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). @ feat/jamba-support

AlpinDale a1f18f17e6 modify the cache engine and model runner/worker to support mamba states		il y a 8 mois
..
__init__.py	04b53d2db5 chore: add initializer files	il y a 1 an
cache_engine.py	a1f18f17e6 modify the cache engine and model runner/worker to support mamba states	il y a 8 mois
cpu_model_runner.py	6e0761ba5d make init_distributed_environment compatible with init_process_group	il y a 9 mois
cpu_worker.py	083ba7b452 roll back chunked prefill changes to SDPA, isolate cpu worker	il y a 9 mois
model_runner.py	a1f18f17e6 modify the cache engine and model runner/worker to support mamba states	il y a 8 mois
neuron_model_runner.py	0f1399c135 feat: attention refactor part 2	il y a 9 mois
neuron_worker.py	4d33ce60da feat: Triton flash attention backend for ROCm (#407)	il y a 9 mois
worker.py	a1f18f17e6 modify the cache engine and model runner/worker to support mamba states	il y a 8 mois
worker_base.py	8c67b37131 fix docstrings	il y a 9 mois