FasterDecoding

FasterDecoding

Geek Repo

Think deeper, decode faster

Location:United States of America

Github PK Tool:Github PK Tool

FasterDecoding's repositories

Medusa

Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads

Language:Jupyter NotebookLicense:Apache-2.0Stargazers:2059Issues:34Issues:79
Language:Jupyter NotebookLicense:Apache-2.0Stargazers:170Issues:4Issues:5

REST

REST: Retrieval-Based Speculative Decoding, NAACL 2024

Language:CLicense:Apache-2.0Stargazers:152Issues:6Issues:13