8000 History for csrc - tongyx361/vllm · GitHub

Repository navigation

Commits

History for

Commits on Aug 22, 2025
- [Kernel] Add FP8 support with FlashMLA backend (#22668)
  MatthewBonanni
  authored
  19fe1a0
Commits on Aug 21, 2025
Commits on Aug 20, 2025
- [Kernel/Quant] Remove the original marlin format and qqq (#23204)
  mgoin
  authored
  0cdbf5e
- [Kernel] CUTLASS MoE FP8: Integrate cuda moe permute/unpermute (#23045)
  shixianc
  authored
  b17109b
Commits on Aug 18, 2025
- [Bugfix][CI] Machete kernels: deterministic ordering for more cache hits (#23055)
  andylolu2
  authored
  b2fd0b8
Commits on Aug 17, 2025
- [Kernel] Add cuda kernel for gpt_oss activation (#22951)
  jeejeelee
  authored
  4d4061b
Commits on Aug 16, 2025
- [Kernel/Quant] Remove AQLM (#22943)
  
  mgoin
  and
  yewentao256
  authored
  4fc722e
Commits on Aug 15, 2025
- [Fix] enable swap_ab for pplx problem size computation (#22991)
  
  shixianc
  and
  Shixian Cui
  authored
  7f89ed2
- [V0 Deprecation] Remove advance_step (#22969)
  WoosukKwon
  authored
  1c859a1
- Revert "[Kernel] Add cuda kernel for gpt_oss activation" (#22948)
  simon-mo
  authored
  f1f0d2f
- [Kernel] Add cuda kernel for gpt_oss activation (#22538)
  jeejeelee
  authored
  81f4b96
Commits on Aug 14, 2025
- [Kernel] [Quantization] Add MXFP4 and bias support for marlin kernel (#22428)
  
  authored
  33c63e9
Commits on Aug 13, 2025
- [ROCm][Bugfix] Fix compilation error in topk softmax fused kernel (#22819)
  kliuae
  authored
  c6cd5ca
- [Perf] Support topk softmax fused kernel for broader num_experts (#22211)
  
  shixianc
  and
  Shixian Cui
  authored
  4c558cf
Commits on Aug 9, 2025
- [ROCm][Misc] Rename the context_len to seq_len in ROCm custom paged attention kernel (#22097)
  charlifu
  authored
  b7c0942
Commits on Aug 8, 2025
- [Kernel] Add support for block FP8 on SM120 (NVIDIA 5090 and RTX PRO 6000) (#22131)
  0xjunhao
  authored
  3303f13
Commits on Aug 7, 2025
- [v1] - Mamba1 Attention Metadata (#21249)
  
  Josephasafg
  and
  asafgardin
  authored
  46a1394
Commits on Aug 5, 2025
- [Feature] Non-contiguous Support for FP8 Quantization (#21961)
  
  yewentao256
  and
  mgoin
  authored
  4771df7
Commits on Aug 3, 2025
- Support CUTLASS NVFP4 (w4a4) for Blackwell Geforce GPUs (SM120) (#21309)
  LopezCastroRoberto
  authored
  789562c
Commits on Aug 1, 2025
- [Perf] Optimize `reshape_and_cache_flash` CUDA Kernel (#22036)
  yewentao256
  authored
  eefbf4a
- [Refactor] Fix Compile Warning #1444-D (#21462)
  yewentao256
  authored
  a59cd9d
Commits on Jul 30, 2025
Commits on Jul 29, 2025
- [AMD][CI/Build][Bugfix] Guarding CUDA specific functions by ifndef ROCM (#21766)
  gshtras
  authored
  12a223e
Commits on Jul 28, 2025
- [Kernel] SM90 CUTLASS FP8 GEMM: add support for swap AB + kernel tuning (#20396)
  
  LyrisZhong
  and
  djmmoss
  authored
  c6c9122
Commits on Jul 27, 2025
- Fix CUDA permute/unpermute for use with DeepGemm Moe (#17934)
  CalebDu
  authored
  57c22e5
Commits on Jul 26, 2025
- [Bugfix][Apple Silicon] fix missing symbols when build from source on Mac with Apple Silicon (#21380)
  zhouyeju
  authored
  9094d11
- [Perf] Cuda Kernel for Int8 Per Token Group Quant (#21476)
  yewentao256
  authored
  75d29cf
Commits on Jul 25, 2025
- [Kernel] Improve machete memory bound perf (#21556)
  czhu-cohere
  authored
  136d750
Commits on Jul 24, 2025
- [Bug] Fix Compressed Tensor NVFP4 `cutlass_fp4_group_mm` illegal memory access (#21465)
  yewentao256
  authored
  e8cb0d0
- [Bugfix][ROCm] Fix for warp_size uses on host (#21205)
  gshtras
  authored
  90eeea8
Commits on Jul 23, 2025
- [Bugfix][ROCm][Build] Fix build regression on ROCm (#21393)
  gshtras
  authored
  3ec7170

0