david/aphrodite-engine: PygmalionAI's large-scale inference engine pygmalion.chat It is designed to serve as the inference endpoint for the PygmalionAI website, and to allow serving the Pygmalion models to a large number of users with blazing fast speeds (thanks to vLLM's Paged Attention). @ eetq_kernels

AlpinDale afb8cfe945 feat: add most of the EETQ kernels, prune later		7 сар өмнө
..
all_reduce	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
attention	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
backup	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	9 сар өмнө
cpu	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
hadamard	5d288aa76c feat: add fast hadamard transformation kernels (#232)	11 сар өмнө
moe	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
punica	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
quantization	afb8cfe945 feat: add most of the EETQ kernels, prune later	7 сар өмнө
activation_kernels.cu	3d6695cfbb feat: add approximate gelu activation kernels (#370)	9 сар өмнө
cache.h	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
cache_kernels.cu	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
cuda_compat.h	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
cuda_utils.h	31c95011a6 feat: FP8 E5M2 KV Cache (#226)	11 сар өмнө
cuda_utils_kernels.cu	31c95011a6 feat: FP8 E5M2 KV Cache (#226)	11 сар өмнө
dispatch_utils.h	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	9 сар өмнө
layernorm_kernels.cu	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
ops.h	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
pos_encoding_kernels.cu	e702f587cf feat: add batched RoPE kernels (#371)	9 сар өмнө
pybind.cpp	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө
reduction.cuh	9d81716bfd [v0.5.3] Release Candidate (#388)	8 сар өмнө