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