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