Repository navigation

Commits

Commits on Sep 23, 2026
- [Perf][Attention] Avoid CPU-GPU sync in DCP sequence lengths (#58169)
  
  ZJY0516
  and
  codex
  authored
  6632ed5
Commits on Sep 17, 2026
- [Test][Core] Add hybrid model prefix cache hit-rate coverage (#57127)
  
  ZJY0516
  and
  codex
  authored
  b1f9ebc
Commits on Sep 16, 2026
- [Refactor][GLM-5.3-Flash] Move sparse_attn_indexer_kpool into the model folder and split AMD/NVIDIA (#55358)
  
  ZJY0516
  and
  tjtanaa
  authored
  4fe9e6f
Commits on Sep 14, 2026
- [Bugfix] Trim stale consequence claims from unannotated-eagle warning (#56791)
  
  ZJY0516
  and
  Kimi Code
  authored
  47fbd36
Commits on Sep 13, 2026
- [Perf][Kernel] Integrate DeepSelect TopK for the DSA sparse indexer (#56464)
  
  authored
  fa008bd
Commits on Sep 12, 2026
- [CI] Bump CUTLASS DSL to 4.7 (#54927)
  
  ZJY0516
  and
  codex
  authored
  ec6b049
Commits on Sep 8, 2026
- [Bugfix] Fix NVFP4 fused SiLU+mul scale allocation and g 10BC8 lobal scale direction (#55643)
  
  authored
  8ebc5b0
Commits on Sep 7, 2026
- [CI] fix pre-commit (#55630)
  ZJY0516
  authored
  f43ef15
Commits on Sep 3, 2026
- [Model] add GLM-5.3-Flash support (#53906)
  ZJY0516
  authored
  98ed085
Commits on Aug 30, 2026
- [Bugfix] Avoid global config lookup in sparse indexer forward (#54400)
  
  ZJY0516
  and
  codex
  authored
  8c51b92
Commits on Aug 27, 2026
- [Bugfix][KV Offload] Handle padded GPU cache storage (#54021)
  
  ZJY0516
  and
  codex
  authored
  478ec3e
Commits on Aug 17, 2026
- [K3] support recoverssm for K3 (#51855)
  
  authored
  70afded
Commits on Aug 11, 2026
- [Bugfix] Align Qwen GDN gates with speculative tokens (#51812)
  
  ZJY0516
  and
  codex
  authored
  5af7c8d
Commits on Aug 10, 2026
- [V1][Scheduler] Apply Mamba alignment before encoder caps (#51603)
  
  authored
  243c63b
Commits on Aug 8, 2026
- [CI] Refresh hybrid Model Runner V2 coverage (#51410)
  ZJY0516
  authored
  44351f8
Commits on Aug 7, 2026
- [CI] Fix Batch Invariance (B200) (#51417)
  ZJY0516
  authored
  27d7303
Commits on Aug 4, 2026
- [Mamba] enable prefix cache by default (#50991)
  
  ZJY0516
  and
  mergify[bot]
  authored
  f9c74b4
Commits on Aug 3, 2026
- [UX] remove torch compile warning when using breakable cudagraph (#50750)
  ZJY0516
  authored
  32c42c4
Commits on Jul 30, 2026
- [New model] Kimi K3 (#50000)
  
  authored
  aeeb36b
Commits on Jul 29, 2026
- [Model] Add Kimi K3 support: model files and kernels [1/N] (#50089)
  
  authored
  7c6729b
Commits on Jul 28, 2026
- [CI] Increase Qwen3.5 MTP GSM8K generation length (#49881)
  
  authored
  25ace8f
Commits on Jul 24, 2026
- [CI] Stabilize memory-sensitive compile and structured output tests (#49749)
  
  ZJY0516
  and
  codex
  authored
  9e6746b
Commits on Jul 12, 2026
- [CI][2/N] reduce CI time (#48394)
  ZJY0516
  authored
  370b678
- [2/N][Core] support partial prefix cache hit for hybrid model (#46384)
  
  ZJY0516
  and
  ivanium
  authored
  481e481
Commits on Jul 11, 2026
- [CI] split tests to reduce CI time (#48219)
  ZJY0516
  authored
  1ef1c7e
Commits on Jul 10, 2026
- [Perf] fuse more rmsnorm and all-reduce in qwen3.5 (#46998)
  ZJY0516
  authored
  300e337
Commits on Jul 1, 2026
- [Attention][DSA] support dcp for FLASHINFER_MLA_SPARSE (#46076)
  
  authored
  c5200d3
Commits on Jun 22, 2026
- [1/N][Core] add partial prefix cache primitives (#45939)
  
  ZJY0516
  and
  ivanium
  authored
  6bc6f2d
Commits on Jun 11, 2026
- [Attention] add triton diff-kv backend for mimo (#41797)
  ZJY0516
  authored
  f81daf8
Commits on Jun 9, 2026
- [Bugfix] fix qwen3.5 ep weight loading (#45002)
  ZJY0516
  authored
  ca4cfd8
- [Perf] fuse qk rmsnorm rope gate for qwen3.5 (#44176)
  ZJY0516
  authored
  7a89b72
Commits on Jun 4, 2026
- [mamba] unify KDA conv states into one cache to match 2-state SSM layout (#44539)
  ZJY0516
  authored
  a55fccf
Commits on Jun 2, 2026
- [Perf] use triton moe backend on hopper by default (#44220)
  ZJY0516
  authored
  dcdfe66
Commits on May 20, 2026
- [Misc] downgrade nvidia-cutlass-dsl to 4.5.0 (#43230)
  ZJY0516
  authored
  6dc0a71
Commits on May 16, 2026
- [Experimental] Breakable CUDA graph (#42304)
  ZJY0516
  authored
  8a56da3

Branch selector

User selector

Datepicker

Commit history

Commits on Sep 23, 2026

Commits on Sep 17, 2026

Commits on Sep 16, 2026

Commits on Sep 14, 2026

Commits on Sep 13, 2026

Commits on Sep 12, 2026

Commits on Sep 8, 2026

Commits on Sep 7, 2026

Commits on Sep 3, 2026

Commits on Aug 30, 2026

Commits on Aug 27, 2026

Commits on Aug 17, 2026

Commits on Aug 11, 2026

Commits on Aug 10, 2026

Commits on Aug 8, 2026

Commits on Aug 7, 2026

Commits on Aug 4, 2026

Commits on Aug 3, 2026

Commits on Jul 30, 2026

Commits on Jul 29, 2026

Commits on Jul 28, 2026

Commits on Jul 24, 2026

Commits on Jul 12, 2026

Commits on Jul 11, 2026

Commits on Jul 10, 2026

Commits on Jul 1, 2026

Commits on Jun 22, 2026

Commits on Jun 11, 2026

Commits on Jun 9, 2026

Commits on Jun 4, 2026

Commits on Jun 2, 2026

Commits on May 20, 2026

Commits on May 16, 2026