Commits

Branch selector

 main

User selector

tianleiwu

Datepicker

All time

Commit History

Commits on Aug 21, 2025

• [CUDA] Add build flag onnxruntime_USE_FPA_INTB_GEMM (#25802)

  Show description for d346333
  tianleiwu

  authored
  d346333

  Copy full SHA for d346333

Commits on Aug 8, 2025

• Skip node output dump for MemcpyToHost (#25651)

  Show description for 067b154
  tianleiwu

  authored
  067b154

  Copy full SHA for 067b154

Commits on Aug 4, 2025

• Update qMoE spec to support block quantization (#25641)

  Show description for 59871e3
  tianleiwu

  authored
  59871e3

  Copy full SHA for 59871e3

Commits on Aug 2, 2025

• Update MoE and qMoE spec (#25619)

  Show description for 562760a
  tianleiwu

  authored
  562760a

  Copy full SHA for 562760a

Commits on Jul 31, 2025

• [CUDA] BF16 MoE and qMoE (#25572)

  Show description for 68b9d9b
  tianleiwu

  authored
  68b9d9b

  Copy full SHA for 68b9d9b

Commits on Jul 28, 2025

• [CUDA] Support SwiGlu in MoE and qMoE (#25530)

  Show description for a2b4546
  tianleiwu

  authored
  a2b4546

  Copy full SHA for a2b4546

Commits on Jul 17, 2025

• [CUDA] Support head_sink in flash attention for GQA (#25432)

  Show description for e6c84b8
  tianleiwu

  authored
  e6c84b8

  Copy full SHA for e6c84b8

Commits on Jul 16, 2025

• Fix Build Error when tensor dumping is enabled (#25414)

  Show description for 2d6a525
  tianleiwu

  authored
  2d6a525

  Copy full SHA for 2d6a525

Commits on Jul 9, 2025

• [CPU] GQA supports head_sink input for smooth softmax (#25269)

  Show description for cd5f91f
  tianleiwu

  authored
  cd5f91f

  Copy full SHA for cd5f91f

• GatherBlockQuantized supports zero points and 8 bits for uint8 dtype (#25214)

  Show description for ce8796d
  tianleiwu

  authored
  ce8796d

  Copy full SHA for ce8796d

Commits on Jul 8, 2025

• Fix cuda 12.9 windows build (#25317)

  Show description for 7c18d89
  tianleiwu

  authored
  7c18d89

  Copy full SHA for 7c18d89

Commits on Jul 3, 2025

• Fix Windows build of cuda 12.8 (#25282)

  Show description for af35340
  tianleiwu

  authored
  af35340

  Copy full SHA for af35340

Commits on Jun 27, 2025

• [CUDA] Exclude lean attention from linux build (#25203)

  Show description for f3c18ed
  tianleiwu

  authored
  f3c18ed

  Copy full SHA for f3c18ed

• Format *.cu and *.cuh with lintrunner (#25189)

  Show description for 7a6cef6
  tianleiwu

  authored
  7a6cef6

  Copy full SHA for 7a6cef6

Commits on Jun 26, 2025

• [CUDA] bfloat16 MatMulNBits (#25161)

  Show description for 7fdd386
  tianleiwu

  authored
  7fdd386

  Copy full SHA for 7fdd386

Commits on Jun 21, 2025

• update CMAKE_CUDA_ARCHITECTURES for packaging pipelines (#25126)

  Show description for c98a0e0
  tianleiwu

  authored
  c98a0e0

  Copy full SHA for c98a0e0

Commits on Jun 20, 2025

• [CUDA] FpA IntB Gemm Kernel Test (#25109)

  Show description for 7268117
  tianleiwu

  authored
  7268117

  Copy full SHA for 7268117

Commits on Jun 14, 2025

• [CUDA] Fix clip operator (#25057)

  Show description for b3c316f
  tianleiwu

  authored
  b3c316f

  Copy full SHA for b3c316f

Commits on Jun 13, 2025

• Revert "Improve Windows ETW callback registration and fix issues" (#25055)

  Show description for 0784e0a
  

  tianleiwu

  and

  yuslepukhin

  authored
  0784e0a

  Copy full SHA for 0784e0a

Commits on Jun 12, 2025

• [CUDA] Suppress MSVC build warnings for sm=90 (#25020)

  Show description for 70f14d7
  tianleiwu

  authored
  70f14d7

  Copy full SHA for 70f14d7

Commits on Jun 8, 2025

• [CUDA] upgrade cudnn front end to 1.12 (#24977)

  Show description for 89258ae
  tianleiwu

  authored
  89258ae

  Copy full SHA for 89258ae

Commits on Jun 5, 2025

• [CUDA] fp16 intB gemm scale only kernel (#24955)

  Show description for ab5ff6a
  tianleiwu

  authored
  ab5ff6a

  Copy full SHA for ab5ff6a

Commits on Jun 4, 2025

• A temporary fix for layout opt level to unblock react native android CI (#24948)

  Show description for f7ed05c
  tianleiwu

  authored
  f7ed05c

  Copy full SHA for f7ed05c

Commits on Jun 3, 2025

• [CUDA] FpA IntB Gemm Weight Conversion in GPU (#24914)

  Show description for 53ee6c5
  tianleiwu

  authored
  53ee6c5

  Copy full SHA for 53ee6c5

Commits on Jun 2, 2025

• [DNNL] update tests (#24921)

  Show description for 22c2581
  tianleiwu

  authored
  22c2581

  Copy full SHA for 22c2581

Commits on May 30, 2025

• [CUDA] fp16 intB gemm (#24854)

  Show description for 9d6546e
  tianleiwu

  authored
  9d6546e

  Copy full SHA for 9d6546e

Commits on May 23, 2025

• Update MatMulBNits spec and Add Input Checks (#24828)

  Show description for 2bdb57b
  tianleiwu

  authored
  2bdb57b

  Copy full SHA for 2bdb57b

Commits on May 21, 2025

• Remove unused tensor dumper functions (#24821)

  Show description for 39767bf
  tianleiwu

  authored
  39767bf

  Copy full SHA for 39767bf

Commits on May 20, 2025

• support more data types in tensor dumper (#24813)

  Show description for ac0195b
  tianleiwu

  authored
  ac0195b

  Copy full SHA for ac0195b

Commits on May 16, 2025

• [CUDA] Upgrade cutlass to 3.9.2 (#24794)

  Show description for 8983424
  tianleiwu

  authored
  8983424

  Copy full SHA for 8983424

Commits on Apr 29, 2025

• [CUDA] Update sm check for flash attention (#24584)

  Show description for 4adef01
  tianleiwu

  authored
  4adef01

  Copy full SHA for 4adef01

• [CUDA] Fix build for sm<53 (#24582)

  Show description for 76cee36
  tianleiwu

  authored
  76cee36

  Copy full SHA for 76cee36

Commits on Apr 26, 2025

• [CUDA] MatMulNBits benchmark (#24564)

  Show description for 1dd9b99
  tianleiwu

  authored
  1dd9b99

  Copy full SHA for 1dd9b99

Commits on Apr 25, 2025

• [Cuda] support 8 bits in MatMulNBits (#24509)

  Show description for 3a7c8b3
  tianleiwu

  authored
  3a7c8b3

  Copy full SHA for 3a7c8b3

Commits on Apr 19, 2025

• Rename matmul_4bits_quantizer.py to matmul_nbits_quantizer.py (#24472)

  Show description for 0d26928
  tianleiwu

  authored
  0d26928

  Copy full SHA for 0d26928

Pagination

Previous

Next