anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity
2,942 Commits 235 Branches 31 Tags
c4e61133fbb5587e0cf8972ce32ffac4e37fa739
Commit Graph
205 Commits
Author SHA1 Message Date
tzhouamandClaude Fable 5.1 6ce8de1b28 [Edge][CPU] KV-cached code predictor, int8-dynamic CPU fast path, model-side CPU sampler 
Measured on a Xeon 8480C (12+4 threads, 0.6B): the 15 residual-codebook sub-steps
re-ran the 5-layer predictor over the whole prefix each time (97 ms per frame of
the 134 ms talker step) and vLLM's stock sampler spent 9 ms per step in its
torch.compile'd random-sample kernel on CPU.

- code predictor: forward_cached (KV cache per layer, 2-token prefill then one
  token per sub-step) and a shape-static forward_static_step (tensor position +
  boolean mask) that torch.compile handles as one graph; incremental decoding
  is the default on CPU (VLLM_OMNI_PREDICTOR_INCREMENTAL), optional compile
  (VLLM_OMNI_PREDICTOR_COMPILE=step), optional thread cap.
- CPU fast path: dynamic int8 Linear layers (fp32 activations) for the predictor
  and its lm heads (VLLM_OMNI_PREDICTOR_INT8, default on; micro-benchmark 70 ->
  36 ms per frame for the 15 sub-steps); VLLM_OMNI_PREDICTOR_PARITY_CHECK=1
  records top-1 agreement / max logit diff against the bf16 model in situ.
- Qwen3-TTS talker: prefer_model_sampler on CPU with a sample() that mirrors
  vLLM's Sampler (logits processors + penalties, temperature, top-k/top-p,
  exponential-noise sampling with per-request generators) without the compiled
  kernel (VLLM_OMNI_CPU_FAST_SAMPLER, default on).
- step_stats: sampler phase timers and sample-metadata timers.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-10 01:56:00 +08:00
tzhouamandClaude Fable 5.1 936c10be4b [Edge][WP8] Time the steady-state talker-MTP graph call (model.mtp_graph_ms) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L1ax5wYLA6XbqEL1i8L9gZ
 2026-09-09 21:42:52 +08:00
tzhouamandClaude Fable 5.1 abf81f4114 [Edge][WP8] Model-level step timers: talker forward, sampling, talker-MTP, code-predictor sub-steps 
Counters model.forward_ms / model.sample_ms / model.talker_mtp_ms and
predictor.substep_ms / predictor.loop_ms (VLLM_OMNI_STEP_STATS_DIR); with
VLLM_OMNI_STEP_STATS_SYNC=1 the accelerator is synchronized around the model
timers so async kernels are attributed (attribution runs only).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L1ax5wYLA6XbqEL1i8L9gZ
 2026-09-09 21:32:15 +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
Zhou TaichangandClaude Fable 5 142151fef5 Parallel stage initialization (admission + SH/EX device locks) (Worker refactor 1/N) (#5224) 
Signed-off-by: Taichang Zhou <tzhouam@connect.ust.hk>
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
 2026-09-05 22:22:41 +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
Inesh Reddy ChappidiandGao Han 66d938100b [Bugfix] Give exact-shape codec models an unpadded input_ids view (#6775) 
Signed-off-by: Inesh Reddy <ineshreddy249@gmail.com>
Signed-off-by: IneshReddy249 <ineshreddy249@gmail.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
 2026-09-04 15:28:49 -07:00
akshatvishu c41925689d [BugFix][Higgs-Audio-V3] Fix transcript and concurrent sampling tests (#6422) 
Signed-off-by: akshatvishu <akshatnayak197@gmail.com>
 2026-08-30 08:52:15 +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
Wallbreazzz 4ad455e61e [Feature][TTS] adaptive chunk ramp (Phase 2 buffer-feedback controller) (#6001) 
Signed-off-by: Wallbreazzz <110282866+Wallbreazzz@users.noreply.github.com>
 2026-08-28 02:11:05 -07:00
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
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
Sy03andsy03 432b2c6ca3 [Model] Add native full-duplex Nemotron VoiceChat serving (#6089) 
Signed-off-by: Sysy <1370724210@qq.com>
Signed-off-by: sy03 <137072421@qq.com>
Co-authored-by: sy03 <137072421@qq.com>
