| .. |
|
all_reduce
|
6305e6f3f2
fix: no repeated IPC registration (#227)
|
11 months ago |
|
attention
|
a788ca33bf
hack in custom bias for attention kernels
|
9 months ago |
|
hadamard
|
5d288aa76c
feat: add fast hadamard transformation kernels (#232)
|
11 months ago |
|
moe
|
7d6ba53602
feat: fused top-k kernels for MoE (#273)
|
10 months ago |
|
punica
|
c0aac15421
feat: S-LoRA support (#222)
|
11 months ago |
|
quantization
|
89c32b40ec
chore: add new imatrix quants (#320)
|
10 months ago |
|
activation_kernels.cu
|
e31c6f0b45
feat: refactor modeling logic and support more models (#274)
|
10 months ago |
|
cache.h
|
9810daa699
feat: INT8 KV Cache (#298)
|
10 months ago |
|
cache_kernels.cu
|
9810daa699
feat: INT8 KV Cache (#298)
|
10 months ago |
|
cuda_compat.h
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
11 months ago |
|
cuda_utils.h
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
11 months ago |
|
cuda_utils_kernels.cu
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
11 months ago |
|
dispatch_utils.h
|
9810daa699
feat: INT8 KV Cache (#298)
|
10 months ago |
|
layernorm_kernels.cu
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
11 months ago |
|
ops.h
|
a788ca33bf
hack in custom bias for attention kernels
|
9 months ago |
|
pos_encoding_kernels.cu
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
11 months ago |
|
pybind.cpp
|
c41462cfcd
feat: exllamav2 quantization (#305)
|
10 months ago |
|
reduction.cuh
|
2755a48d51
merge dev branch into main (#153)
|
1 year ago |
AlpinDale
a788ca33bf
hack in custom bias for attention kernels