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 d9d287a288 rocm: enable multi-step scheduling for rocm (#1071)		3 veckor sedan
..
__init__.py	9d81716bfd [v0.5.3] Release Candidate (#388)	8 månader sedan
abstract.py	a985143768 core: add cuda graph support for encoder-decoder models (#1051)	3 veckor sedan
blocksparse_attn.py	1405051912 attention: add `AttentionState` abstraction (#863)	1 månad sedan
flash_attn.py	1390915778 multi-step: add support for flashinfer attention backend (#1033)	4 veckor sedan
flashinfer.py	a985143768 core: add cuda graph support for encoder-decoder models (#1051)	3 veckor sedan
ipex_attn.py	6951928522 xpu: bump IPEX to 2.3, support GQA (#1042)	4 veckor sedan
openvino.py	1405051912 attention: add `AttentionState` abstraction (#863)	1 månad sedan
pallas.py	032974a28a tpu: fix TPU type api (#975)	1 månad sedan
placeholder_attn.py	3bb0f07461 chore: rename `task_handler` to `worker` (#985)	1 månad sedan
rocm_flash_attn.py	d9d287a288 rocm: enable multi-step scheduling for rocm (#1071)	3 veckor sedan
torch_sdpa.py	1405051912 attention: add `AttentionState` abstraction (#863)	1 månad sedan
utils.py	a985143768 core: add cuda graph support for encoder-decoder models (#1051)	3 veckor sedan
xformers.py	1405051912 attention: add `AttentionState` abstraction (#863)	1 månad sedan