 2026-08-26 13:19:20 -07:00
wy17003andHongsheng Liu b454e34966 [Refactor][2/N Scheduler]Replace _FULL_PAYLOAD_INPUT_STAGES with a resolved stage transport capability. (#6149) 
Signed-off-by: wy17003 <yaow17003@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-08-25 15:24:59 +08:00
wangyu 1e2a6446f7 [CI] Remove AudioX and MammothModa2 support (#6353) 
Signed-off-by: wangyu <410167048@qq.com>
 2026-08-23 04:22:05 +00:00
NATUREandamy-why-3459 4f7c08c0eb [Bugfix][MiniCPM-o] Fix async-chunk snapshot replacement and prompt cleanup (#6406) 
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-08-21 16:37:21 +08:00
Inesh Reddy Chappidi cd3867c373 [Model] Suppress silence codec tokens for the first N Qwen3-TTS decod… (#5048) 
Signed-off-by: Inesh Reddy <ineshreddy249@gmail.com>
 2026-08-20 09:22:37 -07:00
Zhou TaichangandClaude Fable 5 10906e3a6d [Refactor] Worker/ModelRunner runner correctness fixes (G2/N) (#5452) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
 2026-08-20 16:47:50 +08:00
amy-why-3459andCursor 9c61d5b67b [Bugfix][MiniCPM-o] Stop Talker mid-utterance truncation and load cod… (#6346) 
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-20 11:50:07 +08:00
Yukim1 6f55a06fab [Docs] Remove stale stage config references (#6270) 
Signed-off-by: zwhzzz0821 <2831474076@qq.com>
 2026-08-20 07:09:12 +08:00
Yueqian LinandHongsheng Liu 611c083587 [BugFix][TTS] R1.1-R1.4: turn silent async-chunk failures into visible ones (#6033) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-08-17 10:51:17 +08:00
mjZhaoElaineClaude Opus 5Yueqian Lin
 		35cacf020c [Model] Add Gepard-1.0 native-AR FSQ/NanoCodec TTS (offline inference) (#5666) 
Signed-off-by: Mengjie Zhao <zmj0129@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
 2026-08-14 22:06:05 -07:00
Yuekai ZhangandClaude Fable 5 5be7541d01 [Model] Support NVIDIA-NemotronLabs-VoiceChat-11B offline speech-to-speech (#5842) 
Signed-off-by: Yuekai Zhang <zhangyuekai@foxmail.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
 2026-08-12 16:25:48 +08:00
王金旭xuruiClaude Opus 4.8
 		9b0df3909a [Feature] timestamps for TTS via a forced-aligner pooling (#4795) 
Signed-off-by: wjinxu <1299461899@qq.com>
Co-authored-by: xurui <xurui1@modelbest.onaliyun.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-08-12 05:33:34 +00:00
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
Zhou TaichangandClaude Opus 5 2ce15ab7a0 [Rebase] Rebase to vllm 0.27.0 (#5976) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-11 15:23:46 +00:00
R0CKSTARHongsheng Liuamy-why-3459
 		e04210d690 [Hardware][MUSA] Enable Qwen3-Omni ModelOpt FP8 inference (#5671) 
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-08-11 17:19:09 +08:00
Caspian ChenandSYLAR 593b404539 [Quantization][Qwen2.5-Omni] Support thinker-only ModelOpt NVFP4 W4A4 checkpoints (#5073) 
Signed-off-by: Caspian443 <scrisis843@gmail.com>
Co-authored-by: SYLAR <125541396+lishunyang12@users.noreply.github.com>
 2026-08-09 02:45:57 +08:00
Yueqian LinandSy0307 e788ef6e76 [Feature] PersonaPlex (Moshi-based full-duplex S2S): native vLLM port + duplex serving (#4771) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Signed-off-by: Sy0307 <24841857+Sy0307@users.noreply.github.com>
Co-authored-by: Sy0307 <24841857+Sy0307@users.noreply.github.com>
 2026-08-04 20:10:33 -07:00
NATURE 1c4aca7f6e [Bugfix][Hunyuan/Bagel]Avoid payload connector for KV-only senders (#5744) 
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
 2026-08-04 14:40:27 +00:00
SYLARandlishunyang12 6f98c807d2 [Bugfix] Restore BAGEL CFG KV transfers (#5620) 
Signed-off-by: lishunyang12 <lishunyang03@gmail.com>
Co-authored-by: lishunyang12 <lishunyang03@gmail.com>
 2026-08-01 15:06:05 +00:00
Hongsheng Liu f775e80732 Revert "[BugFix] Fix 'GPUARWorker' attribute error" (#5507) 2026-07-28 21:00:49 +08:00
Ting FU f4640d9a2e [BugFix] Fix 'GPUARWorker' attribute error (#5478) 
