Kwokho TsuiandKwokho Tsui 7e152ca64b [Bugfix] Fix HSDP + FP8 online quantization compatibility (#4494) ...

Signed-off-by: Kwokho Tsui <kh_tsui@163.com>
Co-authored-by: Kwokho Tsui <kh_tsui@163.com>

2026-06-19 15:06:27 +08:00

..

attention

[TTS][Perf] Optimize Fish Speech S2 Pro high-concurrency serving (#3773)

2026-05-27 13:25:49 -05:00

benchmarks

[BugFix] Fix the issue of incorrect generated tokens (#4461)

2026-06-17 02:31:48 +00:00

config

[Model] Add Ming-omni-tts dense 0.5B pipeline (#2906)

2026-06-10 20:11:04 +08:00

core

[Perf][PrefixCache] Avoid per-step blocking write in OmniTensorPrefix… (#4106)

2026-06-17 16:53:47 +08:00

deploy

[Perf][DreamZero]dreamzero supports CUDA graph, torch.compile, and DiT Caching. (#4213)

2026-06-19 14:24:37 +08:00

diffusion

[Bugfix] Fix HSDP + FP8 online quantization compatibility (#4494)

2026-06-19 15:06:27 +08:00

distributed

[Refactor] Omni Stage Runtime and Distributed Replica Control Plane (#3855)

2026-06-17 14:13:20 +08:00

engine

[Perf][DreamZero]dreamzero supports CUDA graph, torch.compile, and DiT Caching. (#4213)

2026-06-19 14:24:37 +08:00

entrypoints

[Chore][TTS] Reference recipes.vllm.ai recipes, remove dead code, and fix stale references (#4567)

2026-06-19 15:05:26 +08:00

inputs

Add Cosmos3 model support (#3454)

2026-06-02 04:04:31 +08:00

lora

[Feature] Diffusion LoRA Adapter Support (PEFT compatible) for vLLM alignment (#758)

2026-01-27 10:22:40 +08:00

metrics

Remove unregister_vllm_metrics no-op patch (#4396)

2026-06-18 10:13:49 +08:00

model_executor

[Chore][TTS] Reference recipes.vllm.ai recipes, remove dead code, and fix stale references (#4567)

2026-06-19 15:05:26 +08:00

model_extras

[Refactor]Base class for dit pipelines with unified parameter declaration[1/N] (#4225)

2026-06-18 17:14:18 +08:00

platforms

[DEBUG]Add has_preprocess branch to _dummy_run for models with preprocess stage (#4189)

2026-06-18 07:56:19 +08:00

plugins

[Hardware] Support platforms and plugin system (#774)

2026-01-28 22:12:52 +08:00

profiler

[Enhancement]modify profiling.md (#3051)

2026-04-24 16:30:24 +08:00

quantization

[Quant][Perf] Default Blackwell FP8 GEMM to quack CuteDSL fused-bias kernel (#4241)

2026-06-08 19:44:20 +00:00

reasoning

Step audio R1 reasoning parser (#2846)

2026-06-10 21:03:19 +00:00

sample

update dirs

2025-09-11 15:43:43 +08:00

tokenizers

add support for MammothModa2 model (#336)

2026-03-04 22:52:56 +08:00

transformers_utils

[Chore][TTS] Reference recipes.vllm.ai recipes, remove dead code, and fix stale references (#4567)

2026-06-19 15:05:26 +08:00

utils

[Refactor] Output Processor Phase 2: separate multimodal output channel (#1601) (#2744)

2026-06-12 15:19:14 +08:00

worker

[Bugfix] MOSS-TTS: fix cross-request audio corruption under batching (#4415)

2026-06-18 18:11:51 +00:00

__init__.py

[Bugfix][Refactor] Migrate Voxtral TTS config and parser registry (#3065)

2026-04-25 18:17:25 +08:00

data_entry_keys.py

[Perf][TTS] Optimize cosyvoice TTFP and throughput using TensorRT (#4168)

2026-06-10 16:42:34 +08:00

errors.py

[refactor] Refactor guardrail error handling - add 400 error code (#4297)

2026-06-11 21:09:24 +00:00

logger.py

[BugFix] Failed to print DEBUG level log (#307)

2025-12-15 07:13:36 +00:00

outputs.py

[Refactor] Output Processor Phase 2: separate multimodal output channel (#1601) (#2744)

2026-06-12 15:19:14 +08:00

patch.py

Remove unregister_vllm_metrics no-op patch (#4396)

2026-06-18 10:13:49 +08:00

request.py

[Rebase] Rebase to vllm v0.21.0 (#3530)

2026-05-16 18:18:43 +08:00

version.py

[Misc] Warn When vLLM / vLLM-Omni Have Mismatched Versions (#2691)

2026-04-17 20:15:16 +08:00