 JartXandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
32daf56b42
|
[Refactor] Rename rocm_moe.py to rocm_moe_rdna.py (#45011)
Signed-off-by: JartX <sagformas@epdcenter.es>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-10 17:02:09 +08:00 |
|
|
Harry Mellor
|
af9f583344
|
Revert "[Bugfix][CI] Gemma3 Transformers multimodal encoder profiling and build prompt-embedding fixtures" (#45029)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-06-10 01:37:03 -07:00 |
|
|
yiheng
|
bd2d83ff31
|
[SpecDecode] Reduce TP communication for large-vocab draft models speculative decoding (#39419)
Signed-off-by: EanWang211123 <wangyiheng@sangfor.com.cn>
|
2026-06-10 07:59:24 +00:00 |
|
|
xiaohuguo2023
|
bb78168b21
|
[ROCm][gpt-oss] Hybrid CDNA4 swizzle gate for A8W4 MoE (#44804)
Signed-off-by: Xiaohu Guo <Xiaohu.Guo@amd.com>
|
2026-06-09 23:59:44 -07:00 |
|
|
Furkan Fandwang.yuqi
|
7fdfa6441d
|
Model/colbert autoweightsloader (#44999)
Signed-off-by: Furkan Fidan <dev@yufufi.com>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
|
2026-06-09 23:58:50 -07:00 |
|
|
bnellnm
|
f4966f8b3d
|
[Bugfix] Fix weight loading issues caused by #41184 (#45054)
Signed-off-by: Bill Nell <bnell@redhat.com>
|
2026-06-10 01:20:13 -04:00 |
|
|
|
6deb05e0e4
|
[Core][Model] Gemma4: Unified FA4 for all layers + FlashAttention mm_prefix support (#42175)
Signed-off-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
|
2026-06-09 17:45:39 -07:00 |
|
|
Jiangyun Zhu
|
ca4cfd8731
|
[Bugfix] fix qwen3.5 ep weight loading (#45002)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
|
2026-06-09 13:55:30 -07:00 |
|
|
Fangzhou AiandvLLM Contributor
|
01d8cd92dd
|
[ROCm][Perf] Use fused softplus-sqrt-topk router under AITER fused-MoE (#44945)
Co-authored-by: vLLM Contributor <contributor@vllm.ai>
|
2026-06-09 17:53:05 +00:00 |
|
|
Michał GanczarenkoandKunshang Ji
|
766ce2bb6b
|
Fix MiDashengLM TP>1 crash in audio encoder attention (#44408)
Signed-off-by: Michał Ganczarenko <michal.ganczarenko@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-06-09 09:29:14 -07:00 |
|
|
Andreas Karatzas
|
b697119800
|
[ROCm][CI] Stabilize ModernBERT token-classification parity against Hugging Face (#44040)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-06-09 16:52:36 +01:00 |
|
|
Raushan Turganbay
|
2ee5106372
|
Remove raw_inputs from transformers backend (#39425)
Signed-off-by: raushan <raushan@huggingface.co>
|
2026-06-09 15:01:04 +00:00 |
|
|
Jiangyun Zhu
|
7a89b72564
|
[Perf] fuse qk rmsnorm rope gate for qwen3.5 (#44176)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
|
2026-06-09 22:12:17 +08:00 |
|
|
Jee Jee Li
|
dc10e467a9
|
[Bugfix] Fix minimax_qk_norm_fusion (#44983)
|
2026-06-09 06:43:46 -07:00 |
|
|
Terrence Zhao
|
3e8afdf785
|
[Cohere] Fix Cohere2MoE weight loading when using Transformers ≥5.10 (#44747)
Signed-off-by: Terrencezzj <terrence@cohere.ai>
|
2026-06-09 06:27:40 -07:00 |
|
|
Qiuyang Yue
|
59401ac9f1
|
[Kernel][Perf] Tune fused_moe FP8 config for Qwen3-Next-80B tp=4 on H100 (+25% at batch 96-512) (#44830)
Signed-off-by: Qiuyang Yue <yueqiuyang1389@gmail.com>
|
2026-06-09 04:15:51 -07:00 |
|
|
Ma Jian
|
70db1488c5
|
[DSV4][XPU] Add MHC fused_post_pre support (#44144)
Signed-off-by: Ma Jian <jian1.ma@intel.com>
|
2026-06-09 17:23:17 +08:00 |
|
|
|
e6fc848d4f
|
[Bugfix][MiniCPM-o] Fix cuda/cpu device mismatch in Resampler2_5 pos_embed (#43844)
Signed-off-by: Parth Ashwin Jain <parthash@amd.com>
Co-authored-by: Parth Ashwin Jain <parthash@amd.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-08 23:28:26 -07:00 |
