AlpinDale 2a60b8f8c9 kernel: do not compile machete for cuda 11 and below (#901) 3 minggu lalu
..
all_reduce f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
attention f831fd8312 rocm: fix compile issues with rocm 6.2 (#893) 3 minggu lalu
backup f8dfac6372 chore: attention refactor and upstream sync apr01 (#365) 9 bulan lalu
core 0256ed236b feat: windows support (#790) 2 bulan lalu
cpu f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
cutlass_extensions 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 1 bulan lalu
hadamard 5d288aa76c feat: add fast hadamard transformation kernels (#232) 11 bulan lalu
mamba 0256ed236b feat: windows support (#790) 2 bulan lalu
moe 0256ed236b feat: windows support (#790) 2 bulan lalu
prepare_inputs f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
punica f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
quantization 2a60b8f8c9 kernel: do not compile machete for cuda 11 and below (#901) 3 minggu lalu
sampling bfc8988116 feat: add cuda sampling kernels for top_k and top_p (#828) 1 bulan lalu
activation_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
cache.h f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
cache_kernels.cu 0256ed236b feat: windows support (#790) 2 bulan lalu
cuda_compat.h f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
cuda_utils.h 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 1 bulan lalu
cuda_utils_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
dispatch_utils.h f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
layernorm_kernels.cu e14223dce5 kernel: use `cub::BlockReduce` instead of custom impl (#895) 3 minggu lalu
ops.h 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 1 bulan lalu
permute_cols.cu 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 1 bulan lalu
pos_encoding_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 4 bulan lalu
torch_bindings.cpp 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 1 bulan lalu