karylab_agents/vllm
1
0
Fork 0
forked from Karylab-cklius/vllm
Code Pull Requests 2 Actions 1 Packages Activity
17,719 Commits 354 Branches 175 Tags
d2fbaf73c19e2bf519e4e4fc954ec8c14ba1e1fc
Commit Graph
45 Commits
Author SHA1 Message Date
Yongye ZhuandClaude Opus 4.8 bb4844dba3 [Model] Add fmha_sm100 MSA indexer backend + fp8 index cache for MiniMax M3 
Add an SM100/Blackwell lightning-indexer impl that computes the per-128-block
QK max-scores with fmha_sm100's score-only (OnlyScore) path and selects the
top-k blocks with the existing Triton minimax_m3_index_topk kernel, mirroring
how the main MSA attention pairs the SM100 attend with Triton. Decode and
prefill requests are split manually (decode-first batch) and each side gets its
own _fmha_sm100_plan / _fmha_sm100 call. Auto-selected on SM100 when
topk_blocks in (4, 8, 16, 32) for both bf16 and fp8 index caches; falls back to
the Triton indexer otherwise. The builder declares AttentionCGSupport.NEVER
(eager; the attention is broken out of the graph by _run_attention).

Extend the fused qknorm+rope+kv-insert kernel to optionally emit fp8 (e4m3) for
the index-K cache and index-Q via a direct cast with no scale tensors (RMSNorm
outputs are O(1) and scalar scales do not change top-k ordering). Only the index
outputs go fp8; q/k/v and q_out stay bf16 and bit-identical to the existing
path. MiniMaxM3IndexerCache now accepts fp8 caches and the model allocates
index_q in the cache dtype.

Tests: test_fmha_sm100_indexer_matches_reference (bf16/fp8 x prefill/decode) and
test_msa_indexer_impl_matches_triton (full impl parity vs the Triton indexer
through the real metadata builders); fp8 fused-kernel parity is covered in
test_fused_minimax_m3_qknorm_rope_kv_insert.

AI assistance (Claude Code) was used for this change.

Signed-off-by: Yongye Zhu <yongye@inferact.ai>
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
 2026-06-17 05:32:35 +00:00
Thien TranGitHubYongye Zhu
 		4c62663315 [M3] Enable FP8 sparse GQA (#45744) 
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
 2026-06-16 21:38:03 -07:00
+1
youkaichaoGitHubOpenAI CodexIsotr0pyThien TranBugen ZhaoJee Jee LiRoger WangfunctionstackxYongye ZhuJee Jee Li
 		0a1c5034f5 [Model] Add MiniMax M3 support (#45381) 
Signed-off-by: youkaichao <youkaichao@gmail.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Signed-off-by: functionstackx <47992694+functionstackx@users.noreply.github.com>
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Signed-off-by: Jee Jee Li <jeejeelee@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: functionstackx <47992694+functionstackx@users.noreply.github.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Jee Jee Li <jeejeelee@inferact.ai>
 2026-06-16 01:01:25 +08:00
Xin HeGitHubmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>Kunshang Ji
 		5ed15f42b9 Fix the E8M0 scale computation in the MXFP4 (W4A4) MOE CUTLASS kernel (#43557) 
Signed-off-by: Xin He <xin3.he@intel.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-06-15 06:04:54 -07:00
Shengqi ChenGitHubXin LiShawRongChange72
 		78e7293bb1 [Build] Fix CUDA arch build coverage gaps (#45277) 
Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Co-authored-by: Xin Li <xinli-sw@users.noreply.github.com>
Co-authored-by: ShawRong <ShawRong@users.noreply.github.com>
Co-authored-by: Change72 <Change72@users.noreply.github.com>
 2026-06-13 22:09:20 -07:00
Wentao YeGitHubmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
 		71b961dd35 [Perf] SM90 cutlass fp8 mm supports odd M by swap_ab, 180~290% kernel performance improvement (#44572) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
 2026-06-13 12:05:45 -07:00
Isotr0pyandGitHub 6635279d8a [Migration] Migrate GGUF quantization support to plugin (#39612) 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
 2026-06-12 12:02:21 -07:00
Chris LeonardGitHubShengqi Chen
 		a014dddbaa [11b/n] Migrate Machete kernels to torch stable ABI (#45304) 
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-06-12 10:36:49 +00:00
Chris LeonardandGitHub 7021be66e8 [11a/n] Migrate Marlin kernels to torch stable ABI (#45176) 
