 Jesus FedericoandClaude Opus 4.6
|
b87575d24b
|
feat: add logit_scale to PoolerConfig for affine score calibration (#39435)
Signed-off-by: Jesus Federico <jefp@amazon.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
2026-04-10 17:21:14 +00:00 |
|
|
Jee Jee Li
|
ecd1ea1363
|
[Kernel] Porting the TRTLLM minimax_allreduce_rms kernels (#37045)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
|
2026-04-11 00:20:20 +08:00 |
|
|
wang.yuqi
|
cb5f7501cb
|
[New Model]: jinaai/jina-reranker-v3 (#38800)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
|
2026-04-10 15:20:40 +00:00 |
|
|
Hexiang Wangandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
1dfd64c1cc
|
[PluggableLayer][3/N] Apply PluggableLayer to llm_head and vocab embedding layer (#33465)
Signed-off-by: whx-sjtu <2952154980@qq.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-04-10 16:12:59 +08:00 |
|
|
Richard Zou
|
f44afef6d6
|
[compile] Allow strings in custom ops without regressing compilation times (#38123)
Signed-off-by: Richard Zou <zou3519@gmail.com>
|
2026-04-10 07:26:37 +00:00 |
|
|
Kunshang Ji
|
55d037e2e5
|
[CT][FP8][Marlin] refactor CompressedTensorsW8A16Fp8 to use kernel abstraction (#38244)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
|
2026-04-10 09:58:35 +08:00 |
|
|
Ibrahim Arshad
|
9853a3c159
|
fix(gdn): Align prefill warmup with real prefill path (#39169)
Signed-off-by: Ibrahim Arshad <38925737+ibrahim1023@users.noreply.github.com>
|
2026-04-10 00:49:50 +00:00 |
|
|
Michael Goin
|
2800706f06
|
[Refactor] Move NVFP4 GEMM management into NvFp4LinearKernel (#39129)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-04-09 15:05:36 -04:00 |
|
|
PikaPikachu
|
827268e98d
|
[Quantization] Support Quark W8A8 INT8 MoE inference (#36320)
Signed-off-by: kangletian <Letian.Kang@amd.com>
|
2026-04-09 17:24:43 +00:00 |
|
|
Richard Zou
|
6c749399b7
|
[BugFix] fix tests/kernels/moe/test_moe_layer.py (#39404)
Signed-off-by: Richard Zou <zou3519@gmail.com>
|
2026-04-09 08:48:59 -04:00 |
|
|
Yongye Zhu
|
2e98406048
|
[Refactor] Improve indexer decode path metadata preparation (#38865)
|
2026-04-08 20:49:15 -07:00 |
|
|
Wei ZhaoandJee Jee Li
|
92fbec391b
|
[Bug] Fix routing bias dtype for trtllm per-block fp8 moe (#38989)
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
|
2026-04-08 19:42:43 -07:00 |
|
|
Maral
|
2e9034c998
|
[W8A8 Block Linear Refactor][2/N] Remove W8A8Fp8BlockLinearOp and adopt Fp8 block linear kernel selections. (#33892)
Signed-off-by: maral <maralbahari.98@gmail.com>
Signed-off-by: Maral <maralbahari.98@gmail.com>
|
2026-04-09 08:50:39 +08:00 |
|
|
Benjamin Chislettandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
8332078cfd
|
[Bugfix] FlashInfer MXINT4 MoE crashes, missing do_finalize (#39315)
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
2026-04-08 20:36:33 -04:00 |
|
|
Wentao Ye
|
2018137242
|
[Feature] Batch invariant nvfp4 linear support (#39322)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-04-08 16:29:13 -04:00 |
|
|
|
a776a48b1c
|
[MoE] Move DEEP_GEMM into experts/ subdirectory (#39005)
Signed-off-by: Jackmin801 <ongjackm@gmail.com>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-08 19:23:08 +00:00 |
|
|
|
b55d830ec7
|
[Perf][Kernel] Persistent TopK scheduler: unified CUDAGraph-safe kernel with dynamic per-row dispatch - DeepSeek-V3.2 DSA decode (#37421)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
|
2026-04-08 13:35:57 -04:00 |
|
|
rasmith
|
78434b923c
|
