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).

AlpinDale 5c3b94de45 spec decode: move ops.advane_step to flash attention backend (#1005)		2 týždňov pred
..
__init__.py	9d81716bfd [v0.5.3] Release Candidate (#388)	8 mesiacov pred
batch_expansion.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
draft_model_runner.py	5c3b94de45 spec decode: move ops.advane_step to flash attention backend (#1005)	2 týždňov pred
interfaces.py	f1d0b77c92 [0.6.0] Release Candidate (#481)	4 mesiacov pred
medusa_worker.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
metrics.py	2f61644f6e SPMD optimizations (#824)	1 mesiac pred
mlp_speculator_worker.py	0dfa6b60ec core: support logprobs with multi-step scheduling (#963)	2 týždňov pred
multi_step_worker.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
ngram_worker.py	0dfa6b60ec core: support logprobs with multi-step scheduling (#963)	2 týždňov pred
proposer_worker_base.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
smaller_tp_proposer_worker.py	0dfa6b60ec core: support logprobs with multi-step scheduling (#963)	2 týždňov pred
spec_decode_worker.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
target_model_runner.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	2 týždňov pred
top1_proposer.py	0dfa6b60ec core: support logprobs with multi-step scheduling (#963)	2 týždňov pred
util.py	0dfa6b60ec core: support logprobs with multi-step scheduling (#963)	2 týždňov pred