Commits

Branch selector

 main

User selector

mgoin

Datepicker

All time

Commit History

Commits on Aug 18, 2025

• Use Blackwell FlashInfer MXFP4 MoE by default if available (#23008)

  Show description for 6d25e3f
  mgoin

  authored
  6d25e3f

  Copy full SHA for 6d25e3f

• [CI Bugfix] Pin `openai<1.100` to unblock CI (#23118)

  Show description for bf75632
  mgoin

  authored
  bf75632

  Copy full SHA for bf75632

Commits on Aug 17, 2025

• [UX] Separate marlin moe config logic from triton moe (#23006)

  mgoin

  authored
  94096a4

  Copy full SHA for 94096a4

Commits on Aug 16, 2025

• [Kernel/Quant] Remove AQLM (#22943)

  Show description for 4fc722e
  

  mgoin

  and

  yewentao256

  authored
  4fc722e

  Copy full SHA for 4fc722e

• [Flaky CI] Increase timeout tolerance for test_mp_crash_detection+test_default_mm_lora_chat_completions (#23028)

  Show description for 3253ae7
  mgoin

  authored
  3253ae7

  Copy full SHA for 3253ae7

• [Bugfix gpt-oss] Fix float32 convert for flashinfer sink support (#23016)

  Show description for 000ccec
  mgoin

  authored
  000ccec

  Copy full SHA for 000ccec

Commits on Aug 15, 2025

• [CI] Speed up Whisper tests by reusing server (#22859)

  Show description for 8a87cd2
  mgoin

  authored
  8a87cd2

  Copy full SHA for 8a87cd2

• Use regex in convert-results-json-to-markdown.py (#22989)

  Show description for a344a1a
  mgoin

  authored
  a344a1a

  Copy full SHA for a344a1a

• [CI Perf] Prune tests in `tests/kernels/attention/` (#22936)

  Show description for 0fe8508
  mgoin

  authored
  0fe8508

  Copy full SHA for 0fe8508

• [CI Perf] Prune tests in `tests/kernels/moe/` (#22939)

  Show description for d2b0e97
  mgoin

  authored
  d2b0e97

  Copy full SHA for d2b0e97

• [CI Perf] Prune tests in `tests/kernels/quantization/` (#22942)

  Show description for 590bddb
  mgoin

  authored
  590bddb

  Copy full SHA for 590bddb

Commits on Aug 13, 2025

• Remove unneeded ROCm platform import when using CUDA (#22765)

  Show description for 4082338
  mgoin

  authored
  4082338

  Copy full SHA for 4082338

• Force TRTLLM attention for gpt-oss on SM100 (#22678)

  Show description for c6b9287
  mgoin

  authored
  c6b9287

  Copy full SHA for c6b9287

• [Bugfix] Fix default enable for CUTLASS MLA on SM100 (#22738)

  Show description for b1361c7
  mgoin

  authored
  b1361c7

  Copy full SHA for b1361c7

• Add hardware plugins to installation doc (#22732)

  Show description for e188592
  

  mgoin

  and

  hmellor

  authored
  e188592

  Copy full SHA for e188592

Commits on Aug 12, 2025

• [CI] Increase timeout for test_completion_with_image_embeds (#22670)

  Show description for 93d0652
  mgoin

  authored
  93d0652

  Copy full SHA for 93d0652

• [CI Failure] Use float32 for tests/entrypoints/openai/test_audio.py (#22686)

  Show description for ea1292a
  mgoin

  authored
  ea1292a

  Copy full SHA for ea1292a

Commits on Aug 7, 2025

• Use float32 for test_completion.py (#22385)

  Show description for a00d8b2
  mgoin

  authored
  a00d8b2

  Copy full SHA for a00d8b2

• Update `flashinfer-python==0.2.10` (#22389)

  Show description for e8961e9
  mgoin

  authored
  e8961e9

  Copy full SHA for e8961e9

Commits on Aug 6, 2025

• [Bugfix] Fix 3D input passed into cutlass_scaled_mm (#22278)

  Show description for 6a51530
  mgoin

  authored
  6a51530

  Copy full SHA for 6a51530

• [Bugfix] Remove faulty test for oot attention backend (#22286)

  Show description for 35509fc
  mgoin

  authored
  35509fc

  Copy full SHA for 35509fc

Commits on Aug 5, 2025

• [CI/Build] Update flashinfer to 0.2.9 (#22233)

  Show description for a7cb610
  mgoin

  authored
  a7cb610

  Copy full SHA for a7cb610

• Use UV_LINK_MODE=copy in Dockerfile to avoid hardlink fail (#22128)

  Show description for c494f96
  mgoin

  authored
  c494f96

  Copy full SHA for c494f96

• [UX] Fail if an invalid attention backend is specified (#22217)

  Show description for e79a12f
  mgoin

  authored
  e79a12f

  Copy full SHA for e79a12f

Commits on Aug 1, 2025

• [CI] Initial tests for SM100 Blackwell runner (#21877)

  Show description for 88faa46
  mgoin

  authored
  88faa46

  Copy full SHA for 88faa46

• Fix pre-commit failure for SECURTIY.md (#22102)

  Show description for 3277e8f
  mgoin

  authored
  3277e8f

  Copy full SHA for 3277e8f

• [Bugfix] Check NVIDIA artifactory is accessible before using flashinfer cubin kernels (#21893)

  mgoin

  authored
  f81c1bb

  Copy full SHA for f81c1bb

• Move flashinfer-python to optional extra `vllm[flashinfer]` (#21959)

  Show description for 0bd409c
  mgoin

  authored
  0bd409c

  Copy full SHA for 0bd409c

Commits on Jul 31, 2025

• [CI Bugfix] Fix CI OOM for `test_shared_storage_connector_hashes` (#21973)

  Show description for 055bd39
  mgoin

  authored
  055bd39

  Copy full SHA for 055bd39

• [UX] Rename CUTLASS_MLA_VLLM_V1 to CUTLASS_MLA (#21966)

  Show description for 6144545
  mgoin

  authored
  6144545

  Copy full SHA for 6144545

• [Example] Add `async_llm_streaming.py` example for AsyncLLM streaming in python (#21763)

  Show description for 9cb497b
  mgoin

  authored
  9cb497b

  Copy full SHA for 9cb497b

Commits on Jul 30, 2025

• [Docs] Update docker.md with HF_TOKEN, new model, and podman fix (#21856)

  mgoin

  authored
  fb58e3a

  Copy full SHA for fb58e3a

Commits on Jul 29, 2025

• Add `flashinfer_python` to CUDA wheel requirements (#21389)

  Show description for a33ea28
  mgoin

  authored
  a33ea28

  Copy full SHA for a33ea28

• [CI] Parallelize Kernels MoE Test (#21764)

  Show description for afa2607
  mgoin

  authored
  afa2607

  Copy full SHA for afa2607

• [Docs] Minimize spacing for supported_hardware.md table (#21779)

  mgoin

  authored
  947e982

  Copy full SHA for 947e982

Pagination

Previous

Next