 Roberto L. Castroandvllm-agent
|
a2ad4b74b9
|
[Perf] Enable DeepGEMM sparse MQA logits on the FP8 indexer cache, DSv4.1 (#62011)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 19:25:19 +00:00 |
|
|
wangxiyuan
|
ca8dbccf20
|
[Bugfix] Forward launch grid to Triton kernel overrides (#63999)
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
|
2026-10-27 14:46:07 +00:00 |
|
|
Nicolò Lucchesi
|
939ff73503
|
[Perf] Drop BasevLLMParameter's Python __torch_function__ (#63785)
Signed-off-by: NickLucche <nicolo.lucchesi@mistral.ai>
|
2026-10-27 15:30:25 +01:00 |
|
|
|
c3941c032d
|
[Bugfix][ROCm] Match the DeepSeek-V4.1 FP8 config in the shared-expert FSE compatibility check (#63099)
Signed-off-by: BaoYunkai <269084791+BaoYunkai@users.noreply.github.com>
Co-authored-by: BaoYunkai <269084791+BaoYunkai@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Shanshan Shen <467638484@qq.com>
|
2026-10-26 18:21:24 +08:00 |
|
|
|
5a224c1ef3
|
[Fast Start] Key the weight cache on MoE and linear kernel backends (#63506)
Signed-off-by: liusy58 <mg21330037@smail.nju.edu.cn>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-26 04:06:24 +00:00 |
|
|
|
2383ff0014
|
[Fast Start] Detect weight cache daemon exit in zero-copy engines (#62734)
Signed-off-by: liusy58 <mg21330037@smail.nju.edu.cn>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-26 04:04:28 +00:00 |
|
|
Pablo LlopisandNicolò Lucchesi
|
2c880f1e68
|
[Bugfix] Preserve FP8 indexer WK pairs across load_weights calls (#59223)
Signed-off-by: Pablo Llopis <pablo.llopis@epfl.ch>
Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@mistral.ai>
|
2026-10-25 19:42:48 +00:00 |
|
|
Chongmin Liandvllm-agent
|
b3466364b6
|
[Bugfix] Release cached cuBLAS workspaces after daemon warmup (#63451)
Signed-off-by: lichongmin <chongminli.cn@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-25 15:12:42 +00:00 |
|
|
|
94a29d316e
|
[Fast Start] Add daemon-to-daemon warmup cache seeding for Preload (#63088)
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Xun Sun <UNIDY2002@outlook.com>
Co-authored-by: liusy58 <mg21330037@smail.nju.edu.cn>
Co-authored-by: Kimi Code <noreply@moonshot.cn>
|
2026-10-25 21:03:40 +08:00 |
|
|
aoshen02andClaude Opus 5.5
|
7dc44a0016
|
[Bugfix][LoRA][MoE] Process reloaded weights of LoRA-wrapped MoE layers (#63319)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-24 00:01:24 +00:00 |
|
|
|
258f42076d
|
[Fast Start] Fix weight cache daemon GPU memory and warm-cache warmup hang (#62583)
Signed-off-by: liusy58 <mg21330037@smail.nju.edu.cn>
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
|
2026-10-23 04:51:57 +00:00 |
|
|
|
2d40c0609f
|
[ROCm][Perf] Triton port of the fused GDN MTP decode kernel (#61855)
Signed-off-by: iseeyuan <myuan@meta.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Simon Danielsson <70206058+simondanielsson@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-10-23 10:41:07 +08:00 |
|
|
|
271941177d
|
[Perf] Bypass DispatchPooler grouping for homogeneous tasks (#62940)
Signed-off-by: Jakub Zakrzewski <jzakrzewski@nvidia.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-22 22:40:37 +08:00 |
|
|
Ronald
|
12c01a22eb
|
[MoE] Revert platform hook for routed-expert capture rows (#62624) (#63013)
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
|
2026-10-22 16:26:33 +08:00 |
|
|
Chauncey
|
124501da07
|
[Bugfix] Load Engram tables on CPU in the InstantTensor loader (#62692)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
|
2026-10-22 07:23:51 +00:00 |
|
|
Hao WangandCodex
|
90c4a1a73e
|
[Bugfix] Preserve image normalization constants during dummy loading (#62767)
Signed-off-by: Hao Wang <haowan@nvidia.com>
Co-authored-by: Codex <noreply@openai.com>
|
2026-10-21 18:34:23 +00:00 |
|
|
|
979be69b2c
|
[Bugfix] Include KV cache quantization settings in IPC weight cache key (#61758)
Signed-off-by: bzsuni <bingzhe.sun@daocloud.io>
