Commit Graph
1720 Commits
Author SHA1 Message Date
psv666 e3be42e052 [Core] Give the OpenAI Realtime wire codec a reusable home (RFC #6592 P0a) (#7640)
Signed-off-by: psv666 <2693925048@qq.com>
2026-09-17 02:33:40 -07:00
MrDongslsandMr.Dong 825ab20b07 [Model][PersonaPlex] add per-row RingKV offsets and active rows (#7670)
Signed-off-by: Mr.Dong <1042542469@qq.com>
Co-authored-by: Mr.Dong <1042542469@qq.com>
2026-09-17 01:50:40 -07:00
WeiQing Chen cb439f3e84 [Diffusion][Model] Add direct FastH3 8-Step V2 checkpoint loading (#7610)
Signed-off-by: david6666666 <530634352@qq.com>
2026-09-17 08:41:52 +00:00
Alex Brooks 8c3d3407d1 [CI] Fix Structured Structured Diffusion Config Wan Test (#7707)
Signed-off-by: Alex Brooks <albrooks@redhat.com>
2026-09-17 00:01:01 -07:00
LOGO127 224152566a [Model][PersonaPlex] Add paced multi-session validation (#7428)
Signed-off-by: luozijian <luozijian0924@gmail.com>
2026-09-16 23:04:37 -07:00
Sy03 4200768f59 [Frontend] Scale Omni serving across API processes (#6923)
Signed-off-by: Sy03 <1370724210@qq.com>
2026-09-16 22:29:31 -07:00
maithilijoshi20 f1a6e7ce2a [Core] Retain structured configs through runtime startup (#6849)
Signed-off-by: maithilijoshi20 <97733343+maithilijoshi20@users.noreply.github.com>
2026-09-16 20:47:38 -07:00
HAAZZZEEEE becfefd440 [Hardware][Ascend] Add Wan2.2 MXFP4 UOS, Smooth and W4A8 fallback (#7210)
Signed-off-by: HAAZZZEEEE <wang-taicheng@qq.com>
2026-09-17 03:11:40 +00:00
Linze Shi 4fab34c8f4 [Bugfix] Update ComfyUI serving-mode mock after helper rename (#7694)
Signed-off-by: Linze-Shi <linzeshi0@gmail.com>
2026-09-17 02:45:31 +00:00
WeiQing Chen 72cc012cfd [CI][bugfix] Fix MiniMax H3 FP8 quality test remote-code loading (#7617) 2026-09-16 18:52:23 -07:00
Linze Shi c8b3a17bd3 [Frontend] Allow explicit turn-based serving (#7675)
Signed-off-by: Linze-Shi <linzeshi0@gmail.com>
2026-09-16 17:29:57 -07:00
Aman c32aaeb2da [Bugfix] Fix Qwen3-TTS word timestamps with async chunking (#7544)
Signed-off-by: bezdarnost <amanurumbekov@gmail.com>
2026-09-16 15:07:52 -07:00
Weiming Liao 63032c91c6 [Frontend] Add ComfyUI Generate Music node with initial MiniMax Music 3 support (#7516)
Signed-off-by: Weiming Liao <liaowm5@gmail.com>
2026-09-16 13:43:43 -07:00
Yueqian Lin 5aadcbb0ab [Bugfix] Fix Breeze-TTS-2 repeated silence and audio EOS handling 2026-09-16 18:40:12 +00:00
akshatvishu a8434a60da [BugFix][TTS] Fix MOSS Realtime serving and generation (#5661)
Signed-off-by: akshatvishu <akshatnayak197@gmail.com>
2026-09-16 11:39:51 -07:00
nodeeeeee 04c55d0164 Preserve audio boundary embeddings in Qwen2.5-Omni interleaved input (#7510)
Signed-off-by: nodeeeeee <zhangkai.nodeee@gmail.com>
2026-09-16 11:39:13 -07:00
shiyichuanandmershi c97b5fc741 [Bugfix][Frontend] Route synthetic aborts through final stage (#7006)
Signed-off-by: mershi <mershi@tencent.com>
Co-authored-by: mershi <mershi@tencent.com>
2026-09-16 11:38:45 -07:00
Lei Ke 38c7eeee4c [BugFix][NPU] Fix Fish Speech S2 Pro NPU support (#7546)
Signed-off-by: Lei Ke <1141466880@qq.com>
2026-09-16 11:38:06 -07:00
Deep Shah 4392af5ce3 [Bugfix][Qwen3-Omni] Include deferred residual in captured Thinker states (#7304)
