GitHub / AnswerDotAI / cold-compress
Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of GPT-Fast, a simple, PyTorch-native generation codebase.
JSON API: http://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AnswerDotAI%2Fcold-compress
PURL: pkg:github/AnswerDotAI/cold-compress
Stars: 138
Forks: 14
Open issues: 12
License: bsd-3-clause
Language: Python
Size: 8.33 MB
Dependencies parsed at: Pending
Created at: about 1 year ago
Updated at: about 1 month ago
Pushed at: 12 months ago
Last synced at: 3 days ago
Topics: inference, llm, llm-inference, nlp, open-source