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