Signed-off-by: Deep Shah <deep@socratic.co>
2026-09-16 11:36:52 -07:00
hywandAlex Brooks 2a68471715 [Bugfix] Reject failed diffusion benchmark warmups (#7092)
Signed-off-by: hyw <yuweih205@gmail.com>
Signed-off-by: HuangYuwei <yuweih205@gmail.com>
Signed-off-by: Alex Brooks <albrooks@redhat.com>
Co-authored-by: Alex Brooks <albrooks@redhat.com>
2026-09-16 17:09:23 +00:00
y-null 3d75254a44 [CI][MiniCPM-o] Lower the Daily-Omni accuracy gate to 0.77 (#7657)
Signed-off-by: y_null <y_null@qq.com>
2026-09-16 16:00:45 +00:00
NumberWanandwangyu 81b5be274d [Bugfix][Qwen-Image] Restore RotaryEmbedding CUDA RoPE for Diffusers e2e (#7513)
Signed-off-by: NumberWan <wantszkin2003@gmail.com>
Co-authored-by: wangyu <53896905+yenuo26@users.noreply.github.com>
2026-09-16 08:54:28 -07:00
psv666andamy-why-3459 1053f7b0bd [Bugfix] Preserve benchmark metric sample counts in results (#7624)
Signed-off-by: psv666 <2693925048@qq.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-16 08:39:01 -07:00
Zhengyuan Su (苏政渊) 538eddd74d [Perf][CosyVoice3] Bounded-window incremental HiFT vocoder streaming (#7521)
Signed-off-by: Zhengyuan Su <su.zhengyuan@u.nus.edu>
2026-09-16 20:49:04 +08:00
fa639b889e [Bugfix] Fail a duplex session that outgrows max_model_len instead of killing the EngineCore (#7277)
Signed-off-by: Tianyao Wu <rayroy31@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-16 20:08:25 +08:00
e6cebf8557 [Bugfix][Omni-DuplexEval] Load local Hugging Face dataset layouts in duplex eval loader (#7331)
Signed-off-by: RyanYun09 <RyanYun09@users.noreply.github.com>
Co-authored-by: RyanYun09 <RyanYun09@users.noreply.github.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-16 20:05:52 +08:00
dongbo910220 217a6af52c [Diffusion] Fuse Qwen-Image-Edit select01 modulation for CUDA (#5921)
Signed-off-by: dongbo910220 <1275604947@qq.com>
2026-09-16 19:06:28 +08:00
dengyunyang 249e84d21b [Boogu-Image] Fuse QKV/FFN projections + switch to diffusion RMSNorm (#6649)
Signed-off-by: dengyunyang <584797741@qq.com>
2026-09-16 03:13:54 -07:00
d554e6b67e [feat] Extend disaggregated MiniMax-H3 stage with VAE encoders (#6939)
Signed-off-by: asukaqaq-s <1311722138@qq.com>
Signed-off-by: Jader <yjader@foxmail.com>
Co-authored-by: Jader <yjader@foxmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-16 16:52:35 +08:00
Linze Shi e78a5d01f7 [Frontend] Add MiniMax-H3 references and Ref2VA workflow (#7483)
Signed-off-by: Linze-Shi <linzeshi0@gmail.com>
2026-09-16 08:30:16 +00:00
y-nullandamy-why-3459 5f7f62e13c [Bugfix][MiniCPM-o] Stabilize code2wav CFM CUDA graph caching with reference audio normalization and mel-frame bucketing (#7416)
Signed-off-by: y_null <y_null@qq.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-16 01:24:31 -07:00
SYLARandlishunyang12 507cb1d834 [Attention] [P1] Keep H3 VSA policy in the model and share sparse primitives (#7535)
Signed-off-by: lishunyang12 <lishunyang12@users.noreply.github.com>
Co-authored-by: lishunyang12 <lishunyang12@users.noreply.github.com>
2026-09-16 15:47:31 +08:00
chickeytonandClaude Opus 5 99ff4f3070 [Core][Frontend] Unified Full-duplex Framework (#7413)