[CI][AMD][BugFix][Kernel] Cast induction variable to int64 on MI350 for chunk_gated_delta_rule_fwd_kernel_h_blockdim64 to avoid illegal memory access (#39087)
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
|
2026-04-08 16:57:18 +08:00 |
|
|
Andrey Talman
|
2111997f96
|
[release 2.11] Update to torch 2.11 (#34644)
|
2026-04-07 18:55:48 -07:00 |
|
|
zofiaandKunshang Ji
|
ad3304425b
|
[XPU] add xpu backend implementation of mxfp8 quant (#38682)
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-04-08 08:30:35 +08:00 |
|
|
Lucas Wilkinson
|
70406eb1dc
|
[Attention][V0 Deprecation] Deprecate accept output buffer (#39125)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
|
2026-04-07 17:14:58 -04:00 |
|
|
rasmith
|
83d09d36b5
|
[CI][Bugfix][AMD][ Ensure weights created when using emulating OCP MXFP4 (#36993)
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
|
2026-04-08 00:37:16 +08:00 |
|
|
Chendi.XueandKunshang Ji
|
92b9afeecd
|
[XPU] Quick fix for TritonMLA to remove cuda hardcode (#39088)
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-04-08 00:17:58 +08:00 |
|
|
Jinzhen Lin
|
7310555482
|
[Bugfix] Fix marlin nvfp4 rescaling (#37502)
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
|
2026-04-07 08:57:17 -07:00 |
|
|
kkyyxhll
|
98e1a43af7
|
[Bugfix][Quantization] Fix PerTensorScale loading with tuple shard_id in MergedColumnParallelLinear (#38517)
Signed-off-by: loukang <loukang@xiaohongshu.com>
|
2026-04-07 11:16:26 -04:00 |
|
|
Wei Zhao
|
0be9516ea4
|
[Bug] Fix Trtllm Fp8 MoE Weight Shuffle Memory Fragamentation (#39054)
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
|
2026-04-07 08:04:08 -04:00 |
|
|
Jiangyun ZhuandLuka Govedič
|
8060bb0333
|
[vLLM IR] rework gemma_rms_norm (#39014)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Signed-off-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
|
2026-04-07 01:37:00 -07:00 |
|
|
Andreas Karatzas
|
2df2c85be4
|
[Kernels][MoE] Fix legacy_routing to use bitmatrix-based routing path (#38504)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-04-07 10:57:09 +08:00 |
|
|
bnellnm
|
b2b2c5239e
|
[MoE Refactor] Split up compressed_tensors_moe.py (#38960)
Signed-off-by: Bill Nell <bnell@redhat.com>
|
2026-04-06 20:07:54 -04:00 |
|
|
fxmarty-amdandKyle Sayers
|
00d7b497b3
|
[NVFP4] Support NVFP4 dense models from modelopt and compressed-tensors on AMD Instinct MI300, MI355X and Hopper through emulation (#35733)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
Signed-off-by: fxmarty-amd <felmarty@amd.com>
Co-authored-by: Kyle Sayers <kylesayrs@gmail.com>
|
2026-04-06 16:18:27 -06:00 |
|
|
|
e8ebbdde83
|
[Quantization] Add FlashInfer CuteDSL batched experts backend for NVFP4 MoE (#38251)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
|
2026-04-06 11:57:53 -07:00 |
|
|
namgyu-youn
|
94fbb09894
|
[EASY] Drop duplicate KV-cache initialization (#38799)
Signed-off-by: namgyu-youn <namgyu.dev@gmail.com>
|
2026-04-06 18:05:39 +00:00 |
|
|
bnellnmandRobert Shaw
|
f01482408c
|
[MoE Refactor][Test] FusedMoE layer test (#24675)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-06 17:17:23 +00:00 |
|
|
bnellnmandRobert Shaw
|
93bada494f
|
[MoE Refactor] Split of DefaultMoERunner class (#35326)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-06 12:41:59 -04:00 |
|
|
Wei ZhaoandRobert Shaw
|
1af6f78ae5
|
[Perf] Change Trtllm fp8 MoE to use Shuffled Weights and BlockMajorK Layout (#38993)
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-05 10:54:31 -04:00 |
|
|
|
228023b3a5
|
[Bugfix][MoE] Fix 6-8% decode regression: prefer multi-stream shared expert overlap (#38990)
