Commit Graph
2318 Commits
Author SHA1 Message Date
Shenglei FuandClaude eb67caaaa9 [Diffusion] Enable request-level batching for Boogu-Image (TI2I) (#7227)
Signed-off-by: Shenglei Fu <sfu@confluent.io>
Signed-off-by: Shenglei Fu <117230642+ShengleiFu@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
2026-09-26 22:53:35 +08:00
R0CKSTAR e75278dccc feat(magi2): add fused SwiGLU7 activation (#7252)
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
2026-09-26 22:53:04 +08:00
Sy03 e1303e5067 [Model] Optimize Qwen3-TTS streaming throughput and first audio on CUDA (#8163)
Signed-off-by: Sy03 <1370724210@qq.com>
2026-09-26 22:51:04 +08:00
Anjie HouandHongsheng Liu 8a49c65bf9 [Core][Diffusion] Add byte backpressure to chunk transfers (#7916)
Signed-off-by: specture724 <specture724@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-26 17:48:32 +08:00
acf6620233 [Refactor][Diffusion] Read the resolved offload policy outside the config layer (#7327)
Signed-off-by: specture724 <specture724@gmail.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-26 17:45:20 +08:00
Gao Han a8576ccb72 [Bugfix] Fix MiniCPM-o 4.5 seeded turns and window E2E inputs (#8160)
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-26 01:42:53 +08:00
Sy03 d31a2a3ba8 [Bugfix] Fix Nemotron VoiceChat duplex nightly handshake (#7884)
Signed-off-by: Sy03 <1370724210@qq.com>
2026-09-25 17:25:15 +00:00
Sy03 5d0963b3dc [Bugfix] Fix MRV2 payload slicing with CUDA Graph padding (#8159)
Signed-off-by: Sy03 <1370724210@qq.com>
2026-09-25 16:07:22 +00:00
chickeytonandClaude Opus 5.5 4be6bbe0e4 [Bugfix][MiniCPM-o] Build the Stage-0 duplex runtime at load time (#8146)
Signed-off-by: chickeyton <ngton2014@gmail.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-25 14:57:23 +00:00
bcsdhjew fe9e22df3d [Bugfix] Restore MiniCPM duplex Seed-TTS startup (#8152)
Signed-off-by: Nolen Liang <nliang@nvidia.com>
Signed-off-by: bcsdhjew <nliang@nvidia.com>
2026-09-25 14:52:46 +00:00
Yuanheng Zhao 3bd5ac9685 [Bugfix] Align Ming-Image prompt templates and RNG, fix number of layer to decompose (#8118)
Signed-off-by: Yuanheng <jonathan.zhaoyh@gmail.com>
2026-09-25 22:47:29 +08:00
b41e6707b1 [Bugfix][MammothModa2] Stop allocating AR KV cache for the replaced language model (#8095)
Signed-off-by: MrlixiangWE <mrdanaer@gmail.com>
Co-authored-by: zijianc2 <157244773+zijianc2@users.noreply.github.com>
Co-authored-by: hsliuustc0106 <222337142+hsliuustc0106@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-25 21:28:59 +08:00
29dacbd3ef [Bugfix][MammothModa2] Keep AR layer 0 in BF16 in the FP8 KV preset (#8083)
Signed-off-by: MrlixiangWE <mrdanaer@gmail.com>
Co-authored-by: zijianc2 <157244773+zijianc2@users.noreply.github.com>
Co-authored-by: hsliuustc0106 <222337142+hsliuustc0106@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-25 21:25:49 +08:00
95781c31e0 [Bugfix][Diffusion] Skip AllGather DLO weight restore on worker shutdown (#8103)
Signed-off-by: specture724 <specture724@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-25 08:33:32 +00:00
R0CKSTARandHongsheng Liu 3bf96080f3 feat(magi2): add fused mHC post-processing (#7261)
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-25 16:25:12 +08:00
0c7042d0d5 [Feature][Diffusion] Add BOOGU Image sequence parallel support (#5718)
Signed-off-by: Xenoryn <timberh2000@outlook.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
2026-09-25 16:18:23 +08:00
NumberWan e5613794ca [BugFix][Cosmos3] Decode uploaded video paths in v2v preprocess (#8106)
Signed-off-by: NumberWan <wantszkin2003@gmail.com>
2026-09-25 07:53:24 +00:00
Gao Han 69de153f6e [Bugfix] Stop cancelled MiniMax-H3 requests at denoising boundaries (#7844)
