|
d6247d7173
|
[Spec Decode][Perf] Replicate DSpark Markov head across TP ranks (#49731)
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
|
2026-07-29 11:43:51 -04:00 |
|
|
|
ad5d29db70
|
[Model] Support Qwen3.5 text-only dense and MoE models (#50210)
Signed-off-by: Perkz Zheng <PerkzZheng@users.noreply.github.com>
Co-authored-by: Perkz Zheng <PerkzZheng@users.noreply.github.com>
|
2026-07-29 08:21:57 +00:00 |
|
|
+13
|
7c6729b769
|
[Model] Add Kimi K3 support: model files and kernels [1/N] (#50089)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Ziming Huang <zelda.huanghuang@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Isotr0py <mozf@inferact.ai>
Co-authored-by: aoshen02 <aoshen02@users.noreply.github.com>
Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Jee Jee Li <jeejeelee@inferact.ai>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Summer Yang <girasoleyang@gmail.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
Co-authored-by: Bowen Wang <abmfy@icloud.com>
Co-authored-by: gnovack <novackgm@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: xiaozhoupy <peiyuanzhou1994@gmail.com>
Co-authored-by: Roy Wang <yasong.wang@inferact.ai>
Co-authored-by: Jeff (Junze) Ma <93145857+majunze2001@users.noreply.github.com>
Co-authored-by: Codex <codex@openai.com>
|
2026-07-29 14:10:58 +08:00 |
|
|
|
6f00a1ae3b
|
fused_moe: add VLLM_TRITON_USE_TD tensor-descriptor path (#42436)
Signed-off-by: Artur Fierka <artur.fierka@intel.com>
Signed-off-by: Lena Onyshchenko <162571002+oonyshch@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Lena Onyshchenko <162571002+oonyshch@users.noreply.github.com>
|
2026-07-29 13:15:10 +08:00 |
|
|
|
32a423ac0a
|
Integrate CuTeDSL MoE for ReLU2 NVFP4 (#49580)
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
|
2026-07-28 19:46:38 -07:00 |
|
|
|
30c2718eaa
|
[CompressedTensors] FP4 Qutlass Integration (#43229)
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
Signed-off-by: Brian Dellabetta <bdellabe@redhat.com>
Signed-off-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>
Co-authored-by: Brian Dellabetta <bdellabe@redhat.com>
Co-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>
Co-authored-by: Dipika Sikka <dipikasikka1@gmail.com>
|
2026-07-28 20:34:20 -06:00 |
|
|
fxmarty-amdandGitHub
|
5369f7b7b8
|
[MXFP8][ROCm] Fix MXFP8 MoE backend selection (#49747)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
|
2026-07-29 07:57:20 +08:00 |
|
|
Julien DebacheandGitHub
|
bb3b61f2fd
|
perf: dispatch non-grouped bias-less topk routing methods to fused path (#49618)
Signed-off-by: jdebache <jdebache@nvidia.com>
|
2026-07-28 14:57:22 -07:00 |
|
|
labAxiaomingandGitHub
|
1db989bbf1
|
[Bugfix][Multimodal] Fix video temporal padding estimates (#49030)
Signed-off-by: xiaoming <1259730330@qq.com>
|
2026-07-29 01:57:31 +08:00 |
|
|
Brian DellabettaandGitHub
|
8a7b3c2990
|
[compressed-tensors] update find_matched_target order to prioritize fused name matches over class match (#49483)
Signed-off-by: Brian Dellabetta <bdellabe@redhat.com>
|
2026-07-28 17:03:08 +00:00 |
|
|
Andreas KaratzasandGitHub
|
05a0814863
|
[ROCm] Fix and optimize GPT-J-style MRoPE (#49906)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-28 10:50:10 -06:00 |
|
|
|
4f56321d7e
|
[ROCm] Cache fp32 upcast of static e8m0 weight scale in AITER scaled_mm (#47773)
Signed-off-by: jiacao-amd <jiahui.cao@amd.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-07-28 11:34:16 -05:00 |
|
|
|
ba702e978e
|
[Attention] Skip sparse indexer scoring for dense short prefills (#48407)
Signed-off-by: Yiliu Dong <91178480+qianlihuang@users.noreply.github.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
|
2026-07-28 16:17:22 +00:00 |
|
|
Nick HillandGitHub
|
0d0504b54c
