44 Commits
Author SHA1 Message Date
Harry MellorClaude Opus 5.5mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>Kevin H. Luu
b965ee564d [Chore] Use upstream Transformers DeepseekV4Config (#61392)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
2026-10-15 01:08:08 -07:00
Ronaldandaoshen02 19aba3e4d5 [Bugfix][RL] Load attention sinks through weight_loader for layerwise reload (#57797)
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
2026-10-14 10:17:58 +00:00
Varun Sundar Rabindranath 5463fe4962 [KV-Offloading][TP] : Expand replicated_layout detection to multi-group MLA (#57652) 2026-09-30 09:11:03 -04:00
Yongye ZhuandClaude Opus 5.5 5ff3bbfb08 [Perf][DSv4] Fuse inverse RoPE + FP8 quant into FlashInfer sparse MLA (#58621)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-24 23:26:50 -07:00
Harry MellorandClaude Opus 5 8c1557a79c Add pydocstyle to the ruff rules (#52136)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Harry Mellor <hej.mellor@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 16:44:29 +00:00
Shanshan Shen a6c5d6d0fc [ROCm][Perf] Enable CSA multi-stream overlap for DeepSeek-V4 (#51794)
Signed-off-by: shen-shanshan <467638484@qq.com>
Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com>
2026-09-14 11:21:30 +00:00
Roberto L. CastroandCodex 120ec4ebd2 [5/N][warmup][DSv4] Migrate NVIDIA CuTeDSL attention kernels (#53566)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: Codex <codex@openai.com>
2026-09-12 00:05:59 -04:00
Roberto L. CastroandCodex 6ddbab03de [4/N][warmup][DSv4] Migrate common attention kernels (#50176)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: Codex <codex@openai.com>
2026-09-08 10:25:48 -04:00
fc8f10792c Fix DeepSeek V4 FlashMLA auto KV cache dtype (#45091)
Signed-off-by: elehayym <huangxh1618@gmail.com>
Signed-off-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
Co-authored-by: elehayym <huangxh1618@gmail.com>
Co-authored-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
2026-09-03 19:29:14 +00:00
Isotr0py 1356635d83 [New model][Multimodal] Add DeepSeek-V4-Flash-Vision-Exp support (#54566)
Signed-off-by: Isotr0py <Isotr0py@outlook.com>
2026-09-02 11:09:09 -07:00
Roberto L. CastroandCodex e16b5e518d [1/N][warmup][DSv4] Migrate generic MLA metadata and indexing kernels (#50175)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: Codex <codex@openai.com>
2026-08-31 23:26:05 -04:00
Shanshan Shen 32ad1400d7 [ROCm][Perf] Fuse SWA q/kv RMSNorm and q FP8 group quant for DeepSeek-V4 (#53540)
Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com>
2026-08-27 09:37:54 -07:00
Woosuk Kwon 0a5ad6f0d4 [Model] Remove unused DeepSeek V4 top-k buffer helper (#53697) 2026-08-26 11:59:09 -07:00
Lucas Wilkinson 8bdc70ec7b [6/N][KV-Cache Layout Refactor] Standardize KV cache layout (#51718)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
2026-08-21 17:47:49 -07:00
Fangzhou Ai fe76112ff2 [ROCm][Perf] Optimize DeepSeek V4 C4A top-k with AITER (#52882) 2026-08-21 10:00:39 -07:00
Woosuk KwonandOpenAI Codex f1178f3a06 Revert DSv4 eager workspace reuse (#52836)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
2026-08-18 17:02:59 -07:00
292187dd8c [Bugfix][DSv4] Keep indexer scoring in breakable graphs (#52492)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
2026-08-17 03:01:13 +00:00
Yongye ZhuandClaude Opus 5 967e104fad [Config] Unify indexer cache dtype under attention_config.indexer_kv_dtype (#52550)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-17 10:20:04 +08:00
Nick Hill 8efa13b700 [Bugfix] Pick the DeepSeek V4 eager cudagraph region per model runner (#52401) 2026-08-15 21:04:08 -07:00
Lucas WilkinsonandClaude 57bd0ed441 [5/N][KV-Cache Layout Refactor] Backend-published KV packing via customize_spec (#51704)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
2026-08-14 09:47:55 -04:00
stefankoncarevicandAndreas Karatzas b369f10d5c [Bugfix][ROCm][CI] Restore the DeepSeek-V4 input GEMM override point (#51821)
Signed-off-by: Stefan Koncarevic <Stefan.Koncarevic@amd.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
2026-08-12 20:33:52 -05:00
Lucas WilkinsonOpenAI CodexClaude Opus 5Benjamin Chislettmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>Nick HillOpenAI Codex
