News

Prime Agent: A self-improving RLM agent

  • None--Primeintellect.ai
  • published date: 2026-08-05 21:11:57 UTC

Prime Agent is our open-source, self-improving coding harness built around two abstractions: the Recursive Language Model (RLM) and the Continual Harness. With Opus 5, it achieves 95.5% on ARC-AGI-3, surpassing the reported human expert baseline.

Today, we are launching Prime Agent, our self-improving coding harness designed around two abstractions, the Recursive Language Model (RLM) [citation] and Continual Harness [citation]. Modern harness… [+23155 chars]