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-21 03:19:26 +00:00 |
|
|
Ronald
|
2a1a3c56c9
|
[Feature][MoE] Add platform hook for routed-expert capture rows (#62624)
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
|
2026-10-21 10:10:03 +08:00 |
|
|
|
cc8a41f328
|
[MyPy][4/N] Fix mypy errors in medium-hard tests/ dirs (#55485)
Signed-off-by: Satvik Shrivas <44926681+theSatvik@users.noreply.github.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-10-20 04:51:15 -07:00 |
|
|
|
f85e39ea01
|
[ROCm][Quantization][MOE] Enable fused shared experts for MXFP8 checkpoints (#57916)
Signed-off-by: Sergei Shliapnikov <sergei.shliapnikov@amd.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-20 01:15:59 -07:00 |
|
|
aoshen02andClaude Opus 5.5
|
b44446c180
|
[Quantization] replace_parameter keeps the parameter object by default (#62326)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-20 02:16:31 -04:00 |
|
|
Isotr0pyandCyrus Leung
|
3b05805d47
|
[Misc] Fix failing weight cache RoPE test on main (#62508)
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
|
2026-10-19 11:45:37 -07:00 |
|
|
stefankoncarevic
|
aa3564ceef
|
[Bugfix][Test] Stop requesting capfd in the subprocess weight cache test (#62461)
Signed-off-by: Stefan Koncarevic <stefan.koncarevic@amd.com>
|
2026-10-19 12:00:21 -05:00 |
|
|
bzsuniandvllm-agent
|
2aa5080fa0
|
[Bugfix][Weight Cache] Include RoPE parameters in IPC cache key (#60994)
Signed-off-by: bzsuni <bingzhe.sun@daocloud.io>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-19 10:07:39 -04:00 |
|
|
|
ffa18e55a4
|
[ROCm][DSv4] Require explicit MXFP4 opt-in for MegaMoE shared-expert fusion (#61345)
Signed-off-by: Yanfei Wang <yanfei.wang@amd.com>
Co-authored-by: Shanshan Shen <467638484@qq.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-19 20:31:00 +08:00 |
|
|
nien-hui
|
d46cb9a6d4
|
[Bugfix][Weight Cache] Key non-MoE DP daemons as DP=1 like the engine (#60179)
Signed-off-by: paul <paulqwer.cs13@nycu.edu.tw>
|
2026-10-19 17:31:22 +08:00 |
|
|
|
b9c48c98de
|
[Fast Start] support ep (#55490)
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Xun Sun <UNIDY2002@outlook.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
|
2026-10-19 05:47:33 +00:00 |
|
|
Fadi Arafehandvllm-agent
|
556ac15eb9
|
[Bugfix] Remove extra args from CPU's compute slot mapping kernel (#61992)
Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-17 12:11:33 +08:00 |
|
|
|
71c20f73f2
|
[FlashInfer] Migrate autotune warmup to flashinfer.autotune_v2 managed cache (#55671)
Signed-off-by: Yang Xu <yanxu@nvidia.com>
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
Signed-off-by: YangXu <yanxu@nvidia.com>
Co-authored-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
Co-authored-by: Kimi Code <noreply@moonshot.cn>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Misha Goin <mgoin64@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-17 02:08:28 +00:00 |
|
|
Ayushman Singh
|
de7eeaa87b
|
[Perf] Share Triton JIT warmup compilation pool across registered kernels (#61422)
|
2026-10-16 08:59:42 -04:00 |
|
|
|
a8be1500ba
|
[Bugfix][MoE] Restore TP-sharded routed-expert capture for DP=1 (#61388)
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
Signed-off-by: Ao Shen <281812371+aoshen02@users.noreply.github.com>
Signed-off-by: Ao Shen <aoshen@inferact.ai>
Co-authored-by: Ao Shen <281812371+aoshen02@users.noreply.github.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: Ao Shen <aoshen@inferact.ai>
|
2026-10-16 00:21:24 +00:00 |
|
|
|
f8460ef236
|
[Kernel][DSv4.1] Bump FlashMLA to the 2026.09.30 release and its interleaved V4.1 KV cache (#61371)
Signed-off-by: juntianl <juntianl@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Kimi Code <noreply@moonshot.ai>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
|
2026-10-15 13:58:25 -07:00 |
|
|
|
4d58fa254a
|
[MoE] Unify DeepGEMM MegaMoE experts behind one feature-matrix layer; add NVFP4 (#61538)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
