Provider-agnostic, open-source evaluation infrastructure for language models
Latest release v0.5.3 · 9 Dec 2025
3 files for Claude Code: lighteval-porter, openbench, openbench CLAUDE.md — 1,099 tokens loaded in every session.
CLAUDE.md A 1,099 tok .claude/settings.json A — .claude/agents/lighteval-porter.md A — These files are groq/openbench's own configuration — they tell Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.