Signed-off-by: Chris Leonard <chleonar@redhat.com>
 2026-06-11 21:22:37 -07:00
Juan Pérez de AlgabaandGitHub f219788f91 [Security] Fix info disclosure via int32 truncation in GGUF dequantize kernels (#44971) 
Signed-off-by: jperezde <jperezde@redhat.com>
 2026-06-11 08:05:14 +00:00
Chris LeonardGitHubShengqi Chen
 		6e64c1bab1 [10c/n] Migrate MoE kernels to torch stable ABI (#44565) 
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-06-10 23:02:26 -07:00
Micah WilliamsonandGitHub e1ed89dbee Revert "[Kernel] Speed up silu_and_mul_per_block_quant with warp-shuf… (#45066) 
Signed-off-by: Micah Williamson <micah.williamson@amd.com>
 2026-06-09 14:12:06 -07:00
SII-yangdianGitHubSII-yangdian
 		a4b14b98c6 [Kernel] Speed up silu_and_mul_per_block_quant with warp-shuffle reduction + vectorized I/O (#44173) 
Signed-off-by: SII-yangdian <yangdian@sii.edu.cn>
Co-authored-by: SII-yangdian <yangdian@sii.edu.cn>
 2026-06-09 10:41:26 -07:00
Charlie FuandGitHub 228bcc436b [ROCm][Kernel] Enable permute_cols for ROCm (#44674) 
Signed-off-by: charlifu <charlifu@amd.com>
 2026-06-07 09:50:03 +00:00
Markus HartikainenGitHubCursorFrida AnderssonDouglas LehrTJian
 		bc5745a00f [ROCm][MLA] Replace torch.cat in sparse-MLA forward_mqa with fused concat_mla_q (#42838) 
Signed-off-by: Markus Hartikainen <markus.hartikainen@amd.com>
Signed-off-by: Frida Andersson <fanderss@amd.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Frida Andersson <fanderss@amd.com>
Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
 2026-06-06 18:20:50 -05:00
TJianandGitHub aa6fb8a329 [Bugfix] [ROCm] [Critical] fallback to regular abi for ROCm (#44648) 
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
 2026-06-05 15:51:17 +00:00
Chris LeonardandGitHub 56aff0dd15 [10/n] Migrate cuda_view and silu_and_mul_per_block_quant kernels to torch stale ABI. (#44334) 2026-06-04 20:14:43 -07:00
Yongye ZhuandGitHub b5235fca2e [DSv4] Adding TRTLLM gen attention kernel (#43827) 2026-06-04 07:35:09 -07:00
Chris LeonardGitHubShengqi Chen
 		59d0236193 [10b/n] Migrate custom all-reduce, DeepSeek V4 fused MLA, MiniMax reduce-RMS, and MXFP8 MoE to libtorch stable ABI (#44365) 
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-06-04 00:29:46 +08:00
Charlie FuandGitHub 71df063c49 Enable perf_token_group_quant/_C_stable_libtorch for ROCm (#42758) 
Signed-off-by: charlifu <charlifu@amd.com>
 2026-06-02 23:23:28 -07:00
Chris LeonardandGitHub 4d93bc35c9 Migrate header files to torch stable abi (#44013) 2026-06-02 08:09:52 -07:00
Lanze LiuGitHubmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
 		124fac10cb [Bugfix] Fix RMSNorm kernels to multiply in weight's native dtype (#42379) 
Signed-off-by: Lanze Liu <lanzetech@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
 2026-05-29 23:16:53 -07:00
Jee Jee LiGitHubYiliu Dong
 		559d6710bf [PERF]MiniMax-M2 gate kernel (#38445) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Signed-off-by: qianlihuang <91178480+qianlihuang@users.noreply.github.com>
Co-authored-by: Yiliu Dong <91178480+qianlihuang@users.noreply.github.com>
 2026-05-29 18:28:34 -07:00
Chris LeonardGitHubMikayla GawareckiShengqi Chen
 		22a58640b4 [9/n] Migrate attention and cache kernels to torch stable ABI (continued) (#43717) 
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-05-29 04:44:45 +00:00
Chris LeonardGitHubMikayla GawareckiShengqi Chen
 		284e6f543d [8/n] Migrate merge_attn_states, mamba, sampler to torch stable ABI (continued) (#43361) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Co-authored-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-05-27 09:35:24 -07:00
Jakub ZakrzewskiGitHubWentao YeYongye Zhu
 		5bb8d2767a [Kernel] Batch invariant NVFP4 linear using cutlass (#39912) 
Signed-off-by: Jakub Zakrzewski <jzakrzewski@nvidia.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
 2026-05-23 09:41:12 -04:00
Chris LeonardGitHubMikayla GawareckiShengqi Chen
 		a7be0f342d [7/n] Migrate pos_encoding and norm kernels to libtorch stable ABI (continued) (#43209) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Co-authored-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-05-23 13:20:00 +08:00
