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). @ 05e45aeb53702dac44913e0241b4ac2dabbe3a87

AlpinDale ad24e74a99 feat: FP8 weight-only quantization support for Ampere GPUs		6 mesiacov pred
..
all_reduce	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
attention	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
backup	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	11 mesiacov pred
cpu	271a680026 feat: inference support for PowerPC ISA	7 mesiacov pred
hadamard	5d288aa76c feat: add fast hadamard transformation kernels (#232)	1 rok pred
mamba	5be90c3859 Mamba infrastrucuture support (#586)	6 mesiacov pred
moe	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
punica	4f42985b5c feat: qwen2 lora shapes	7 mesiacov pred
quantization	ad24e74a99 feat: FP8 weight-only quantization support for Ampere GPUs	6 mesiacov pred
activation_kernels.cu	c0c336aaa3 refactor: registry for processing model inputs; quick_gelu; clip model support	7 mesiacov pred
cache.h	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
cache_kernels.cu	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
cuda_compat.h	00acf371f9 rocm: fused topk softmax	7 mesiacov pred
cuda_utils.h	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
cuda_utils_kernels.cu	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
dispatch_utils.h	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
layernorm_kernels.cu	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
ops.h	5be90c3859 Mamba infrastrucuture support (#586)	6 mesiacov pred
pos_encoding_kernels.cu	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
reduction.cuh	aba03b4756 feat: dynamic per-token activation quantization	7 mesiacov pred
registration.h	156f577f79 feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)	7 mesiacov pred
torch_bindings.cpp	ad24e74a99 feat: FP8 weight-only quantization support for Ampere GPUs	6 mesiacov pred