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). @ a2d476183f95115d19b4a57a8b4de6142a46f769

AlpinDale a2d476183f fix: remove scipy and re-implement CSR matrix		7 月之前
..
blocksparse_attention	a2d476183f fix: remove scipy and re-implement CSR matrix	7 月之前
__init__.py	9d81716bfd [v0.5.3] Release Candidate (#388)	10 月之前
ipex_attn.py	805fa8721d feat: use intel_extension_for_pytorch for CPU backend	7 月之前
paged_attn.py	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 月之前
prefix_prefill.py	cda0e93a10 abstract away the platform for device capability	7 月之前
triton_flash_attn.py	8e11259e90 missing triton autoconfig for rocm flash attn	8 月之前