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). @ ec5b99d075c728d0ab4188ce1f277d24b45d389c

AlpinDale ec5b99d075 fix: use named args		7 bulan lalu
..
attention	75f97bc25d bump flash-attn to remove unnecessary copies in the backend	7 bulan lalu
common	237fa59aea feat: support CPU/GPU swapping in BlockManagerV2	7 bulan lalu
distributed	b2fd915c35 improve p2p access check	7 bulan lalu
endpoints	d00a7517e6 fix: tokenizer delay with using LLM class	7 bulan lalu
engine	ec5b99d075 fix: use named args	7 bulan lalu
executor	05d6e43244 fix: `torch.compile()` with mp executor backend	7 bulan lalu
kv_quant	e42a78381a feat: switch from pylint to ruff (#322)	1 tahun lalu
lora	5fecc6b025 when was this deprecated?	7 bulan lalu
modeling	39b36efabf fix: mixtral fp8 ckpt loading	7 bulan lalu
multimodal	75f97bc25d bump flash-attn to remove unnecessary copies in the backend	7 bulan lalu
processing	237fa59aea feat: support CPU/GPU swapping in BlockManagerV2	7 bulan lalu
quantization	39b36efabf fix: mixtral fp8 ckpt loading	7 bulan lalu
spec_decode	ec5b99d075 fix: use named args	7 bulan lalu
task_handler	e321d80e4e fix: `prompt_logprobs==0` case	7 bulan lalu
transformers_utils	8d77c69cbd feat: support image processor and add llava example	7 bulan lalu
__init__.py	be8154a8a0 feat: proper embeddings API with e5-mistral-7b support	7 bulan lalu
py.typed	1c988a48b2 fix logging and add py.typed	1 tahun lalu