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). @ 7d0884de9ab1b8cb0a8cd7578bd0c6a7ef7f02f4

AlpinDale 7d0884de9a fix mistral v0.3 weight loading		7 月之前
..
attention	f6250c5516 move dockerfiles to root; fix cpu build	7 月之前
common	e8b7f53321 allow prompt token IDs in the logits processor api	7 月之前
distributed	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
endpoints	e8b7f53321 allow prompt token IDs in the logits processor api	7 月之前
engine	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
executor	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
kv_quant	e42a78381a feat: switch from pylint to ruff (#322)	1 年之前
lora	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
modeling	7d0884de9a fix mistral v0.3 weight loading	7 月之前
processing	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
quantization	e8b7f53321 allow prompt token IDs in the logits processor api	7 月之前
spec_decode	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
task_handler	5b0c11d190 support pipeline parallel pynccl groups	7 月之前
transformers_utils	60e74e92fd add rope_scaling arg	7 月之前
__init__.py	be8154a8a0 feat: proper embeddings API with e5-mistral-7b support	7 月之前
py.typed	1c988a48b2 fix logging and add py.typed	1 年之前