Commit Graph
11774 Commits
Author SHA1 Message Date
Mick de6a1dbd7a [diffusion] feat: support hybrid conditioning for minimax h3 (#36080) 2026-08-23 21:51:25 +08:00
pllimaxandsglang-npu-bot 8014d9d062 [npu] Kill evalscope session by process group and fix report score parsing (#35988)
Co-authored-by: sglang-npu-bot <sglangnpu@163.com>
2026-08-23 21:31:42 +08:00
Mick 939c00a7e3 [diffusion] feat: support compact qwen3-vl conditioning for minimax h3 (#36076) 2026-08-23 21:25:11 +08:00
MickandClaude Fable 5 886e37a649 [diffusion] CI: guard the anonymous-host budget alongside peak VRAM (#36051)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-08-23 20:13:46 +08:00
Mick dd15fb57b5 [diffusion] feat: automatically infer comfy fp8 activation scaling (#36060) 2026-08-23 18:43:33 +08:00
Sahithi ChigurupatiandMohammad Miadh Angkad 44db041700 [NVIDIA] Fix SM107 MXFP8 activation prep (#35405)
Signed-off-by: Sahithi Chigurupati <chigurupati.sahithi@gmail.com>
Co-authored-by: Mohammad Miadh Angkad <176301910+mmangkad@users.noreply.github.com>
2026-08-23 18:17:03 +08:00
27aa48bca1 [Fix] lfm2 detector: recover tool calls dropped by common model-outpu… (#34237)
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
2026-08-23 18:09:43 +08:00
Mick e3a008a9db [diffusion] UX: clean up startup and offload logs (#36034) 2026-08-23 16:50:25 +08:00
Cheng WanandClaude Opus 5 340391a297 config: publish before the launcher reads effective configuration (#35910)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-23 01:20:20 -07:00
Cheng WanandClaude Opus 5 362c2ee849 config: borrowed-record reads follow the config bags (#35908)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-23 01:19:20 -07:00
Cheng WanandClaude Opus 5 64aa859da2 config: constructing a config no longer resolves it (#35907)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-23 01:18:53 -07:00
Cheng WanandClaude Opus 5 4bc79a1b49 config: project the config bags from the resolution result (#35906)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-23 01:18:24 -07:00
Cheng WanandClaude Opus 5 0e22777572 config: record resolution writes in a declaration stash (#35905)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-23 01:17:27 -07:00
Cheng Wan 6218d6ce3f config: a defensive publish must not re-project over a live process (#35904) 2026-08-23 01:16:39 -07:00
MickandClaude Fable 5 bd3cc97e7e [diffusion] CI: let the 5090 consumer case runs two warm requests on the full recipe (#36032)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-08-23 15:39:32 +08:00
karverma-amd 155aa26c19 [AMD][DSV4] perf: use full 1024-thread block for indexer top-k on ROCm (#36004) 2026-08-23 00:22:30 -07:00
Xinyi Song edd675cecf [AMD] Add Radix-4 MoE top-k router kernel for Kimi-K3 routing (#34490) 2026-08-22 23:57:35 -07:00
Shangming Cai 8270ac4621 refactor(disagg): move _is_watermark_ready into StagingManagerMixin (#36030) 2026-08-23 14:18:41 +08:00
Shangming Cai 849ce71976 refactor(disagg): dedupe mooncake failure_exception into a mixin (#36031) 2026-08-23 14:17:44 +08:00
Shangming Cai bd72190728 refactor(disagg): register SGLANG_ENCODER_MM_LOAD_WORKERS in Envs (#36006) 2026-08-23 12:47:38 +08:00
Mick 97ae27893a [diffusion] feat: release a layerwise component's non-layer weights between uses (#35734) 2026-08-23 12:05:24 +08:00
Mick 70319a0881 [diffusion] feat: support loading serialized comfy convrot int8 native encoders (#36023) 2026-08-23 12:00:38 +08:00
Mick bbbcbf9418 [diffusion] fix: stabilize ltx-2.3 two-stage cold requests (#35997) 2026-08-23 10:04:00 +08:00
MickandClaude Fable 5 59cdc9de7d [diffusion] chore: re-home decode-dtype vae weights to a file-backed store (#35986)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-08-23 09:32:04 +08:00
Mick a36c0746b9 [diffusion] feat: support serialized comfy convrot int8 dits (#35994) 2026-08-23 09:29:59 +08:00
Chungman LeeandMohammad Miadh Angkad 7f30d66045 [Kimi-K3] Fix "wrong grids" crash in DP-sharded vision preprocessing (#35305)
