Bloom turns a single named behavior into many generated scenarios and scores how often it appears, so you can build evalA repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.Full definition → in days. Its outputs correlate with hand labels. The original repo is frozen and development moved to Meridian Labs.
Terms in this piece · Glossary
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
articleREAP: Automatic Curation of Coding Agent Benchmarks from Interactive Production UsageSmriti Jha, Matteo Paltenghi, Chandra Maddila, Vijayaraghavan Murali, Shubham Ugare, Satish Chandra
articleDemystifying evals for AI agentsAnthropic Engineering
videoEverything Is a Rollout — Alex Shaw + Ryan Marten, Terminal-Bench, Harbor, Laude InstituteAI Engineer