CWM：一款面向代码生成与世界模型研究的开源权重大型语言模型

摘要

我们发布了Code World Model（CWM），一个拥有320亿参数的开源权重大型语言模型（LLM），旨在推动基于世界模型的代码生成研究。为了超越仅从静态代码训练中获得的代码理解能力，我们对CWM进行了中期训练，使用了大量来自Python解释器及自主Docker环境的观察-行动轨迹数据，并在可验证编码、数学及多轮软件工程环境中进行了广泛的多任务推理强化学习（RL）。通过CWM，我们为研究者提供了一个强大的测试平台，以探索世界模型在计算环境中通过推理与规划提升代码生成的潜力。我们初步展示了世界模型如何助力自主编码，实现Python代码执行的逐步模拟，并展示了推理如何从后者中获益的早期成果。CWM是一个密集的、仅解码器架构的LLM，训练上下文长度可达131k个标记。独立于其世界建模能力，CWM在通用编码与数学任务上表现出色：在SWE-bench Verified（含测试时扩展）上达到65.8%的pass@1分数，在LiveCodeBench上为68.6%，在Math-500上高达96.6%，在AIME 2024上则为76.0%。为支持代码世界建模的进一步研究，我们公开了中期训练、监督微调（SFT）及强化学习后的模型检查点。

English

We release Code World Model (CWM), a 32-billion-parameter open-weights LLM, to advance research on code generation with world models. To improve code understanding beyond what can be learned from training on static code alone, we mid-train CWM on a large amount of observation-action trajectories from Python interpreter and agentic Docker environments, and perform extensive multi-task reasoning RL in verifiable coding, math, and multi-turn software engineering environments. With CWM, we provide a strong testbed for researchers to explore the opportunities world modeling affords for improving code generation with reasoning and planning in computational environments. We present first steps of how world models can benefit agentic coding, enable step-by-step simulation of Python code execution, and show early results of how reasoning can benefit from the latter. CWM is a dense, decoder-only LLM trained with a context size of up to 131k tokens. Independent of its world modeling capabilities, CWM offers strong performance on general coding and math tasks: it reaches pass@1 scores of 65.8% on SWE-bench Verified (with test-time scaling), 68.6% on LiveCodeBench, 96.6% on Math-500, and 76.0% on AIME 2024. To support further research on code world modeling, we release model checkpoints after mid-training, SFT, and RL.

CWM：一款面向代码生成与世界模型研究的开源权重大型语言模型

CWM: An Open-Weights LLM for Research on Code Generation with World Models

摘要

Support