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). @ bitnet

AlpinDale 5b0c11d190 support pipeline parallel pynccl groups		5 ヶ月前
..
__init__.py	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	9 ヶ月前
cpu_executor.py	ef733aee43 implement ExecuteModelData to reduce executor complexity	6 ヶ月前
distributed_gpu_executor.py	de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead	5 ヶ月前
executor_base.py	de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead	5 ヶ月前
gpu_executor.py	236be273e5 feat: tensor parallel speculative decoding (#554)	6 ヶ月前
multiproc_gpu_executor.py	5b0c11d190 support pipeline parallel pynccl groups	5 ヶ月前
multiproc_worker_utils.py	eaa06fdd14 fix some f-strings	6 ヶ月前
neuron_executor.py	ef733aee43 implement ExecuteModelData to reduce executor complexity	6 ヶ月前
ray_gpu_executor.py	de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead	5 ヶ月前
ray_utils.py	c6a501f682 add multiprocessing executor; make ray optional	6 ヶ月前