We present INTELLECT-3, a 106B-parameter Mixture-of-Experts model (12B active) trained with large-scale reinforcement learning on our end-to-end RL infrastructure stack. INTELLECT-3 achieves state of the art performance for its size across math, code, science and reasoning benchmarks, outperforming many larger frontier models. We open-source the model together with the full infrastructure stack used to create it, including RL frameworks, complete recipe, and a wide collection of environments, built with the verifiers library, for training and evaluation from our Environments Hub community platform. Built for this effort, we introduce prime-rl, an open framework for large-scale asynchronous reinforcement learning, which scales seamlessly from a single node to thousands of GPUs, and is tailored for agentic RL with first-class support for multi-turn interactions and tool use. Using this stack, we run both SFT and RL training on top of the GLM-4.5-Air-Base model, scaling RL training up to 512 H200s with high training efficiency.
@article{arxiv.2512.16144,
title = {INTELLECT-3: Technical Report},
author = {Prime Intellect Team and Mika Senghaas and Fares Obeid and Sami Jaghouar and William Brown and Jack Min Ong and Daniel Auras and Matej Sirovatka and Jannik Straube and Andrew Baker and Sebastian Müller and Justus Mattern and Manveer Basra and Aiman Ismail and Dominik Scherm and Cooper Miller and Ameen Patel and Simon Kirsten and Mario Sieg and Christian Reetz and Kemal Erdem and Vincent Weisser and Johannes Hagemann},
journal= {arXiv preprint arXiv:2512.16144},
year = {2025}
}