pgasawa/continual-learning-bench

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Measuring and evolving with the frontier of agent work

EdgeBench: Unveiling scaling laws of learning from real-world environments

Infrastructure for continually self‑improving agents

Evaluate and improve models and agents using environments

Research infra for creating RL environments, post-training, and evals.

MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

General agent evaluation framework