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). @ 5175605f8dd052de54a9e270cc499241a9f37008

AlpinDale 5175605f8d fix: yarn (#112)		преди 1 година
..
attention	4e71bd1d12 feat: add PagedAttention V2 kernels (#76)	преди 1 година
quantization	887e03669a feat: add exllamav2 for GPTQ (#99)	преди 1 година
activation.cpp	32844c1522 add GELU kernels and remove compile bloat	преди 1 година
activation_kernels.cu	5175605f8d fix: yarn (#112)	преди 1 година
attention.cpp	4e71bd1d12 feat: add PagedAttention V2 kernels (#76)	преди 1 година
cache.cpp	081545bde6 fix: various CUDA kernel tweaks	преди 1 година
cache_kernels.cu	3d72f05c7b feat: flattened 1D tensor -> 2D tensor (#85)	преди 1 година
cuda_utils.cpp	75c27d3e65 massive overhaul	преди 1 година
cuda_utils_kernels.cu	75c27d3e65 massive overhaul	преди 1 година
dispatch_utils.h	32844c1522 add GELU kernels and remove compile bloat	преди 1 година
layernorm.cpp	081545bde6 fix: various CUDA kernel tweaks	преди 1 година
layernorm_kernels.cu	3d72f05c7b feat: flattened 1D tensor -> 2D tensor (#85)	преди 1 година
pos_encoding.cpp	45f6d9f923 initial refactor commit	преди 1 година
pos_encoding_kernels.cu	5175605f8d fix: yarn (#112)	преди 1 година
quantization.cpp	887e03669a feat: add exllamav2 for GPTQ (#99)	преди 1 година
reduction.cuh	081545bde6 fix: various CUDA kernel tweaks	преди 1 година