|
eaf1627c01
|
[ROCm][aiter] officially supporting aiter Triton kernels+dsv4 on RDNA3 (#52970)
Signed-off-by: xawang <xawang@amd.com>
Signed-off-by: skysnow2001 <skysnow9285@gmail.com>
Co-authored-by: skysnow2001 <skysnow9285@gmail.com>
Co-authored-by: keneoneth <keneoneth18@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-10-27 17:01:13 +00:00 |
|
|
|
94ad4dfbae
|
[ROCm][Perf] W4A16: opt-in dense weight cache for RDNA hybrid prefill (#60710)
Signed-off-by: Zohaib Moti <zohaib.moti@amd.com>
Co-authored-by: roberteg16 <roberteg16@users.noreply.github.com>
Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
|
2026-10-27 16:57:52 +00:00 |
|
|
|
dc4d977d97
|
[MRV2] Let the runner decide whether to profile CUDA graph memory (#63778)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 12:52:47 -04:00 |
|
|
|
6aa70c4bec
|
[ROCm][Perf][Kimi-K3] Fold q_b_proj and W_UK into one GEMM for the MLA decode query (#60110)
Signed-off-by: Nicholas Susanto <nicholas.susanto@amd.com>
Signed-off-by: Omar Muhammad <omar.muhammad@amd.com>
Co-authored-by: Omar Muhammad <omar.muhammad@amd.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 11:34:59 -05:00 |
|
|
|
889d635dd7
|
[Perf][ROCm][Attention] Fuse Kimi-K3 MLA chunked-context gather on AITER (#53166)
Signed-off-by: Nicholas Susanto <nicholas.susanto@amd.com>
Signed-off-by: Omar Muhammad <omar.muhammad@amd.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
Co-authored-by: Omar Muhammad <omar.muhammad@amd.com>
|
2026-10-27 11:29:57 -05:00 |
|
|
Carl Persson
|
228b1e3d23
|
[ROCm][DSV4.1][Perf] Single-launch DSA decode candidate mask (#59668)
Signed-off-by: Carl Persson <carl.persson@amd.com>
|
2026-10-27 16:08:14 +00:00 |
|
|
Woosuk KwonandCodex
|
e3ba8d00db
|
[Bugfix][Bench] Handle packed hybrid KV and speculative recurrent state (#63873)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Codex <noreply@openai.com>
|
2026-10-27 09:04:59 -07:00 |
|
|
Robert EsclapezandDouglas Lehr
|
43111fd1f0
|
[Bugfix][Spec Decode] Fix eagle draft sharing with a quantized lm_head (#53104)
Signed-off-by: Robert Esclapez Garcia <robert.garcia@amd.com>
Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
|
2026-10-27 15:47:22 +00:00 |
|
|
|
83909ad84b
|
[ROCm][Perf] W4A16: widen skinny GEMM LDS activation loads to b128 (#57015)
Signed-off-by: Robert Esclapez Garcia <robert.garcia@amd.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
|
2026-10-27 15:46:42 +00:00 |
|
|
Isotr0pyandvllm-agent
|
a9b727a612
|
[UX] Remove --hf-overrides necessity for Qwen3-ASR-ForceAligner (#63274)
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 23:20:03 +08:00 |
|
|
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 |
|
|
Yusen Gao
|
bf4b3b5cb8
|
[CI][CPU] Initialize DFlash test FC after loading weights (#63973)
Signed-off-by: Yusen Gao <amenomorikou@gmail.com>
|
2026-10-27 22:41:31 +08:00 |
|
|
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 |
|
|
|
803bc13cf9
|
[Rust Frontend] Add --reasoning-effort-map to rewrite request reasoning efforts (#62312)
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
|
2026-10-27 22:04:08 +08:00 |
|
|
|
4216a234cf
|
[Rust Frontend] Add per-request speculative decoding metrics (#61877)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 22:03:10 +08:00 |
|
|
|
447d28eef2
|
[Rust Frontend] Group prompt-cache usage and report created_cache_tokens (#63292)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Signed-off-by: zhongdet <thomas980930@gmail.com>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 21:55:35 +08:00 |
|
|
|
2fb5111bba
|
[Perf][Rust Frontend] Skip detokenization for gRPC requests that read no text (#63789)
Signed-off-by: Alec Flowers <aflowers@nvidia.com>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 21:50:59 +08:00 |
|
|
Hemant Kumarandhemuku90
|
40fe396b60
|
[Bugfix][Multimodal] Return 400 when local media access is disabled (#64032)
