AISimon Willisonabout 2 hours ago
smevals: Small Eval Suite for Models and Prompts
2 min read
smevals is a new tool for running small eval suites across model configurations and grading results. It supports commands like uvx smevals run, grade, serve, and build. The tool was developed with Jesse Vincent's Prime Radiant lab and is Simon Willison's third iteration on evals.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