|
[Core] Warm up runner-owned Triton kernels before the first request (#49903)
|
2026-07-28 07:58:02 -07:00 |
|
|
|
948107acf7
|
[Bugfix] Enhance extra_config handling for layer name suffix matching (#48589)
Signed-off-by: Xin He <xin3.he@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-07-28 20:37:55 +08:00 |
|
|
Itay AlroyandGitHub
|
35efdf6b34
|
[Elastic EP] Async preparation (#47288)
Signed-off-by: Itay Alroy <ialroy@nvidia.com>
|
2026-07-28 05:19:13 -07:00 |
|
|
|
03a2d03367
|
[Bugfix] Respect cgroup memory limits on all platforms (#49966)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-28 14:38:20 +08:00 |
|
|
afriedriandGitHub
|
f472ab0a4c
|
Remove triton per group quant [ROCm] [Bugfix] (#49621)
Signed-off-by: Andy Friedrich <afriedri@amd.com>
|
2026-07-28 05:53:46 +00:00 |
|
|
Nick HillandGitHub
|
d223c900d8
|
[Bugfix] Only pad transformers backend value when it is narrower (#50060)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
|
2026-07-27 23:36:45 -05:00 |
|
|
|
33fe71a4d3
|
[AMD] Revert Mxfp4MoeBackend.TRITON_UNFUSED fallback (#46491)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
Co-authored-by: Felix Marty <Felix.Marty@amd.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-27 22:31:23 -05:00 |
|
|
nvbfalkandGitHub
|
60915c972c
|
[Feature] Add VidCom2 video token pruning (#47750)
Signed-off-by: Benedikt Falk <bfalk@nvidia.com>
|
2026-07-28 03:20:17 +00:00 |
|
|
|
7aea73d83d
|
[ROCm][Quark][6/N] Use MXFP4 linear kernel abstraction for aiter backend (#49348)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-27 21:35:57 -05:00 |
|
|
Netanel HaberandGitHub
|
1e34a13539
|
Fix Humming non-gated MoE (#49096)
|
2026-07-27 22:56:44 +00:00 |
|
|
|
ebcef33766
|
Fix MQA with tensor parallelism on transformers modeling backend (#49987)
Signed-off-by: microslaw <milosz.grunwald@intel.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-07-27 22:51:17 +00:00 |
|
|
|
28158b2fc3
|
[ROCm] [BugFix] Fix Quark GLM-5.2 Checkpoint inference: indexer wk per-channel FP8 dequant + missing sparse-MLA metadata fields (#48886)
Signed-off-by: Colin Zeng <Colin.Zeng@amd.com>
Signed-off-by: ColinZ22 <Colin.Zeng@amd.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
Co-authored-by: fanxingran <xingran.fan@amd.com>
|
2026-07-27 22:38:22 +00:00 |
|
|
|
1053e248f0
|
[ROCm][Quantization][5/N] Refactor quark_moe w8a8-int8 w/ oracle (#46765)
Signed-off-by: amd-sourjya <amd-sourjya@users.noreply.github.com>
Co-authored-by: amd-sourjya <amd-sourjya@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-27 16:01:34 -05:00 |
|
|
Wentao YeandGitHub
|
b5bcb3ce88
|
[Refactor] Remove dead code in multiple files (#49745)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-07-27 15:58:26 -04:00 |
|
|
|
ed13deb376
|
[Bugfix][CPU] Fall back to torch for unaligned swigluoai on NEON/vec MoE (#49985)
Signed-off-by: oops-oom <73481342@qq.com>
Co-authored-by: oops-oom <73481342@qq.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
2026-07-27 18:35:57 +00:00 |
|
|
|
99de48e98f
|
Fix MLA padding and grouped topk routing in the Transformers modelling backend (#49982)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
|
2026-07-27 18:32:32 +00:00 |
|
|
Guan-Ming ChiuandGitHub
|
ef9975d021
|
[Bugfix] Reject pipeline parallelism for DiffusionGemma (#45828)
Signed-off-by: Guan-Ming (Wesley) Chiu <105915352+guan404ming@users.noreply.github.com>
|
2026-07-27 14:37:54 +00:00 |
|
|
Roberto L. CastroandGitHub
|
56c96b0d91
|
[Perf] Tune LL BF16 Router GEMM (#48774)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
|
2026-07-27 10:25:37 -04:00 |
|
|
Rui "Garry" GaoandGitHub
|
dbccc5ae32
|
[Model] Enable EVS for Qwen3.5 (#48912)
Signed-off-by: Rui "Garry" Gao <garrygaogg@gmail.com>
|