Chris LeonardGitHubMikayla GawareckiShengqi Chen
 		07aeaf9d4d [6/n] Migrate activation kernels, gptq, gguf, non cutlass w8a8 to libtorch stable ABI (continued) (#42663) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Signed-off-by: Chris Leonard <chleonar@redhat.com>
Co-authored-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
 2026-05-20 00:18:12 -07:00
Wentao YeandGitHub 37ece593c1 [Perf] Padded nvfp4 quant kernel to remove additional copy, 2.4%~5.7% e2e performance improvement (#42774) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-05-18 16:38:12 -07:00
Blake LeddenGitHubClaude Opus 4.6Pavani Majety
 		06d020bb6e [Bugfix] Fix SM121 (DGX Spark) exclusion from Marlin/CUTLASS FP8 paths (#35568) 
Signed-off-by: Blake Ledden <blake@secondnaturecomputing.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Pavani Majety <pmajety@nvidia.com>
 2026-05-15 10:59:00 -07:00
Chris LeonardGitHubMikayla Gawarecki
 		85b2fecab7 [5/n] Migrate CUTLASS MLA, hadamard, awq, allspark and DSV3 fused a gemm to torch stable ABI (continued) (#42339) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
Co-authored-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
 2026-05-13 07:24:39 +00:00
Jiahan Chang (Cyrus)GitHubYongye Zhu
 		dd6b3a5ef5 [Perf] Use 2D-grid to eliminate divmod in W8W8 group quant (#42153) 
Signed-off-by: jiahanc <173873397+jiahanc@users.noreply.github.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
 2026-05-12 10:01:30 -04:00
Wentao YeandGitHub 0d453e2336 [Perf] Batch invariance with Cutlass fp8 support, 28.9% E2E latency improvement (#40408) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Signed-off-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
 2026-05-11 12:20:58 -04:00
Yongye ZhuGitHubRoger Wang
 		a3c83ff2fd Faster per-token fp8 group quant packed kernel for blackwell (#41326) 
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
 2026-04-30 18:09:55 -07:00
Cyrus LeungandGitHub 4a42aba380 [CI/Build] Enable FP8 on NVIDIA Thor (#39712) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-04-29 09:48:52 -07:00
Jakub ZakrzewskiandGitHub 6fbec8ed47 [Bugfix][Kernel] nvfp4 cutlass MoE: fix nvfp4 experts quant out-of-bounds read for expert counts not divisible by 4 or 16 (#40351) 
Signed-off-by: Jakub Zakrzewski <jzakrzewski@nvidia.com>
 2026-04-21 19:06:09 +00:00
mysterious hhhhGitHubmgoinClaude Opus 4.7
 		d0359f3e04 [Bugfix] Guard mxfp4_experts_quant bindings on ENABLE_NVFP4_SM100 (#40191) 
Signed-off-by: ultranationalism <www913363043@gmail.com>
Signed-off-by: mgoin <mike.goin12@gmail.com>
Co-authored-by: mgoin <mike.goin12@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-18 13:58:46 -07:00
Michael GoinandGitHub a8bffaa133 [Kernel] Add MXFP4 W4A4 CUTLASS MoE kernel for SM100 (#37463) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2026-04-17 16:42:32 -07:00
Wei ZhaoandGitHub 59b2f7b640 [Perf] Fuse Zero Initializer for FP8 DeepGemm Block Quant Kernel (#39547) 
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
 2026-04-11 07:16:51 -07:00
NecofishandGitHub 580090db6b [Kernel] Add swapAB support for SM120 CUTLASS blockwise FP8 GEMM (#38325) 2026-04-03 15:49:59 +02:00
Wentao YeandGitHub c9a9db0e02 [Compile] Fix nvfp4 compile warning (#38573) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-04-01 18:28:57 +00:00
mikaylagawareckiandGitHub 7c080dd3c5 [4/n] Migrate FP4/W4A8 CUTLASS kernels to torch stable ABI (#37503) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
 2026-03-31 10:21:13 -07:00
mikaylagawareckiandGitHub ab1a6a43fa [3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
 2026-03-30 11:20:13 -07:00
mikaylagawareckiandGitHub bf4cc9ed2d [2/n] Migrate per_token_group_quant to torch stable ABI (#36058) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
 2026-03-25 10:15:13 -07:00
mikaylagawareckiandGitHub 8b10e4fb31 [1/n] Migrate permute_cols to libtorch stable ABI (#31509) 
Signed-off-by: Mikayla Gawarecki <mikaylagawarecki@gmail.com>
 2026-03-19 11:27:26 -04:00