
Signed-off-by: Dnoob <dxpouo@gmail.com>
 2026-05-26 16:13:31 -10:00
dengyunyang 36eb8afba5 [HunyuanImage][End2End Performance CI] Add hunyuan end2end test (#3849) 
Signed-off-by: dengyunyang <584797741@qq.com>
 2026-05-26 16:45:05 +08:00
Yuanheng Zhao 2bc9cb7f13 [Feat] Support online quantization (fp8/int8) for LTX-2 (#3700) 
Signed-off-by: Yuanheng Zhao <jonathan.zhaoyh@gmail.com>
Signed-off-by: yuanheng <jonathan.zhaoyh@gmail.com>
 2026-05-21 19:52:21 +08:00
NumberWan 474710dda9 [Benchmark] Route i2i/ti2i to POST /v1/images/edits in diffusion_benchmark_serving (#3693) 
Signed-off-by: NumberWan <wantszkin2003@gmail.com>
 2026-05-21 17:18:02 +08:00
Yueqian Lin fc8486cf20 [Doc][TTS] CosyVoice3 online docs + residual TTS yaml cleanup + remove VoxCPM v1 (#3748) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
 2026-05-20 15:24:06 +08:00
Yueqian LinandHongsheng Liu f99e8b2d3c [Bench] Add audio-streaming continuity metric for TTS (#3618) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-05-20 05:08:56 +00:00
bjf-frzandbjfwhite 1fa734419e [Refactor]Rename diffusion benchmark backend to endpoint (#3137) 
Signed-off-by: bjf-frz <frz123db@gmail.com>
Signed-off-by: bjfwhite <baijingfan1@huawei.com>
Co-authored-by: bjfwhite <baijingfan1@huawei.com>
 2026-05-19 11:30:33 +08:00
Lancer dbc589dbca [Bugfix] fix diffusion quantization benchmarking for Omni outputs (#3653) 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
 2026-05-18 17:23:40 +08:00