amy-why-3459andHongsheng Liu 2696751003 [Perf] Qwen3-Omni performance optimization (#3878) ...

Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>

2026-05-29 22:15:23 +08:00

..

bagel

[Core] Support Async & Sync AutoRegressive Scheduling (#3306)

2026-05-05 12:09:45 +08:00

common

[Refactor] Unify Snake/SnakeBeta and alias-free activation into common modules (#3886)

2026-05-27 23:41:35 -05:00

cosyvoice3

[3/5][core]refactor communication layer: PR 3 of 5, all other models in non async mode (#3719)

2026-05-28 14:06:41 +08:00

covo_audio

[3/5][core]refactor communication layer: PR 3 of 5, all other models in non async mode (#3719)

2026-05-28 14:06:41 +08:00

dynin_omni

[Rebase] Rebase to vllm 0.20.0 (#3232)

2026-04-30 09:56:59 +12:00

fish_speech

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

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

glm_image

[Bugfix] Fix GLM-Image prior token debug logging (#3165)

2026-05-04 03:54:33 +00:00

glm_tts

[Model] Add GLM-TTS text-to-speech model support (#3141)

2026-05-24 05:42:26 +08:00

higgs_audio_v2

[Higgs-Audio] bosonai/higgs-audio-v2-generation-3B-base TTS model support (#3762)

2026-05-25 23:16:07 +08:00

hunyuan_image3

（Phase 2）Add ModelOpt mixed FP8/NVFP4 support for image generation (#3570)

2026-05-22 10:39:10 +08:00

mammoth_moda2

[Refactor] Standardize data entry key names to {type}.{qualifier} format (#1829)

2026-04-25 18:15:38 +08:00

mimo_audio

[3/5][core]refactor communication layer: PR 3 of 5, all other models in non async mode (#3719)

2026-05-28 14:06:41 +08:00

ming_flash_omni

[3/5][core]refactor communication layer: PR 3 of 5, all other models in non async mode (#3719)

2026-05-28 14:06:41 +08:00

moss_tts_nano

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

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

omnivoice

[Feat]Support voice clone for omnivoice in online serving & add seed parameter for reproducible (#3668)

2026-05-29 07:42:08 +00:00

qwen2_5_omni

[3/5][core]refactor communication layer: PR 3 of 5, all other models in non async mode (#3719)

2026-05-28 14:06:41 +08:00

qwen3_omni

[Perf] Qwen3-Omni performance optimization (#3878)

2026-05-29 22:15:23 +08:00

qwen3_tts

[Refactor][Qwen3-TTS] Construct speech tokenizer encoder natively (#3360)

2026-05-28 16:20:49 +08:00

voxcpm2

[Perf][VoxCPM2][Ming-Flash-Omni] Use global CUDA graph pool (#3361)

2026-05-20 05:01:42 +00:00

voxtral_tts

[Feature] Add FP8 quantization for Voxtral TTS (#3036)

2026-05-08 22:40:36 -04:00

__init__.py

[Core] Unified quantization framework (#1764)

2026-03-24 18:56:55 +08:00

output_templates.py

[Refactor] Standardize data entry key names to {type}.{qualifier} format (#1829)

2026-04-25 18:15:38 +08:00

registry.py

[Higgs-Audio] bosonai/higgs-audio-v2-generation-3B-base TTS model support (#3762)

2026-05-25 23:16:07 +08:00

utils.py

[Bugfix][MOSS-TTS-Nano] Drop fictional voice presets, require ref_audio (#3192)

2026-05-03 16:35:35 +12:00

whisper_utils.py

[Model] Add Ming-flash-omni-2.0 Thinker Stage (#1822)

2026-04-17 19:10:02 +08:00