Signed-off-by: hemuku90 <25220495+hemuku90@users.noreply.github.com>
Co-authored-by: hemuku90 <25220495+hemuku90@users.noreply.github.com>
|
2026-10-27 13:21:21 +00:00 |
|
|
|
e7fa26d5c8
|
[Model][Spec Decode] Refactor #41834 with useful improvements (#63996)
Signed-off-by: lazymio <mio@lazym.io>
Co-authored-by: alexbi29 <alexbi29@users.noreply.github.com>
Co-authored-by: jasl <jasl9187@hotmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
|
2026-10-27 20:13:39 +08:00 |
|
|
|
a91a1d06b7
|
[Bugfix][Model] Qwen3.8-Flash-Next: load checkpoints with quantized q/k/v and a BF16 QSA indexer (#62318)
Signed-off-by: Seungryong Kim <137975233+srkim-business@users.noreply.github.com>
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Kimi <noreply@moonshot.ai>
|
2026-10-27 11:54:14 +00:00 |
|
|
Hyukjoon LeeandClaude Opus 5.5
|
f8f7634cb4
|
[ROCm][Perf][Kimi-K3] Use block_m=128 for the fused KDA prefill conv1d (#63342)
Signed-off-by: hyukjlee <hyukjlee@amd.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 19:24:00 +08:00 |
|
|
|
b29caba14e
|
[CI] Shard draft decoding, Qwen eval and scheduling automatically (#61832)
Signed-off-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
Signed-off-by: Thang Nguyen <thangnguyenvn647@gmail.com>
Co-authored-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 04:21:50 -07:00 |
|
|
|
8d88bc5d81
|
[ROCm][DSv4/4.1] Fuse inverse RoPE and MXFP8 quant into the aiter sparse MLA and sparse MLA gating changes (#60154)
Signed-off-by: Mehmet Cagri Kaymak <mehmet.kaymak@amd.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Simon Danielsson <70206058+simondanielsson@users.noreply.github.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 19:15:04 +08:00 |
|
|
Harry MellorandClaude Opus 5.5
|
9e0bc7a651
|
[CI] Remove codeowner for pre-commit allowlists (#64021)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 11:11:37 +00:00 |
|
|
|
5f1a564181
|
[Recipe][AMD] Refresh Kimi-K3 PD configurations (#63726)
Signed-off-by: Yichao Zhu <Yichao.Zhu@amd.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Shanshan Shen <467638484@qq.com>
|
2026-10-27 19:05:02 +08:00 |
|
|
|
de58886274
|
[CI] Move B200 Kernels and MoE Refactor evals from manual to automatic shards (#62120)
Signed-off-by: Thang Nguyen <thangnguyenvn647@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 03:57:47 -07:00 |
|
|
Bugen ZhaoandClaude Opus 5.5
|
7fe793a5d5
|
[Rust Frontend] Support --enable-log-outputs and --enable-log-deltas (#63769)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 18:53:50 +08:00 |
|
|
|
984a4181de
|
[Rust Frontend] Keep completions top_logprobs and prompt_logprobs in the engine's order like Python (#63763)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 18:52:32 +08:00 |
|
|
|
dacd7e2cb8
|
[Rust Frontend] Forward profiler_kwargs from /start_profile (#63771)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 18:50:45 +08:00 |
|
|
Thang NguyenandClaude Opus 5.5
|
eb5e8d1cc8
|
[CI] Move L4 Distributed Models and LoRA TP and their AMD mirrors from fixed to automatic shards (#62537)
Signed-off-by: Thang Nguyen <thangnguyenvn647@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 03:49:07 -07:00 |
|
|
|
659defd662
|
[CI] Shard the AMD mirrors of Basic Models (Other) and PCP automatically (#61904)
Signed-off-by: Thang Nguyen <thangnguyenvn647@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 03:47:24 -07:00 |
|
|
Thang NguyenandClaude Opus 5.5
|
f62e80ff04
|
[CI] Enroll 4 steps and 2 AMD mirrors in automatic sharding (#63907)
Signed-off-by: Thang Nguyen <thangnguyenvn647@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 03:44:27 -07:00 |
|
|
nrenvo
|
a657fd6082
|
[Test][Multimodal] Add end-to-end CLI test for vllm bench mm-processor (#62497)
Signed-off-by: nrenvo <nrenvo@gmail.com>
|
2026-10-27 10:38:08 +00:00 |
|
|
Divakar Verma
|
7c713621f3
|