|
|
Andreas Karatzas
|
f843ac1a1c
|
[Bugfix][CI] Gemma3 Transformers multimodal encoder profiling and build prompt-embedding fixtures (#44952)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-06-09 05:49:30 +00:00 |
|
|
|
9f153aa781
|
[MM][Perf][CG] Support ViT full CUDA graph for glm4_1v image and video inference (#40576)
Signed-off-by: grYe99 <guorongye99@gmail.com>
Co-authored-by: grYe99 <guorongye99@gmail.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-06-09 11:13:56 +08:00 |
|
|
Lanze Liu
|
540aaf2140
|
[Bugfix][Model] Qwen3-Omni: move cu_seqlens to GPU before VIT attention (#44264)
Signed-off-by: Lanze Liu <lanzetech@gmail.com>
|
2026-06-08 20:02:27 -07:00 |
|
|
Tyler Michael SmithandClaude Opus 4.6
|
e2f993dc41
|
[WideEP] Integrate DeepEP v2 (#41183)
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
2026-06-08 18:07:29 -07:00 |
|
|
Michael Goin
|
6afa25000c
|
[Bugfix] Canonicalize FP8 weight layout to (K, N) at the source (#44735)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-06-08 14:37:36 -06:00 |
|
|
Mohammad Miadh Angkad
|
823a0ab754
|
[Bugfix][MoE] Fix fused MoE expert mapping helper call sites (#44897)
Signed-off-by: Mohammad Miadh Angkad <176301910+mmangkad@users.noreply.github.com>
|
2026-06-08 13:35:04 -07:00 |
|
|
bnellnm
|
dc68bd8c41
|
[MoE Refactor] FusedMoE/MoERunner inversion refactor (#41184)
Signed-off-by: Bill Nell <bnell@redhat.com>
|
2026-06-08 10:42:58 -04:00 |
|
|
Walter Beller-Moralesandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
753e9d55e6
|
[Quantization] add online fp8 ptpc (#44132)
Signed-off-by: walterbm <walter.beller.morales@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-08 22:42:11 +08:00 |
|
|
Sungjae Lee
|
469f3dcf1d
|
[BugFix] Use served model name in gemma4 audio-tower error message (#44828)
Signed-off-by: Sungjae Lee <33976427+llsj14@users.noreply.github.com>
Signed-off-by: Sungjae Lee <sung-jae.lee@navercorp.com>
|
2026-06-08 06:58:31 +00:00 |
|
|
Yan Ma
|
54c660c3a6
|
[XPU][Minor] format moe kernel name and add in kernel list (#44771)
Signed-off-by: Yan Ma <yan.ma@intel.com>
|
2026-06-08 13:58:16 +08:00 |
|
|
Shanshan Shen
|
8fb0274415
|
[MM][CG] Simplify ViT CUDA graph interfaces (#44484)
Signed-off-by: shen-shanshan <467638484@qq.com>
|
2026-06-08 05:57:06 +00:00 |
|
|
Ma Jian
|
eebce65756
|
[XPU]feat: add DeepSeek-V4 XPU attention decode path (#42953)
Signed-off-by: Ma Jian <jian1.ma@intel.com>
|
2026-06-08 13:27:12 +08:00 |
|
|
Taneem Ibrahim
|
5633405964
|
Added extra_repr() to pooler classes to improve debuggability (#44805)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
|
2026-06-08 03:19:31 +00:00 |
|
|
SunskyXHandRoger Wang
|
6124a98a9b
|
[Bugfix] Fix FunASR-Nano crash during initialization (#44215)
Signed-off-by: SunskyXH <sunskyxh@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
|
2026-06-07 20:00:02 -07:00 |
|
|
Jared Wenandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
32f34d3935
|
[feature] add index share feature for DSA MTP (#44420)
Signed-off-by: JaredforReal <w13431838023@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-06 22:04:14 -07:00 |
|
|
zofiaandKunshang Ji
|
6181e80fe0
|
[XPU] add xpu branch in compressed_tensors_moe_w4a4_mxfp4 (#44540)
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
Co-authored-by: Kunshang Ji <jikunshang95@gmail.com>
|
2026-06-07 12:27:34 +08:00 |
|
|
JartXandTJian
|
062b05ff3a
|
[ROCm][Perf] Fused MoE W4A16 HIP kernel for AMD RDNA3 (gfx1100) (#44075)
