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). @ sampling-kernels

AlpinDale 709628a74d fix		5 bulan lalu
..
fused_moe	9be43994fe feat: fbgemm quantization support (#601)	5 bulan lalu
mamba	2dfa4e47e6 chore: set seed for dummy weights init	5 bulan lalu
ops	fca911ee0a vLLM Upstream Sync (#526)	7 bulan lalu
__init__.py	07aa2a492f upstream: add option to specify tokenizer	1 tahun lalu
activation.py	c0c336aaa3 refactor: registry for processing model inputs; quick_gelu; clip model support	6 bulan lalu
layernorm.py	5761ef8c35 feat: gemma-2 support	5 bulan lalu
linear.py	ba371fbbbd feat: AWQ marlin kernels (#603)	5 bulan lalu
logits_processor.py	5761ef8c35 feat: gemma-2 support	5 bulan lalu
pooler.py	be8154a8a0 feat: proper embeddings API with e5-mistral-7b support	6 bulan lalu
rejection_sampler.py	2c653a2268 fix: make speculative decoding work with per-request seed	5 bulan lalu
rotary_embedding.py	5761ef8c35 feat: gemma-2 support	5 bulan lalu
sampler.py	709628a74d fix	5 bulan lalu
spec_decode_base_sampler.py	2c653a2268 fix: make speculative decoding work with per-request seed	5 bulan lalu
typical_acceptance_sampler.py	2c653a2268 fix: make speculative decoding work with per-request seed	5 bulan lalu
vocab_parallel_embedding.py	9be43994fe feat: fbgemm quantization support (#601)	5 bulan lalu