[CI][ROCm] Preflight idle-GPU check in test_pp_cudagraph (#63850)
Signed-off-by: Divakar Verma <divakar.verma@amd.com>
|
2026-10-27 10:32:24 +00:00 |
|
|
|
f03b704da2
|
[Bugfix][Kernel] Fix OAI Triton MXFP4 MoE on SM110 (#62133)
Signed-off-by: lazymio <mio@lazym.io>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
|
2026-10-27 10:27:09 +00:00 |
|
|
Jee Jee Li
|
08d94f0176
|
[Model] Add JIT warmup for Muse Spark NVFP4 MegaMoE input preparation (#63920)
Signed-off-by: Jee Jee Li <jeejeelee@inferact.ai>
|
2026-10-27 10:21:21 +00:00 |
|
|
Chaunceyandmergify[bot]
|
08b989da99
|
[Perf][DSpark] Stack DeepSeek V41 context WKV projections (#63063)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-10-27 10:10:22 +00:00 |
|
|
Chaojun Zhangandvllm-agent
|
ef2f382fd1
|
[Bugfix][LoRA] Avoid double rounding in fused MoE expand (#63894)
Signed-off-by: Chaojun Zhang <chaojun.zhang@intel.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 10:09:57 +00:00 |
|
|
|
90309f6b74
|
[Bugfix][Qwen4Exp] Reserve QSA indexer prefill workspace during memory profiling (#60490)
Signed-off-by: Nick Kuhn <95254386+nkuhn-vmw@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
|
2026-10-27 09:38:03 +00:00 |
|
|
|
de096755ad
|
[Bugfix] Fuse GLM-5.3 KDA q|k|v convs into one conv1d parameter (#60991)
Signed-off-by: Jared Wen <w13431838023@gmail.com>
Co-authored-by: Jyotirmoy Roy <jyotirmoyroy649@gmail.com>
Co-authored-by: Mustafa YILDIRIM <2341474+mustafayildirim@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
|
2026-10-27 09:36:17 +00:00 |
|
|
|
a5b257a34b
|
[Docs][KV Connector] Document standalone Mooncake EGM deployment (#63064)
Signed-off-by: wugou.cyf <wugou.cyf@antfin.com>
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: wugou.cyf <wugou.cyf@antfin.com>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Codex <noreply@openai.com>
|
2026-10-27 09:34:36 +00:00 |
|
|
Bugen ZhaoandClaude Opus 5.5
|
22d1bc8814
|
[Rust Frontend] Stream kv_transfer_params on the final chunk like Python (#63772)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
|
2026-10-27 17:31:03 +08:00 |
|
|
Louie Tsai
|
115b59a2da
|
[Recipes Tool] Fix TP auto-detection for heterogeneous per-layer attention models (#63857)
Signed-off-by: louie-tsai <louie.tsai@intel.com>
|
2026-10-27 17:28:59 +08:00 |
|
|
karma_osandvllm-agent
|
a23d7dc3f1
|
[Bugfix][CPU] Run encoder-only attention bidirectionally in CPU_ATTN (#63738)
Signed-off-by: gustav <gusangmo0605@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 17:26:20 +08:00 |
|
|
NVShreyasandThien Tran
|
8e26dec5f5
|
[Kernel][MiniMax-M3] Add Triton query-tiled sparse prefill tuned on Hopper (#57420)
Signed-off-by: Shreyas Misra <shreyasm@nvidia.com>
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
|
2026-10-27 16:26:03 +07:00 |
|
|
|
5314c58281
|
[ROCm][Recipe] Qwen3.8-Flash-Next: add MI355X Quark MXFP4 variant (TP1/TP2) (#63332)
Signed-off-by: Mikko Tukiainen <Mikko.Tukiainen@amd.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-10-27 09:20:39 +00:00 |
|
|
|
0076619a94
|
[Perf][Kernel] Add a CuteDSL BF16x3 router GEMM for fp32 router weights on Hopper (#61903)
Signed-off-by: Chen Cheng <85005591+ischencheng@users.noreply.github.com>
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 02:16:44 -07:00 |
|
|
Reidandvllm-agent
|
f09d5e30c3
|
[Rust][Benchmark] Fix multi-turn input token accounting (#51109)
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 09:03:38 +00:00 |
|
|
|
c56cdbea15
|
[Bugfix][Rust Frontend] Decode current NIXL connector stats (#60856)
Signed-off-by: Markov Ilya <markovilya197@gmail.com>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: vllm-agent <claw@inferact.ai>
|
2026-10-27 16:57:15 +08:00 |
|
|
|
5c1eb11cb3
|
[Bugfix] Allow pipeline parallelism for Qwen4Exp PLE models and their MTP drafter (#60234)
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
|
2026-10-27 08:37:20 +00:00 |
|