Signed-off-by: Semmer <semmer@live.cn>
 2026-07-28 20:13:33 +08:00
Yuekai ZhangClaude Fable 5Yueqian LinYueqian Lin
 		5014fea4a0 [New Model] support nemotron Audex (#4976) 
Signed-off-by: Yuekai Zhang <zhangyuekai@foxmail.com>
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
 2026-07-28 03:58:22 +00:00
Zhou TaichangClaude Fable 5tzhouam
 		cbab1d4d8f [Rebase] Rebase to vllm 0.26.0 (#5443) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Signed-off-by: tzhouam <tzhouam@gmail.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: tzhouam <tzhouam@gmail.com>
 2026-07-27 23:35:49 +08:00
Wallbreazzz e28e389a76 feat(qwen3-tts): add codec_chunk_ramp for gradual chunk size warmup (#5152) 
Signed-off-by: Wallbreazzz <110282866+Wallbreazzz@users.noreply.github.com>
 2026-07-27 04:15:43 +00:00
rein yangandSy03 ad3a162eb8 [perf] Minicpm o4.5 optimization (#5228) 
Signed-off-by: R2-Y <ruiruyang2@gmail.com>
Co-authored-by: Sy03 <1370724210@qq.com>
 2026-07-24 17:20:39 +08:00
Sy03lishunyang12Yueqian LinStormMelodyGao HanHongsheng Liu
 		05b794f29b [Full Duplex] Feat: Support Full-Duplex realtime runtime & add MiniCPM-o 4.5 demo (#3907) 
Signed-off-by: Sy03 <1370724210@qq.com>
Signed-off-by: lishunyang12 <lishunyang03@gmail.com>
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: lishunyang12 <lishunyang03@gmail.com>
Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: StormMelody <22371184@zju.edu.cn>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-07-23 16:45:18 +08:00
vOvandClaude Opus 4.8 cc1939120f [Bugfix] Skip non-positive scheduled token spans in omni stage runners (#5269) 
Signed-off-by: cr-gao <gaochenrui@sjtu.edu.cn>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
 2026-07-22 16:01:45 +00:00
Bingyu (Spencer) Liu dcf8b07d39 [Core] Cleanup Omni connector runtime and transfer backends (#5096) 
Signed-off-by: spencerr221 <liubingyu62@gmail.com>
 2026-07-22 09:44:15 +08:00
Canlin Guoandsunnyxyli 5c3900960b [Feature] Support MOSS-TTS-Local-v1.5 vocoder graph (#4929) 
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
Signed-off-by: sunnyxyli <1550693033@qq.com>
Co-authored-by: sunnyxyli <1550693033@qq.com>
 2026-07-16 15:02:15 +08:00
Zhou TaichangandClaude Opus 4.8 d3c47efc4f [Rebase] Rebase to vllm v0.25.0 (#5042) 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
 2026-07-12 23:39:53 +08:00
amy-why-3459 9a44e7e019 [Refactor][Phase 1]Remove redundant functions and logs (#4986) 
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-07-10 15:13:35 +08:00
Yueqian Lin 6d50348ef1 [Perf][Qwen3-TTS] Skip per-step hidden-state D2H via hidden pooler payload opt-out (#4879) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
 2026-07-08 11:59:41 -07:00
Yueqian Lin da15a4cb1c [Bugfix] Drop out-of-vocabulary stop ids from min-tokens masking (qwen3-tts min_tokens engine crash) (#4971) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
 2026-07-08 11:24:31 -07:00
amy-why-3459 9ceca4ec85 [Bugfix] Fix full-payload mm splitting for dual hidden/scheduled batch axes (#4870 follow-up) (#4910) 
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-07-06 15:08:05 +08:00
Yueqian Lin d7b365df64 [Bugfix][Qwen3-TTS] Keep seeded residual MTP sampling batched (#4889) 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
 2026-07-05 21:47:06 +08:00
Sy03andYueqian Lin 1b31f96240 [Bugfix] Fix Qwen3-TTS no-async-chunk payload splitting (#4870) 
Signed-off-by: Sy03 <1370724210@qq.com>
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
 2026-07-04 19:57:55 +00:00
Sy03andYueqian Lin 9eaca72382 [Perf][VoxCPM2] Add unified decode graph optimization (#4606) 
Signed-off-by: Sy03 <1370724210@qq.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
 2026-07-04 05:30:48 +00:00
amy-why-3459 20d43ce24d [BugFix] Adapt to pin_memory in vLLM v0.24.0 (#4860) 
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
 2026-07-03 21:35:26 +08:00