biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity
14,985 Commits 2 Branches 140 Tags
738d0a281fab2e151a67b370c26b4e4360362f8f
Commit Graph

738 Commits

Author SHA1 Message Date
Xin Yang
b1169d7be8 [Kernel] Add gpt-oss Router GEMM kernel (#37205) 
Signed-off-by: Xin Yang <xyangx@amazon.com>
 2026-03-18 08:15:56 -07:00
Michael Goin
09e4576f65 [Kernel] Add non-gated support for NVFP4 CUTLASS MoE (#37320) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2026-03-17 18:12:04 -04:00
Terry Gao
3e6a1e1686 [Custom Ops] Add functional + out variant for scaled_fp4_quant (#34389) 
Signed-off-by: tianrengao <terrygao87@gmail.com>
 2026-03-16 18:51:46 -04:00
Krish Gupta
c0f011918d [Bugfix] opcheck false mutation error in rms_norm_per_block_quant (#36688) (#36779) 
Signed-off-by: Krish Gupta <krishom70@gmail.com>
 2026-03-16 21:11:33 +00:00
Matthew Bonanni
c88ea8338b [MTP][Sparse MLA] Take advantage of native MTP support in indexer when possible (#36982) 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
 2026-03-16 13:51:21 -04:00
Wentao Ye
ce8cf9161d [Compile] Fix compile warning st256_cs in cuda_vec_utils.cuh (#36693) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-16 11:12:15 -04:00
xjx
18be11fd59 [BUGFIX]fix CUDA OOM ERROR : invalid argument at cumem_allocator.cpp:119 (#35594) 
Signed-off-by: xjx <493337577@qq.com>
 2026-03-16 15:10:42 +00:00
Wentao Ye
e855d380fa [Compile] Fix compile warning in moe_permute (#36529) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-16 10:16:14 -04:00
Luka Govedič
9556af87d5 [torch.compile] Add support for non-contiguous fused RMSNorm + group quant (#36551) 
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ProExpertProg <11367180+ProExpertProg@users.noreply.github.com>
 2026-03-11 10:56:55 -07:00
Julien Denize
a5d06dc557 Add 320 dimension size support to MLA (#36161) 
Signed-off-by: Julien Denize <julien.denize@mistral.ai>
 2026-03-11 10:21:22 -07:00
typer-J
4184653775 feat: add RISC-V support for CPU backend (v2) (#36578) 
Signed-off-by: typer-J <2236066784@qq.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
 2026-03-10 21:51:39 -07:00
Hashem Hashemi
721ae79f50 Improvements to wvSplitKrc skinny GEMM solution (#34304) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-03-10 09:14:27 -07:00
Roberto L. Castro
580864d81e [Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917) 
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
 2026-03-09 09:50:36 -07:00
Roberto L. Castro
2b28b9b269 [Attention][Perf] Optimize cp_gather_and_upconvert_fp8_kv_cache - DeepSeek-v3.2 (#35290) 
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
 2026-03-09 09:46:57 -07:00
Jiayi Yan
6a895197fa [Bugfix][CI] fix typos (#34934) 
Signed-off-by: 1195343015 <1195343015@qq.com>
Signed-off-by: Jiayi Yan <66017932+1195343015@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-03-05 17:05:46 +00:00
Tianmu Li
8e7820131e [Perf] Use dummy M for weight prepacking on x86 (#35890) 
Signed-off-by: Li, Tianmu <tianmu.li@intel.com>
 2026-03-05 04:56:49 +00:00
EdalatiAli
cb21972a97 [Kernel] Integrate SM100 MXFP8 blockscaled grouped MM and quant kernels (#34448) 
Signed-off-by: EdalatiAli <aliedalati@cohere.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-03-01 23:31:19 -08:00
Asaf Gardin
bbf81f9a92 [Mamba1] - Kernel Level Chunk Alignment for Prefix Caching (#34798) 
Signed-off-by: Josephasafg <ajgard7@gmail.com>
 2026-03-01 20:40:23 +08:00
Hashem Hashemi
7600642eae Add padding support to wvSplitK solution for skinny GEMMs (#33762) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-02-28 09:02:05 +00:00
Ma Jian
90805ff464 [CI/Build] CPU release supports both of AVX2 and AVX512 (#35466) 
Signed-off-by: jiang1.li <jiang1.li@intel.com>
Co-authored-by: jiang1.li <jiang1.li@intel.com>
 2026-02-28 04:35:21 +00:00
Roberto L. Castro
a201ad72d8 [Refactor][Kernel] Add global helper to deduplicate vectorized memory ops (#35105) 
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: LopezCastroRoberto <roberto.lopez.castro@udc.es>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
 2026-02-27 16:28:17 -08:00
roikoren755
38c498b8e3 [Performance] Cublas Bf16 Gate with Fp32 Output (#35121) 
Signed-off-by: Roi Koren <roik@nvidia.com>
 2026-02-26 16:51:28 -08:00
Tyler Michael Smith
eb19955c37 [WideEP] Remove pplx all2all backend (#33724) 
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 14:30:10 -08:00
Asaf Gardin
ec13e549d3 [Bugfix] Fix uint32 overflow in Mamba selective scan state pointer arithmetic (#35275) 
Signed-off-by: Josephasafg <ajgard7@gmail.com>
 2026-02-26 12:22:06 +00:00
Roberto L. Castro
86c3b5a808 [BugFix] Fix fp4 quant kernel on CUDA 12.8 (#35210) 
