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). @ linear_weights_light

AlpinDale 8cf8fa3f09 don't assign meta tensor to a cuda param		vor 7 Monaten
..
all_reduce	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
attention	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
backup	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	vor 9 Monaten
cpu	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
hadamard	5d288aa76c feat: add fast hadamard transformation kernels (#232)	vor 11 Monaten
moe	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
punica	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
quantization	8cf8fa3f09 don't assign meta tensor to a cuda param	vor 7 Monaten
activation_kernels.cu	3d6695cfbb feat: add approximate gelu activation kernels (#370)	vor 9 Monaten
cache.h	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
cache_kernels.cu	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
cuda_compat.h	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
cuda_utils.h	31c95011a6 feat: FP8 E5M2 KV Cache (#226)	vor 11 Monaten
cuda_utils_kernels.cu	31c95011a6 feat: FP8 E5M2 KV Cache (#226)	vor 11 Monaten
dispatch_utils.h	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	vor 9 Monaten
layernorm_kernels.cu	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
ops.h	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
pos_encoding_kernels.cu	e702f587cf feat: add batched RoPE kernels (#371)	vor 9 Monaten
pybind.cpp	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten
reduction.cuh	9d81716bfd [v0.5.3] Release Candidate (#388)	vor 8 Monaten