2026-07-27 13:42:35 +00:00 |
|
|
neweyesandGitHub
|
96fa3f42c9
|
[Perf] Skip ll_bf16 router GEMM warmup for non-MoE models (#49659)
Signed-off-by: neweyes <328719365@qq.com>
|
2026-07-27 05:16:42 -07:00 |
|
|
Harry MellorandGitHub
|
92e8518d37
|
Improve Transformers modelling backend fx tracer (#49957)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-07-27 12:51:16 +01:00 |
|
|
Andreas KaratzasandGitHub
|
30fbd05537
|
[ROCm] Use backend-default dot precision for ReplaySSM (#49909)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-07-27 18:11:06 +08:00 |
|
|
|
eb290ab673
|
[Bugfix][CPU] Zero-pad MoE intermediate size for grouped-gemm TP alignment (#49591)
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
|
2026-07-27 16:32:23 +08:00 |
|
|
|
5d07e268b1
|
[Quantization][INC]Add MXFP8 Linear Support (#47514)
Signed-off-by: Zhenzhong1 <zhenzhong.xu@intel.com>
Signed-off-by: Zhenzhong Xu <zhenzhong.xu@intel.com>
Co-authored-by: Yi Liu <yi4.liu@intel.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-07-27 14:26:31 +08:00 |
|
|
|
544cb724c8
|
[CPU][Spec Decode] Optimize GDN conv path for speculative decoding (#48577)
Signed-off-by: Li, Tianmu <tianmu.li@intel.com>
Co-authored-by: Codex <codex@openai.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
|
2026-07-27 06:20:14 +00:00 |
|
|
|
c314af1abf
|
[CPU][Perf] INT8 Fused MoE Kernel for Arm CPUs (#48637)
Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
|
2026-07-27 05:53:09 +00:00 |
|
|
Nick HillandGitHub
|
74d3b799e1
|
[Bugfix] Fix mHC block-M prenorm GEMM cross-row reduction carry-over (#49429)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
|
2026-07-27 04:42:47 +00:00 |
|
|
Schwinn SaereesitthipitakandGitHub
|
b5b61c622c
|
[Core][Distributed] Add process-checkpoint lifecycle hooks for communicators (starting with Flashinfer) (#46877)
Signed-off-by: Schwinn Saereesitthipitak <schwinns@nvidia.com>
|
2026-07-26 14:47:50 -04:00 |
|
|
Taneem IbrahimandGitHub
|
0da6e7f3d6
|
[Bugfix] Reject contradictory custom-op directives (#49134)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
|
2026-07-26 08:42:14 -04:00 |
|
|
Guan-Ming ChiuandGitHub
|
21fd9e85a0
|
[Model] Support top_k and top_p sampling for DiffusionGemma (#45429)
Signed-off-by: Guan-Ming (Wesley) Chiu <105915352+guan404ming@users.noreply.github.com>
|
2026-07-26 08:39:25 +00:00 |
|
|
|
30b0714031
|
[Perf] DeepSeek-OCR-2 TTFT Optimize (#49531)
Signed-off-by: RED <outofthewoods@qq.com>
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
|
2026-07-26 05:53:06 +00:00 |
|
|
Athrael SojuandGitHub
|
1240c74c0a
|
[Bugfix] Respect declared attention contract for ColQwen3.5 retrievers (#49372)
Signed-off-by: Athrael Soju <athrael.soju@gmail.com>
|
2026-07-26 04:08:07 +00:00 |
|
|
|
0111002323
|
[Kernel] TD operand loads for batched MoE GEMM (moe_mmk) on XPU (#46340)
Signed-off-by: oonyshch <xonyshch@gmail.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-07-26 08:50:07 +08:00 |
|
|
|
70009fb934
|
[MM][CG] Support ViT CUDA Graph for Gemma-4 (#46837)
Signed-off-by: Anthony Su <xsuanthony@gmail.com>
Co-authored-by: Linkun Chen <github@lkchen.net>
|
2026-07-25 15:02:09 -05:00 |
|
|
|
9321aff536
|
[Bugfix] Wait for the linear bias before layerwise online processing (#49805)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
2026-07-25 17:56:32 +00:00 |
|
|
Harry MellorandGitHub
|
26d725c334
|
[Model] Add VaultGemma via Transformers modeling backend (#49803)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-07-25 16:54:15 +00:00 |
|
|
Wentao YeandGitHub
|
7fe6d3c76b
|
[Perf] Fix moe reduce_scatter perf regression by removing additional comm, 5% E2E throughput gain back. (#48763)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-07-25 16:36:19 +00:00 |
|