Co-authored-by: Mohammad Miadh Angkad <176301910+mmangkad@users.noreply.github.com>
2026-08-23 09:20:51 +08:00
Hồ Sỹ Thế eec794bce0 [AMD][Spec] Fix aiter GQA packing + split-KV routing in NEXTN spec attention (verify & draft_extend) (#30105) 2026-08-22 14:22:43 -07:00
Shangming Cai cce0a1244b refactor(disagg): hoist staging helper imports out of the bootstrap loops (#35980) 2026-08-23 01:24:58 +08:00
b98d472158 [NPU] [Diffusion] Fix critical Ascend NPU Diffusion regression/bugs & restore 2-NPU CI testcase (#34855)
Co-authored-by: Elizaveta Martirosian <elizabet3000@mail.ru>
Co-authored-by: Arseniy Mironov <98156294+Napkin-AI@users.noreply.github.com>
Co-authored-by: Alexandr <117110413+Allor-maker@users.noreply.github.com>
Co-authored-by: P_Alex_Tr <aleksandr.smyshlaev@yandex.ru>
2026-08-22 18:32:20 +03:00
Mick 0b064e3739 [diffusion] comfyui: add a minimax-h3 node and a generic extra-fields passthrough (#35352) 2026-08-22 23:11:49 +08:00
Mick 453b98c490 [diffusion] feat: support single-file component weight overrides (#35979) 2026-08-22 23:09:35 +08:00
Mick 46cb12ab45 [diffusion] fix: fix hunyuan3d stale extension lock hangs (#35989) 2026-08-22 23:08:44 +08:00
d315eb7250 [AMD] DeepSeek-V4: add aiter fused mHC post+pre with cross-layer boundary dispatch (#32577)
Co-authored-by: 1am9trash <1am9trash@gmail.com>
Co-authored-by: HAI <hixiao@gmail.com>
2026-08-22 07:34:57 -07:00
MickandClaude Opus 5 db570fe619 [diffusion] chore: let the auto policy select h3's dit for layerwise offload (#35812)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-22 21:36:46 +08:00
MickandClaude Fable 5 61981e1fcd [diffusion] optimization: keep vae decoder weights in their decode dtype from load (#35967)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-08-22 21:31:06 +08:00
MickandClaude Opus 5 a8c16b2e55 [diffusion] feature: use the directory for the vae mapping gate (#35946)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-22 19:12:48 +08:00
Mick 489e605b35 [diffusion] feat: admit compatible quantized native encoders (#35962) 2026-08-22 18:51:01 +08:00
MickandClaude Opus 5 382343f860 [diffusion] optimization: transfer mapped layers through a courier thread (#35882)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-08-22 17:17:14 +08:00
siyuandliusy58 cb10ca16dd [FEAT] Weight Daemon abstraction (#33279)
Co-authored-by: liusy58 <liusy58@smail.nju.edu.cn>
2026-08-22 02:13:46 -07:00
Shangming Cai 5c03069d4b refactor(disagg): drop dead placeholder overrides in Common KV sender/receiver (#35950) 2026-08-22 15:45:18 +08:00
Shangming Cai 15a4398320 refactor(disagg): hoist duplicated _handle_staging_req into a mixin (#35948) 2026-08-22 15:35:55 +08:00
Mick b391ef171f [diffusion] feat: load serialized bnb4 components with transformers (#35945) 2026-08-22 15:33:53 +08:00
Mohammad Miadh Angkad 90354326c7 [VLM] Split Pixtral multi-image features before the CUDA IPC wrap (#35463) 2026-08-22 15:17:55 +08:00
Lianmin ZhengandYichao Fu 6fd0384d42 Make draft attention backends extensible (#35932)
Co-authored-by: Yichao Fu <yichaofu@meta.com>
2026-08-21 23:54:44 -07:00
Jianfei Wangandmiraclezqc af39ad9349 [Model] Complete dots.note.omni support with native encoders, video preprocessing, and MTP decoding (#33829)
Co-authored-by: miraclezqc <dysania@pku.edu.cn>
2026-08-22 14:19:14 +08:00
Zhiqiang Xie c35683fda0 [HiCache] Clamp tombstoned SWA locs in UnifiedSWAKVPool translation (#35933) 2026-08-21 23:08:56 -07:00
Xiaoyu Zhang 96bfd2476c [diffusion] Enable SANA-Video breakable CUDA graphs (#35729) 2026-08-22 12:57:09 +08:00
Xiaoyu Zhang 83e9ece672 [diffusion] Fuse SANA-Video interleaved RoPE (#35695) 2026-08-22 12:56:45 +08:00
Mick 5290327025 [diffusion] feat: resolve hub component subfolders (#35939) 2026-08-22 12:01:44 +08:00
Shangming Cai ac179eec11 fix(disagg): PD transfer-failure injection was silently inert (#35890) 2026-08-22 11:51:16 +08:00