Geçmişin Kaydedilmesi

Yazar SHA1 Mesaj Tarih
  Tri Dao 7bc3f031a4 Compile for both Sm80 and Sm90 1 gün önce
  Tri Dao 68bf390920 Update Cutlass to fix mem fence 2 gün önce
  Tri Dao 7a802796e1 Big refactor and update 2 gün önce
  Driss Guessous 36dddb891c Remove unused 224 configs (#1425) 3 gün önce
  XiaobingZhang 0dfb281743 don't save inputs buffer of FlashAttenFunc to reduce memory usage for inference mode (#1383) 4 hafta önce
  Tri Dao f86e3dd919 [CI] Use MAX_JOBS=1 with nvcc 12.3, don't need OLD_GENERATOR_PATH 1 ay önce
  Tri Dao b7d29fb3b7 Bump to v2.7.2 1 ay önce
  Tri Dao d3b1cd1c13 [CI] Use MAX_JOBS=2 even with nvcc 12.3 1 ay önce
  Tri Dao 61a23ea8a2 Update to Cutlass 3.6.0 1 ay önce
  Tri Dao 9375ac9322 [CI] Don't include <ATen/cuda/CUDAGraphsUtils.cuh> 1 ay önce
  Tri Dao e782d28692 [CI] Change torch #include to make it work with torch 2.1 Philox 1 ay önce
  Tri Dao 073afd5931 [CI] Use torch 2.6.0.dev20241001, reduce torch #include 1 ay önce
  Tri Dao cf0f4c38ef [CI] Fix CUDA version for torch 2.6 1 ay önce
  Tri Dao cc408f9fbf Bump to v2.7.1 1 ay önce
  Tri Dao 88786928ed [CI] Drop Python 3.8, add Python 3.13, add pytorch 2.6.0.dev20241010 1 ay önce
  Adam Louly df1a744887 Add how to import FA3 to documentation (#1112) 1 ay önce
  Michael Melesse b518517cb8 [AMD] Triton Backend for ROCm (#1203) 1 ay önce
  sclarkson 1feb711f46 Fix compilation with clang on ARM64 (#1285) 1 ay önce
  Kai Londenberg 0823cf7b5d Fix FA3 Varlen Performance regression (#1361) 1 ay önce
  Alexander Gessler ca71144d59 flash_bwd_kernel.h: add maybe_unused annotation to suppress lengthy warnings about dtanh not being used in some compile permutations. (#1363) 1 ay önce
  Neil Tenenholtz 7153673c1a Fix swiglu backwards return type (#1337) 1 ay önce
  Tri Dao 641db759ab [CI] Pytorch 2.5.1 does not support python 3.8 1 ay önce
  Tri Dao 7435839e3d Update README for FA3 1 ay önce
  Tri Dao 241c682c9f [CI] Switch back to CUDA 12.4 1 ay önce
  Tri Dao c555642172 Bump to v2.7.0 1 ay önce
  Tri Dao 6ffeb572b1 [CI] Still use CUDA 12.3 but pull the right pytorch version 1 ay önce
  Ethan Steinberg 42f2b8be34 Use CUDA 12.4 in the build system (#1326) 1 ay önce
  Tri Dao 2f6c633179 Drop support for Pytorch 2.0 1 ay önce
  rocking 88d1657a14 [AMD ROCm] Fix KVcache bug and improve performance (#1328) 1 ay önce
  Kai Londenberg 284e2c6e5b Make FA3 paged attention ready for upgrade to Cutlass 3.6 (#1331) 1 ay önce