Signed-off-by: JartX <sagformas@epdcenter.es>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-06-06 15:30:39 -05:00 |
|
|
Vadim Gimpelsonandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
fa27d4e9cf
|
[PERF] [Qwen3.5] Split mixed prefill+decode batches: route decodes to the recurrent kernel (#44700)
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-06 22:13:50 +08:00 |
|
|
|
00d1fb7747
|
[Bugfix][ROCm] ApplyRotaryEmb: fall back to native when flash_attn rotary grid would exceed the HIP per-dim limit (#43684)
Signed-off-by: vLLM ROCm fix <noreply@example.com>
Signed-off-by: amd-fuweiy <fuweiy@amd.com>
Co-authored-by: vLLM ROCm fix <noreply@example.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-06-06 02:29:16 -07:00 |
|
|
aoshen02andClaude Opus 4.8
|
f87df1df9e
|
[Bugfix][MoE] Snapshot max_cudagraph_capture_size into FusedMoEConfig (#44613)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-05 23:14:41 -07:00 |
|
|
Taneem Ibrahim
|
eafbb06331
|
[Misc] Replaced asserts with proper exceptions to improve UX for pooling (#44593)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
|
2026-06-06 05:57:26 +00:00 |
|
|
Yongye ZhuandClaude
|
ec0a31d4aa
|
[Bugfix][Kernel] Fix mHC fused-RMSNorm big-fuse miscompile for hidden_size != 4096 (#44692)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
2026-06-06 10:44:21 +08:00 |
|
|
Vadim GimpelsonandCursor
|
4765f0f189
|
[Bugfix] Fix sequence_parallel_chunk_impl custom op aliasing its input (#44130)
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
|
2026-06-05 23:56:36 +00:00 |
|
|
Terrence Zhao
|
a50e675b0d
|
[Cohere] fix RoutingMethodType (#44021)
Signed-off-by: Terrencezzj <terrence@cohere.ai>
|
2026-06-05 16:25:53 -07:00 |
|
|
Harry Mellor
|
ef0df7dbd6
|
[CI] Bump mypy version 1.19.1 -> 1.20.2 (#44647)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-06-05 14:56:27 +00:00 |
|
|
Harry Mellor
|
a80af24356
|
Speed up docs build (#44635)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-06-05 14:51:44 +00:00 |
|
|
Harry Mellorandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
62215e72c6
|
Remove KV cache scale boilerplate from model weight loading methods (#43167)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-06-05 05:19:04 -07:00 |
|
|
|
165b7864d0
|
[ROCM] [FEAT] Integrate Aiter hipBLASLt GEMM online tuning (#40426)
Signed-off-by: hanlin12 <hanlin12@amd.com>
Signed-off-by: Han Lin <hanlin12@amd.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-06-04 23:45:36 -07:00 |
|
|
Luciano MartinsandLuciano Martins
|
da1daf40bf
|
[Bugfix] Exclude vision embedder from quantization in Gemma4 Unified (#44571)
Signed-off-by: Luciano Martins <lucianommartins@users.noreply.github.com>
Co-authored-by: Luciano Martins <lucianommartins@users.noreply.github.com>
|
2026-06-04 20:47:38 -07:00 |
|
|
zofia
|
063ce98fb7
|
[XPU][MoE] support block_fp8_moe on xpu (#42139)
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
Signed-off-by: zofia <110436990+zufangzhu@users.noreply.github.com>
|
2026-06-05 08:36:58 +08:00 |
|
|
Schwinn Saereesitthipitak
|
b7c5baf63d
|
fix: keep DeepSeek V4 RoPE cache on inv_freq device (#43926)
Signed-off-by: Schwinn Saereesitthipitak <schwinns@nvidia.com>
Signed-off-by: Schwinn Saereesitthipitak <17022745+galletas1712@users.noreply.github.com>
|
2026-06-05 02:30:29 +04:00 |
|
|
Jiangyun Zhu
|
a55fccfc7c
|
[mamba] unify KDA conv states into one cache to match 2-state SSM layout (#44539)
|
2026-06-04 20:38:05 +02:00 |
|