| .. |
|
all_reduce
|
6305e6f3f2
fix: no repeated IPC registration (#227)
|
1 yıl önce |
|
attention
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
hadamard
|
5d288aa76c
feat: add fast hadamard transformation kernels (#232)
|
1 yıl önce |
|
punica
|
c0aac15421
feat: S-LoRA support (#222)
|
1 yıl önce |
|
quantization
|
c3a221eb02
feat: GGUF, QuIP#, and Marlin support (#228)
|
1 yıl önce |
|
activation_kernels.cu
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
1 yıl önce |
|
cache.h
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
cache_kernels.cu
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
cuda_compat.h
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
1 yıl önce |
|
cuda_utils.h
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
cuda_utils_kernels.cu
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
dispatch_utils.h
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
1 yıl önce |
|
layernorm_kernels.cu
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
1 yıl önce |
|
ops.h
|
c3a221eb02
feat: GGUF, QuIP#, and Marlin support (#228)
|
1 yıl önce |
|
pos_encoding_kernels.cu
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
1 yıl önce |
|
pybind.cpp
|
c3a221eb02
feat: GGUF, QuIP#, and Marlin support (#228)
|
1 yıl önce |
|
reduction.cuh
|
2755a48d51
merge dev branch into main (#153)
|
1 yıl önce |
AlpinDale
5d288aa76c
feat: add fast hadamard transformation kernels (#232)