|
2026-10-15 19:02:01 +00:00 |
|
|
Mikko TukiainenandCursor
|
47fd76e4ce
|
[ROCm][Perf] Route softmax MoE top-k, including fused shared-expert sigmoid, through AITER topk_gating (#56226)
Signed-off-by: Mikko Tukiainen <Mikko.Tukiainen@amd.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
|
2026-10-15 17:53:18 +00:00 |
|
|
Matthias Gehre
|
011fa68ca9
|
[ROCm][Perf] Pad fp16/bf16 GEMM weight row stride for L2 cache (#55090)
Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
|
2026-10-15 14:26:32 +00:00 |
|
|
|
b965ee564d
|
[Chore] Use upstream Transformers DeepseekV4Config (#61392)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
|
2026-10-15 01:08:08 -07:00 |
|
|
Haiyu ZuoandKevin H. Luu
|
ff5a91a456
|
[Feature][Model Loader] Apply model-owned MTP shard selection to RunAI ModelStreamer (#60759)
Signed-off-by: 浩韵 <zuohaiyu.zhy@alibaba-inc.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
|
2026-10-15 12:23:11 +08:00 |
|
|
amd-xavierwangandJartX
|
06265beb25
|
[ROCm] Enable fp8 query for Triton attention (#60296)
Signed-off-by: xawang <xawang@amd.com>
Co-authored-by: JartX <sagformas@epdcenter.es>
|
2026-10-14 15:03:20 -04:00 |
|
|
|
13dbf96093
|
[Model Loader] Apply EP weight filtering in InstantTensor (#61167)
Signed-off-by: Itay Etelis <92247226+Etelis@users.noreply.github.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Misha Goin <mgoin64@gmail.com>
|
2026-10-14 09:10:10 -07:00 |
|
|
Ronaldandaoshen02
|
19aba3e4d5
|
[Bugfix][RL] Load attention sinks through weight_loader for layerwise reload (#57797)
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
|
2026-10-14 10:17:58 +00:00 |
|
|
|
6849ca696a
|
[Quantization] Add ModelOpt IQ2_XS, IQ2_XXS, and Q8_0 support via b12x (#59604)
Signed-off-by: Andrii Skliar <askliar@nvidia.com>
Signed-off-by: Luke Alonso <lalonso@gmail.com>
Co-authored-by: Andrii Skliar <askliar@nvidia.com>
Co-authored-by: Luke Alonso <lalonso@gmail.com>
|
2026-10-13 09:06:33 -07:00 |
|
|
Woosuk KwonandClaude Opus 5.5
|
ac71b17517
|
[Model Loader] Keep norm layers neutral under dummy weight init (#61323)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-13 08:54:24 -07:00 |
|
|
|
253e02b95e
|
[Perf] Warm up DeepGEMM MegaMoE kernels before serving (#60468)
Signed-off-by: Simon Veitner <sveitner@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
|
2026-10-13 20:17:48 +08:00 |
|
|
Linze Shi
|
e6e84f350c
|
[Feature][Pooling] Add offline fixed-length late chunking (#60164)
Signed-off-by: Linze-Shi <linzeshi0@gmail.com>
|
2026-10-13 11:16:21 +00:00 |
|
|
Yongye ZhuandClaude Opus 5.5
|
a6009ea85b
|
[Kernel][Build] Bump DeepGEMM to the 26/09/30 release and adapt the sparse indexer (#61149)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
|
2026-10-12 23:24:43 -07:00 |
|
|
Woosuk KwonandClaude Opus 5.5
|
f47792af7f
|
[Model Loader] Make dummy weight init dtype-aware (#61134)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-12 18:33:26 -07:00 |
|
|
|
189bbfb9f4
|
Add MoNe experts pruning support (#50415)
Signed-off-by: Ilya Boytsov <ieboytsov@nebius.com>
Signed-off-by: Misha Goin <mgoin64@gmail.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Ilya Boytsov <ieboytsov@nebius.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Misha Goin <mgoin64@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-12 23:18:43 +00:00 |
|
|
|
092c35dd15
|
[Model] Support Cloudflare/clef-omni on /v1/systemone (#61064)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
|
2026-10-12 21:27:34 +08:00 |
|
|
Nick Hill
|
3fe917abb1
|
[Bugfix][MRV2] Fix FlashInfer attention and ReplaySSM warmups on MRV2 (#61013)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
|
2026-10-11 16:04:41 -07:00 |
|
|
tc-mbandCyrus Leung
|
c85616efc6
|
[Model] Support MiniCPM-V 4.7 (#58674)
Signed-off-by: tc-mb <tianchi_cai@icloud.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
|
2026-10-11 10:21:10 +00:00 |
|