news
266

Prime Agent: A self-improving RLM agent
2026-08-06Prime Agent is an open-source, self-improving coding harness built on Recursive Language Model and Continual Harness abstractions. Powered by Opus 5, the system achieved a 95.5% score on the ARC-AGI-3 benchmark, successfully surpassing the previously reported human expert baseline.
Comments
0Discuss on Hacker NewsOnly full members can comment
Log in to express your opinion.
No comments yet.