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). @ 71a26f0998ce6eac8a8ec6f49484a1a857e18fbb

AlpinDale 71a26f0998 chore: use pytorch sdpa backend to do naive attention for rocm		7 месяцев назад
..
backends	71a26f0998 chore: use pytorch sdpa backend to do naive attention for rocm	7 месяцев назад
ops	ced1b36b8b feat: support head size of 192	7 месяцев назад
__init__.py	a94de94c44 refactor: combine the prefill and decode into a single API (#553)	7 месяцев назад
layer.py	ac79d115b3 add guards for prefix caching, fp8, chunked, etc	7 месяцев назад
selector.py	696f2cd59c add phi3_small support with blocksparse attention	7 месяцев назад