Signed-off-by: chickeyton <ngton2014@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 15:35:36 +08:00
linzhenpl07andClaude Opus 5 9005d78903 [LingBot World] Reuse the text encode for a prompt already encoded (#6845)
Signed-off-by: linzhenpl07 <linzhenpl07@gmail.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-09-16 15:01:11 +08:00
Zhou Taichang 5e9024171d [Bugfix] Skip generic text-only warmup for DreamZero (#7548)
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
2026-09-16 14:43:18 +08:00
vOv 21d86ec923 [Diffusion] MAGI-2: regional compile coverage (#7174)
Signed-off-by: cr-gao <gaochenrui@sjtu.edu.cn>
2026-09-16 13:01:26 +08:00
GXIN a2138b830f [Feature][Qwen3-TTS] Make Code2Wav honor the stage dtype (#6059)
Signed-off-by: XIN GAO <1037396230@qq.com>
2026-09-16 04:50:00 +00:00
Zheng Wengangandwuli666 275720b378 [Bugfix] Keep async video jobs queued until inference and abort on delete (#6759)
Signed-off-by: ZhengWG <zwg0606@gmail.com>
Signed-off-by: wuli666 <421774554@qq.com>
Co-authored-by: wuli666 <421774554@qq.com>
2026-09-15 20:43:30 -07:00
Yueqian Lin 5dbae2b182 [Bugfix] Per-key accumulation strategy for Qwen3-TTS codec frame outputs (#7608)
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
2026-09-15 20:08:49 -07:00
psv666 7bceb1a76b [Bugfix] Align video frame consumption reporting with prompt sampling (#7329) 2026-09-16 02:49:07 +00:00
HAAZZZEEEE 3eb5c3b4e6 [Bugfix][Wan2.2] Apply UniPC flow shift once (#7576)
Signed-off-by: HAAZZZEEEE <wang-taicheng@qq.com>
2026-09-16 10:06:41 +08:00
tlysanhuo 2185c4e26f [Bugfix] Tolerate a bounded TTS tail in the audio-text similarity gate (#7358)
Signed-off-by: tly <2200895168@qq.com>
2026-09-15 18:51:49 -07:00
Deep Shah c197cd2422 [Bugfix] Fix Qwen3-Omni streaming crops in mixed-length batches (#7340)
Signed-off-by: Deep Shah <deep@socratic.co>
2026-09-15 18:26:43 -07:00
Clodagh Walsh a1a234a0b4 [Voxtral] Raise BadRequest for ref_audio (#7012)
Signed-off-by: Clodagh Walsh <clodaghwalsh17@gmail.com>
2026-09-15 18:26:08 -07:00
leo 45a3e77351 [Bugfix] Route Music3 float32 attention to SDPA (#7354)
Signed-off-by: liuyao0322 <yaoliu548926@gmail.com>
2026-09-15 18:08:06 -07:00
Weiming Liao 5527f3f407 [CI][Bugfix] Update Qwen3-TTS CI Validation Message (#7276)
Signed-off-by: Weiming Liao <liaowm5@gmail.com>
2026-09-15 18:07:44 -07:00
ChangXiong Denganddengchangxiong f420cfcfbf [NPU][CosyVoice3] Add CPU STFT fallback (#7520)
Signed-off-by: dengchangxiong <dengchangxiong1@huawei.com>
Co-authored-by: dengchangxiong <dengchangxiong1@huawei.com>
2026-09-15 18:06:41 -07:00
LOGO127 65693b30fb [Bugfix][Core] Keep sample-rate snapshots bounded in delta output (#7448)
Signed-off-by: luozijian <luozijian0924@gmail.com>
2026-09-15 16:50:32 -07:00
yzhautouskay 6040bb97e2 [Perf][Cosmos3] Add SeaCache support for Cosmos3 (#6922)
Signed-off-by: Yuliya Zhautouskaya <yzhautouskay@nvidia.com>
2026-09-15 14:57:15 -07:00
Mengjie ZhaoandCursor c17126c219 [Model] Add Gepard-1.0 OpenAI /v1/audio/speech serving (#7499)
Signed-off-by: Mengjie Zhao <zmj0129@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-15 14:56:51 -07:00