INTELLECT-3技术报告

摘要

我们推出INTELLECT-3——一个基于端到端强化学习基础设施栈训练、拥有1060亿参数（激活120亿）的混合专家模型。该模型在数学、编程、科学和推理等基准测试中，以同等规模实现了最先进的性能表现，超越了许多参数更大的前沿模型。我们将模型连同其完整构建基础设施开源发布，包括强化学习框架、完整训练方案，以及通过验证器库构建的、来自Environments Hub社区平台的丰富训练评估环境集合。为此我们全新开发了prime-rl框架，这是一个支持大规模异步强化学习的开源架构，可实现从单节点到数千张GPU的无缝扩展，并专门为智能体强化学习设计，原生支持多轮交互与工具调用功能。基于该技术栈，我们在GLM-4.5-Air-Base模型基础上同步进行了监督微调与强化学习训练，最终在512张H200显卡上实现了高训练效率的大规模强化学习训练。

English

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.

INTELLECT-3技术报告

INTELLECT-3: Technical Report

摘要

Support