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-25 13:36:09 +08:00
BeatSeatandamy-why-3459 b308e196c9 [Model] Support request-level batching for MammothModa2 DiT on shared runtime (#7476)
Signed-off-by: BeatSeat <wendavid552@gmail.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-25 10:34:59 +08:00
9e7e3a802c [Bugfix] Compensate cancelled duplex open, append and resume; fix serving README drift (#7636 Issues 1, 2, 3, 21) (#7785)
Signed-off-by: chickeyton <ngton2014@gmail.com>
Signed-off-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
2026-09-24 19:23:08 -07:00
Sy03 551a03102f [Bugfix][Qwen3-TTS] Decode full payload from model buffer (#8132)
Signed-off-by: Sy03 <1370724210@qq.com>
2026-09-25 10:03:58 +08:00
f46c958b67 [Bugfix][Qwen3-Omni] Restore the talker sampling seed (#7474)
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Signed-off-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
2026-09-24 17:41:03 -07:00
513dba842a [BugFix] Add modalities and logprobs validation for /v1/chat/completions (#4743)
Signed-off-by: Shaun Walsh <shaunwalsh24@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Alex Brooks <albrooks@redhat.com>
2026-09-24 18:09:19 -06:00
Jim BanandYueqian Lin 4aac1294f1 [Bugfix][Frontend] Surface streaming terminal submit failures (#7011)
Signed-off-by: BANANASJIM <bananasjim1@gmail.com>
Signed-off-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com>
2026-09-24 16:50:43 -07:00
Joshna-Medisetty b539282a9a [Bugfix][XPU] Enable LTX-2 family on XPU (#7961)
Signed-off-by: Joshna Medisetty <joshna.medisetty@intel.com>
2026-09-24 13:57:38 -05:00
ChoHeeandGao Han af46abebb5 [Bugfix][Diffusion] Bound diffusion execute RPCs with the DP wave timeout (#7745)
Signed-off-by: ChoHee15 <cc5281@126.com>
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-24 16:56:41 +00:00
Aman f58ef57ac4 [Bugfix] Enable pipeline profiling for LingBot Video (#7744)
Signed-off-by: bezdarnost <amanurumbekov@gmail.com>
2026-09-24 16:22:17 +00:00
NumberWanandOliver7th 9376c21edd [Bugfix][Cosmos3] Resolve per-component transformer quantization config (#8101)
Signed-off-by: NumberWan <wantszkin2003@gmail.com>
Co-authored-by: Oliver7th <fangzhengyi@huawei.com>
2026-09-25 00:18:34 +08:00
yzong-rh 56da589a3e [Bugfix] Sync TeaCache skip decision across SP to prevent deadlock (#6228)
Signed-off-by: Yifan Zong <yzong@redhat.com>
2026-09-24 16:17:27 +00:00
28c87cde19 [Bugfix] Replace hardcoded Vivian speaker default with config-ordered default (#5814)
Signed-off-by: khairulkabir1661 <khairulkabir1661@users.noreply.github.com>
Co-authored-by: khairulkabir1661 <khairulkabir1661@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Alex Brooks <albrooks@redhat.com>
2026-09-24 09:10:01 -07:00
psv666 2d00d86e5e [CI/Build] Fix Qwen3-Omni Talker benchmark workload (#8057)
Signed-off-by: psv666 <2693925048@qq.com>
2026-09-24 23:56:53 +08:00
0295eeb6f3 [Orchestrator] Implement cache reset RPC forwarding to AR stage engine cores (#6766)
Signed-off-by: Asthenia <asthenia0412@gmail.com>
Signed-off-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Asthenia <asthenia0412@gmail.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-24 15:32:07 +00:00
535bd35b42 [Perf] Add reusable exact projection caching with MiniMax-H3 integration (#7987)
Signed-off-by: lishunyang12 <lishunyang12@users.noreply.github.com>
Signed-off-by: Armaan Amatya <armaanamatya2014@gmail.com>
Co-authored-by: lishunyang12 <lishunyang12@users.noreply.github.com>
Co-authored-by: Armaan Amatya <38102001+armaanamatya@users.noreply.github.com>
2026-09-24 23:30:42 +08:00
