AlpinDale 3bb0f07461 chore: rename `task_handler` to `worker` (#985) 2 ay önce
..
all_reduce f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
attention 3bb0f07461 chore: rename `task_handler` to `worker` (#985) 2 ay önce
backup 3bb0f07461 chore: rename `task_handler` to `worker` (#985) 2 ay önce
core f1ea7711bd core: do not compile ScalarType for torch < 2.4.0 (#938) 2 ay önce
cpu f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
cutlass_extensions 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 3 ay önce
hadamard 5d288aa76c feat: add fast hadamard transformation kernels (#232) 1 yıl önce
mamba 0256ed236b feat: windows support (#790) 4 ay önce
moe 9f3e7c86e2 feat: add fused Marlin MoE kernel (#934) 2 ay önce
prepare_inputs f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
punica f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
quantization 2a60b8f8c9 kernel: do not compile machete for cuda 11 and below (#901) 2 ay önce
sampling bfc8988116 feat: add cuda sampling kernels for top_k and top_p (#828) 3 ay önce
activation_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
cache.h f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
cache_kernels.cu 0256ed236b feat: windows support (#790) 4 ay önce
cuda_compat.h f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
cuda_utils.h 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 3 ay önce
cuda_utils_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
dispatch_utils.h f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
layernorm_kernels.cu e14223dce5 kernel: use `cub::BlockReduce` instead of custom impl (#895) 2 ay önce
ops.h 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 3 ay önce
permute_cols.cu 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 3 ay önce
pos_encoding_kernels.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 6 ay önce
torch_bindings.cpp 93bc863591 feat: Machete Kernels for Hopper GPUs (#842) 3 ay önce