8000 History for csrc - yx-lamini/vllm · GitHub

Repository navigation

Commits

History for

Commits on Dec 19, 2024
- [Kernel] Refactor Cutlass c3x (#10049)
  
  varun-sundar-rabindranath
  and
  Varun Sundar Rabindranath
  authored
  8936316
- [Bugfix][Build/CI] Fix sparse CUTLASS compilation on CUDA [12.0, 12.2) (#11311)
  tlrmchlsmth
  authored
  5a9da2e
Commits on Dec 18, 2024
- [Kernel]: Cutlass 2:4 Sparsity + FP8/Int8 Quant Support (#10995)
  
  authored
  60508ff
Commits on Dec 13, 2024
- [torch.compile] Dynamic fp8 + rms_norm fusion (#10906)
  
  ProExpertProg
  and
  Varun Sundar Rabindranath
  authored
  30870b4
Commits on Dec 9, 2024
- [V1] Further reduce CPU overheads in flash-attn (#10989)
  WoosukKwon
  authored
  3b61cb4
Commits on Dec 7, 2024
- [BugFix][Kernel]: fix illegal memory access in causal_conv1d when conv_states is None (#10928)
  xffxff
  authored
  78029b3
- [Build] Fix for the Wswitch-bool clang warning (#10060)
  gshtras
  authored
  f13cf9a
Commits on Nov 27, 2024
- [Kernel] Remove if-else with identical branches in marlin 2:4 (#10687)
  tlrmchlsmth
  authored
  e225110
Commits on Nov 26, 2024
- [Feature] vLLM ARM Enablement for AARCH64 CPUs (#9228)
  
  authored
  a6760f6
Commits on Nov 23, 2024
- [AMD] Add support for GGUF quantization on ROCm (#10254)
  kliuae
  authored
  7c25fe4
Commits on Nov 20, 2024
- [Bugfix] Marlin 2:4 temp fix for large M dim (>256) (#10464)
  LucasWilkinson
  authored
  d200972
Commits on Nov 19, 2024
- [Model][Quantization] HQQ support through Marlin kernel expansion (#9766)
  ElizaWszola
  authored
  b00b33d
- Fix: Build error seen on Power Architecture (#10421)
  
  authored
  1ea291a
Commits on Nov 18, 2024
- [Kernel] Initial Machete W4A8 support + Refactors (#9855)
  LucasWilkinson
  authored
  96d999f
Commits on Nov 14, 2024
- Support Roberta embedding models (#9387)
  
  maxdebayser
  and
  flaviabeo
  authored
  4a18fd1
Commits on Nov 13, 2024
- [Core] Flashinfer - Remove advance step size restriction (#10282)
  pavanimajety
  authored
  b6dde33
Commits on Nov 12, 2024
- Splitting attention kernel file (#10091)
  
  maleksan85
  and
  Aleksandr Malyshev
  authored
  812c981
Commits on Nov 8, 2024
- [torch.compile] Fuse RMSNorm with quant (#9138)
  
  ProExpertProg
  and
  youkaichao
  authored
  4f93dfe
Commits on Nov 7, 2024
Commits on Nov 6, 2024
- [CI/Build] drop support for Python 3.8 EOL (#8464)
  aarnphm
  authored
  21063c1
Commits on Oct 31, 2024
- [BugFix][Kernel] Fix Illegal memory access in causal_conv1d in H100 (#9838)
  mzusman
  authored
  9fb12f7
Commits on Oct 27, 2024
- [core] cudagraph output with tensor weak reference (#9724)
  youkaichao
  authored
  8549c82
Commits on Oct 24, 2024
- [Performance][Kernel] Fused_moe Performance Improvement (#9384)
  charlifu
  authored
  5944909
- [Kernel] add kernel for FATReLU (#9610)
  jeejeelee
  authored
  295a061
Commits on Oct 22, 2024
- [Bugfix] Fix spurious "No compiled cutlass_scaled_mm ..." for W8A8 on Turing (#9487)
  LucasWilkinson
  authored
  d1e8240
Commits on Oct 17, 2024
Commits on Oct 16, 2024
- [Bugfix][Kernel] Prevent integer overflow in fp8 dynamic per-token quantize kernel (#9425)
  tlrmchlsmth
  authored
  c3fab5f
- [Kernel][Model] Improve continuous batching for Jamba and Mamba (#9189)
  mzusman
  authored
  fb60ae9
Commits on Oct 10, 2024
- [Bugfix] Fix Machete unittests failing with `NotImplementedError` (#9218)
  LucasWilkinson
  authored
  18511ae
- [Bugfix] Machete garbage results for some models (large K dim) (#9212)
  LucasWilkinson
  authored
  a64e7b9
Commits on Oct 4, 2024
- [Kernel] Zero point support in fused MarlinMoE kernel + AWQ Fused MoE (#8973)
  
  ElizaWszola
  and
  dsikka
  authored
  05d6864

0