Khairul Kabirandkhairulkabir1661 63bbd19a1d [Bugfix] Read hidden layer key from model config instead of hardcoding 24 (#5825)
Signed-off-by: khairulkabir1661 <khairulkabir1661@users.noreply.github.com>
Co-authored-by: khairulkabir1661 <khairulkabir1661@users.noreply.github.com>
2026-09-24 09:19:41 -06:00
andyluo7 9b08ac8203 [Bugfix][Diffusion] Clean up inline worker lifecycle (#7986)
Signed-off-by: andyluo7 <andy.luo@amd.com>
2026-09-24 21:43:25 +08:00
Zheng Wengangandamy-why-3459 ed5fc2b83b [Hotfix][Core] PrefixCache: re-plan same-step prefix hits whose rows are written after the early prefetch (#7877)
Signed-off-by: ZhengWG <zwg0606@gmail.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
2026-09-24 12:35:24 +00:00
9c1656c607 [Model] Reduce MiniCPM-o 4.5 reference setup and Talker overhead (#7968)
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: xjwu <wuxianjin633@gmail.com>
Co-authored-by: Eric(Qun) <186020953+0xtoward@users.noreply.github.com>
2026-09-24 05:18:56 -07:00
Zhou TaichangandClaude Fable 5.1 ba3d48e1c7 [LingBot World] Speed up streaming VAE decode and RGB frame delivery (#7749)
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-24 03:56:44 -07:00
NumberWan 7650020618 [BugFix][Cosmos3] Register opt-in T2I pipeline so deploy-config applies (#7971)
Signed-off-by: NumberWan <wantszkin2003@gmail.com>
2026-09-24 10:08:46 +00:00
NATUREandliuqihao 7a8d956ece [Bugfix] Fix MiniCPM-o 4.5 native duplex speech alignment (#7974)
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: liuqihao <liuqihao970610@gmail.com>
2026-09-24 10:04:28 +00:00
NATUREandGagan Dhakrey d486bf097a [Bugfix] Reclaim aborted chunk SHM and consumed connector state (#8082)
Signed-off-by: natureofnature <wzliu@connect.hku.hk>
Co-authored-by: Gagan Dhakrey <gagandhakrey@gmail.com>
2026-09-24 17:59:43 +08:00
z2z23n0andCanlin Guo 3b02e5338a [Bugfix][Frontend] Return 500 for unexpected speech failures (#6487)
Signed-off-by: z2z23n0 <51999056+z2z23n0@users.noreply.github.com>
Co-authored-by: Canlin Guo <canlinguosdu@gmail.com>
2026-09-24 01:38:05 -07:00
ZacheryAUandGao Han 926ba72793 [Frontend][Benchmark] Add judger and accuracy metrics for OmniInterac… (#7317)
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-24 08:34:23 +00:00
Haozhe Jiang dbd6a35dd2 [Bugfix] Reject partially bound diffusion LoRA adapters (#8008)
Signed-off-by: Haozhe Jiang <162801044+provoke210@users.noreply.github.com>
2026-09-24 15:28:00 +08:00
wangyuandCursor 3d43571b94 [CI/Build] Cut L3 entrypoint BAGEL cold starts via L4 small-model sleep suites (#6851) (#7359)
Signed-off-by: wangyu <410167048@qq.com>
Signed-off-by: [Your Name] <your.email@example.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-23 22:06:58 -07:00
Honghan ZhuandGao Han 67aa30c966 [Bugfix] Align MiniMax-H3 inference steps with denoiser evaluations (#7219)
Signed-off-by: zhuhh97 <zhuhonghan@huawei.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
2026-09-24 03:56:38 +00:00
wuli666 6daf5b30f0 [Bugfix] Preserve cancellation in duplex session control queues (#7782)
Signed-off-by: wuli666 <421774554@qq.com>
2026-09-23 19:32:40 +00:00
Joshna-Medisetty 5e6b9047c9 [Bugfix] Drain the omni connector output on the main thread in the AR runner (#7750)
Signed-off-by: Joshna Medisetty <joshna.medisetty@intel.com>
2026-09-23 11:36:56 -07:00
akshatvishu b4737d3066 [BugFix] Download Anima assets for README CI test (#7928)
Signed-off-by: akshatvishu <akshatnayak197@gmail.com>
2026-09-23 17:04:51 +00:00
Zhou Taichang d38566ae80 [Bugfix][Core] Fix stage initialization spawn-lock/device-lock inversion (#7926)
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
2026-09-23 17:03:06 +00:00