7f7a32cfec [Spec Decode] DSpark confidence-scheduled verification (#47808)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Signed-off-by: Lucas Wilkinson <wilkinson.lucas@gmail.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
2026-08-12 12:27:48 -07:00
Woosuk KwonandOpenAI Codex 79c865b838 [Perf] Narrow DeepSeek V4 eager CUDA graph region (#51430)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
2026-08-10 10:52:10 -07:00
Wentao Yeandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com> df71917cf1 [DSv4 Perf] Optimize workspace reuse for eager break, 3.9% E2E TTFT improvement. (#49236)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-07-31 08:45:58 -07:00
12a34a6bc7 [ROCm][DSV4] B-preshuffle the attention fp8 projections (#46720)
Signed-off-by: Mehmet Cagri Kaymak <mehmet.kaymak@amd.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
2026-07-30 13:59:52 -05:00
+13 aeeb36b1f1 [New model] Kimi K3 (#50000)
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: Jee Jee Li <jeejeelee@inferact.ai>
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Signed-off-by: wzhao18 <wzhao18.sz@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>
Co-authored-by: Wei Zhao <51183510+wzhao18@users.noreply.github.com>
2026-07-30 03:49:58 -07:00
Wentao Ye b0cb1da1bd [DSv4 Perf] Skip topk and router when not needed, 3.4% E2E TTFT improvement for Decode case (#49486)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
2026-07-23 13:08:08 -04:00
Summer Yang f3a920a076 [Core][DSV4] Compact MXFP4 indexer KV cache and packed group overlays (#48993) 2026-07-22 16:58:44 -07:00
Jeff (Junze) MaandLucas Wilkinson 80eb01e93d [Bugfix] DSV4 TP16 garbage output (#47493)
Signed-off-by: Jeff Ma <jeffjma@umich.edu>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
2026-07-07 21:04:33 -07:00
04adc8843b [Bugfix]Fix DeepSeek-V4 fp8_ds_mla KV cache reshape (#47716)
Co-authored-by: yy-fighting <23518844576@qq.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
2026-07-06 12:56:44 -07:00
Kunshang JiandHarry Mellor 07f9baf756 Revert "[Platform] Replace torch.cuda.Event with torch.Event (#47140)" (#47668)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2026-07-06 14:18:33 +01:00
Kunshang JiandHarry Mellor e840f0d3f5 [Platform] Replace torch.cuda.Event with torch.Event (#47140)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2026-06-30 08:39:59 -07:00
44d95069e9 Enable DeepSeek V4 and GLM-5.1 on SM120 (#43477)
Signed-off-by: Zihua Wu <13583761+lucifer1004@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
2026-06-22 11:54:14 -07:00
Woosuk KwonandClaude Opus 4.8 1797576237 Revert "[DSV4 Perf] Optimize dsv4 cudagraph by reducing eager_break_during_capture" (#45309) (#45972)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-17 17:20:59 -07:00
Wentao Ye 2a47a9ff0f [DSV4 Perf] Optimize dsv4 cudagraph by reducing eager_break_during_capture, 26.8% ~ 27.9% E2E TTFT improvement (#45309)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
2026-06-17 09:34:53 -07:00
Woosuk Kwon 02d2da0748 [DSV4] Move more ops out of eager breakpoint (#44561) 2026-06-05 06:42:41 -07:00
Woosuk Kwon 4efd6ffde0 [DSV4] Refactor DeepseekV4Attention (#44569)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
2026-06-04 20:23:07 -07:00
Yongye Zhu b5235fca2e [DSv4] Adding TRTLLM gen attention kernel (#43827) 2026-06-04 07:35:09 -07:00
Woosuk Kwon 8c3cc98cff [DSV4] Remove unncessary classes & functions (#44246)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
2026-06-01 14:43:00 -07:00
Woosuk Kwon 69b8956dcd [Model Refactoring] Remove unncessary torch op registration for DSv4 (#43891)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
2026-05-28 14:04:55 -07:00
Yongye Zhu 6ab6ffb428 [Feat][DSV4] Fuse q pad into deepseek v4 fused kernel (#43162) 2026-05-26 05:12:54 -10:00
Yongye Zhu 843715739b [Refactor] Extract DeepSeek V4 sparse MLA impl into model folder (#43149) 2026-05-22 10:06:31 -07:00
Woosuk Kwon b14be81c1f [Model Refactoring] Move deepseek_v4_ops to models/deepseek_v4 [3/N] (#43073)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
2026-05-19 00:52:54 -07:00
Woosuk Kwon 87b08c5f64 [Model Refactoring] Move DeepSeek V4 layers to models/deepseek_v4/ [2/N] (#43039)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
2026-05-18 21:00:58 -07:00