{"ID":2827372,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.16144","arxiv_id":"2512.16144","title":"INTELLECT-3: Technical Report","abstract":"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.","short_abstract":"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...","url_abs":"https://arxiv.org/abs/2512.16144","url_pdf":"https://arxiv.org/pdf/2512.16144v1","authors":"[\"Prime Intellect Team\",\"Mika Senghaas\",\"Fares Obeid\",\"Sami Jaghouar\",\"William Brown\",\"Jack Min Ong\",\"Daniel Auras\",\"Matej Sirovatka\",\"Jannik Straube\",\"Andrew Baker\",\"Sebastian Müller\",\"Justus Mattern\",\"Manveer Basra\",\"Aiman Ismail\",\"Dominik Scherm\",\"Cooper Miller\",\"Ameen Patel\",\"Simon Kirsten\",\"Mario Sieg\",\"Christian Reetz\",\"Kemal Erdem\",\"Vincent Weisser\",\"Johannes Hagemann\"]","published":"2025-12-18T03:57:01Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}