Signed-off-by: Martin Vit <martin@voipmonitor.org>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-05 10:28:31 -04:00 |
|
|
Robert ShawandRobert Shaw
|
968ed02ace
|
[Quantization][Deprecation] Remove Petit NVFP4 (#32694)
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
|
2026-04-05 00:07:45 +00:00 |
|
|
Robert Shaw
|
7d266abb22
|
Revert "[vLLM IR] gemma_rms_norm" (#38998)
|
2026-04-04 17:48:08 -04:00 |
|
|
Xiaoshuang Wang
|
156405d243
|
[vLLM IR] gemma_rms_norm (#38780)
Signed-off-by: Icey <1790571317@qq.com>
|
2026-04-04 13:55:52 -04:00 |
|
|
Artem Perevedentsevandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
99e5539a67
|
[Perf][GDN] Align TMA usage with upstream FLA (#38981)
Signed-off-by: Artem Perevedentsev <aperevedents@nvidia.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
2026-04-05 00:38:02 +08:00 |
|
|
LinkunandLuka Govedič
|
a88ce94bbb
|
[IR][RmsNorm] pass None if not has_weight (#38961)
Signed-off-by: Linkun Chen <github@lkchen.net>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
|
2026-04-04 11:02:30 -04:00 |
|
|
yzong-rh
|
a5a623d961
|
[Bugfix] Re-enable Renormalize routing for TRT-LLM MoE experts (#38859)
Signed-off-by: Yifan Zong <yzong@redhat.com>
|
2026-04-04 01:48:17 +08:00 |
|
|
Vasiliy Kuznetsov
|
7b1a7423be
|
[Frontend] new online quantization frontend (#38138)
Signed-off-by: Vasiliy Kuznetsov <vasiliy@meta.com>
|
2026-04-03 11:58:39 -04:00 |
|
|
Yusuf Mohammadandyusuf
|
46f02e00f2
|
[Bugfix] Fix AWQ models batch invariance issues (#38670)
Signed-off-by: yusuf <yusuf@deeplearningmachine.mynet>
Signed-off-by: <>
Co-authored-by: yusuf <yusuf@deeplearningmachine.mynet>
|
2026-04-03 14:54:15 +00:00 |
|
|
Qiming ZhangandKunshang Ji
|
6b4872240f
|
[XPU] bump up xpu-kernel v0.1.5, transpose moe weights (#38342)
Signed-off-by: mayuyuace <qiming1.zhang@intel.com>
Signed-off-by: Qiming Zhang <qiming1.zhang@intel.com>
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-04-03 14:10:02 +00:00 |
|
|
Artem Perevedentsev
|
cb10b7e80b
|
[GDN] Eliminate GPU->CPU sync in prepare_chunk_indices during prefill (#38361)
Signed-off-by: Artem Perevedentsev <aperevedents@nvidia.com>
Signed-off-by: Vadim Gimpelson <156319763+vadiklyutiy@users.noreply.github.com>
|
2026-04-03 13:38:02 +00:00 |
|
|
|
bf8b022e60
|
[Intel][Triton] Support round_int8 for Intel backend (#38825)
Signed-off-by: Mieszko Dziadowiec <mdziadowiec@habana.ai>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Stefano Castagnetta <scastagnetta@nvidia.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Stefano Castagnetta <scastagnetta@nvidia.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-04-03 20:47:35 +08:00 |
|
|
Bowen BaoandRobert Shaw
|
103f0de565
|
[ROCm][Quantization][1/N] Refactor quark_moe w_mxfp4 w/ oracle (#38774)
Signed-off-by: Bowen Bao <bowenbao@amd.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-03 03:29:57 +00:00 |
|
|
Carl YandClaude Opus 4.6
|
1f5ec2889c
|
[mla] Support fused FP8/NVFP4 output quantization in MLA attention (#35792) (#36205)
Signed-off-by: Carl You <4531192+carlyou@users.noreply.github.com>
Signed-off-by: Carl Y <4531192+carlyou@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
|
2026-04-02 21:16:11 -04:00 |
|
|
|
ee3cf45739
|
[XPU] Initial support for GDN attention on Qwen3-next/Qwen3.5 (#33657)
Signed-off-by: Yan Ma <yan.ma@intel.com>
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Chendi Xue <chendi.xue@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-04-03 08:59:11 +08:00 |
|