Latest Updates fromPrime Intellect.

verifiers v1: Decomposing Tasksets and Harnesses for Agentic RL & Evaluations

prime-rl gets an Algorithms layer

RL at 1T Scale: prime-rl Performance Deep Dive

True Agents Model the World

Systematic Reward Hacking and Prime Sprints

General Agent: A Self-Evolving, Synthetic Agent Environment

Autonomous AI research for nanogpt speedrun

renderers: Token-Level Templating for Agentic RL

Recursive Language Models: the paradigm of 2026