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
 2026-02-25 18:32:50 -08:00
wenshuai
cd43673668 [Perf] Optimize FP8 gemm of sm120. (#34424) 
Signed-off-by: wenshuai <wenshuai@xiaomi.com>
 2026-02-24 22:25:24 -08:00
Xin Yang
3bbb2046ff [Bugfix] Fix expert_ids padding values in moe_align_block_size kernel (#35161) 
Signed-off-by: Xin Yang <xyangx@amazon.com>
 2026-02-24 17:14:24 -08:00
Hashem Hashemi
a0e50a4260 Convert wvSplitKQ to 16x16 MFMA in prep for mi4xx. (#34100) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-02-24 23:35:21 +00:00
R3hankhan
34ce0ffd1f [CPU][Perf] Accelerate Attention head for s390x using vector intrinsics (#34434) 
Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
 2026-02-24 07:25:39 -08:00
Michael Goin
3ef9fd0f98 [Bugfix] Fix DSV3 kernels breaking _C and _moe_C on unsupported arches (#35123) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2026-02-23 17:11:27 -08:00
Robert Shaw
8435b2e049 [ModelBash][DSV3] Add TRTLLM DSV3 Router GEMM kernel (6% B1 Speedup) (#34302) 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
 2026-02-23 14:02:26 +00:00
Huy Do
272b535ab3 [Bugfix] Gate 256-bit instructions to CUDA 12.9+ (#34791) 
Signed-off-by: Huy Do <huydhn@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
 2026-02-21 04:48:14 -08:00
Xin Yang
b1c4f0b265 [Kernel] Optimize grouped topk kernel (#34206) 
Signed-off-by: Xin Yang <xyangx@amazon.com>
 2026-02-20 01:34:45 -08:00
Robert Shaw
6874638bc4 [Model Bash] DeepSeek R1 BF16 Min Latency QKV A GEMM (0.5% E2E Speedup) (#34758) 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
 2026-02-18 07:42:36 -08:00
ElizaWszola
a88b3be7c4 [Bugfix] Fix quant RMS norm fusion for quantization with TMA-aligned scales (#33255) 
Signed-off-by: ElizaWszola <ewszola@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2026-02-17 23:35:04 -08:00
Hongxia Yang
4a00a511bb [BugFix] [Build] fix string literals comparison in indexer_k_quant_and_cache calling site (#34653) 
Signed-off-by: Hongxia Yang <hongxiay.yang@amd.com>
Co-authored-by: Hongxia Yang <hongxiay.yang@amd.com>
 2026-02-17 19:19:41 -08:00
Pushpinder Singh
bcd65c1f6a [Bugfix] Replace c10::optional with std::optional in topk kernel (#34467) 
Signed-off-by: Pushpinder Singh <pushpindersingh135@gmail.com>
 2026-02-13 08:30:23 -08:00
Wei Zhao
59d53066d8 [Feature] Support CPU Offloading without Pytorch Pinned Memory that leads to doubled allocation (#32993) 
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-02-13 08:11:26 -08:00
Hashem Hashemi
fac4e96940 small adjustment to wvSplitKrc (#34410) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-02-12 20:26:36 +00:00
Kyle Sayers
e9cd691132 [Bugfix] Fix Sparse24 Compressed Tensors models (#33446) 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-02-11 23:15:16 -08:00
Li, Jiang
05339a7b20 [Bugfix][CPU] Fix llama4 inference on CPU (#34321) 
Signed-off-by: jiang1.li <jiang1.li@intel.com>
 2026-02-11 19:07:23 +08:00
R3hankhan
d1b837f0ae [CPU] Enable FP16 (Half dtype) support for s390x (#34116) 
Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>
 2026-02-11 14:41:42 +08:00
Дзержи́нский
1485396abb [Kernel] Apply 256bit LDG/STG To Activation Kernels (#33022) 
Signed-off-by: Dzerzhinsky <256908701+AstroVoyager7@users.noreply.github.com>
Signed-off-by: Дзержи́нский <256908701+AstroVoyager7@users.noreply.github.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
 2026-02-10 19:31:51 -08:00
Kebe
5ee5c86eeb [Bugfix][DeepSeek-V3.2] fix fp8 kvcache type cast (#33884) 
Signed-off-by: Kebe <mail@kebe7jun.com>
 2026-02-10 19:31:36 -08:00
Roberto L. Castro
afdce12c89 [Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680) 
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>
 2026-02-10 10:29:52 -05:00
Nikhil Gupta
caad9f1e01 [Fix] [CPU Backend] : Prepack weights for w8a8 oneDNN matmul (#33901) 
Signed-off-by: nikhil-arm <nikhil.gupta2@arm.com>
 2026-02-09 18:04:41 +08:00
ihb2032
5a5c43511a fix(cpu): fix mla_decode compilation on x86 without AVX512 (#34052) 
Signed-off-by: ihb2032 <hebome@foxmail.com>
Co-authored-by: root <root@LAPTOP-FKNHV411.localdomain>
 2026-02-09 08:55:41 +00:00
Hashem Hashemi
ed17f54c8b Perf tuning and expansion of cases covered for wvSplitKrc (#33493) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-02-07 05:33:11 -08:00
Vel
bc32444b23 [Kernel] Add enable_sm120_or_later for SM121 (DGX Spark) CUTLASS support (#33517) 
Signed-off-by: code4me2 <velvetmoon222999@gmail.com>
 2026-02-06 20:28:01 -08:00
Wentao Ye
77c09e1130 [Refactor] Remove align block size logic in moe_permute (#33449) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-02-06 10:57:06 -08:00