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). @ multinode_dynatemp

AlpinDale c2be1b9f29 formatting		5 mēneši atpakaļ
..
attention	0d3562a7f9 MQA in triton FA	5 mēneši atpakaļ
common	3ab36e6b2d feat: extended RoPE for Llama 3.1 (#543)	5 mēneši atpakaļ
distributed	d1a3c7bc2c chore: simplify try-finally logic in pynccl	6 mēneši atpakaļ
endpoints	46159b107a formatting: pt1	6 mēneši atpakaļ
engine	aed64884c6 feat: prompt logprobs with chunked prefill (#539)	5 mēneši atpakaļ
executor	c9d6f9f164 fix formatting	5 mēneši atpakaļ
kv_quant	e42a78381a feat: switch from pylint to ruff (#322)	10 mēneši atpakaļ
lora	b178ae4b4a chore: generalize linear_method to be quant_method (#540)	5 mēneši atpakaļ
modeling	c2be1b9f29 formatting	5 mēneši atpakaļ
processing	aed64884c6 feat: prompt logprobs with chunked prefill (#539)	5 mēneši atpakaļ
quantization	36660b55c2 chore: mixtral fp8 w/ static scales (#542)	5 mēneši atpakaļ
spec_decode	9d81716bfd [v0.5.3] Release Candidate (#388)	8 mēneši atpakaļ
task_handler	aed64884c6 feat: prompt logprobs with chunked prefill (#539)	5 mēneši atpakaļ
transformers_utils	ed759f065d chore: tokenizer_revision -> revision	6 mēneši atpakaļ
__init__.py	199e776722 chore: move ray utils to executor dir	6 mēneši atpakaļ
py.typed	1c988a48b2 fix logging and add py.typed	1 gadu atpakaļ