miromind-ai/MiroThinker-14B-SFT-v0.2

Introduction

MiroThinker is an open-source agentic model series. Designed as a research agent for complex, long-horizon problem solving, it integrates strong capabilities in task decomposition, multi-hop reasoning, retrieval-augmented generation, code execution, web browsing, and document/file processing, enabling a wide range of real-world applications.

In MiroThinker-v0.2, we introduced three key improvements:

• Richer training data from both English and Chinese sources, yielding significant gains in benchmark performance and generalization.
• Unified DPO training with a single preference dataset across all models.
• Extended context length from 40k to 64k for more challenging multi-turn tool-use tasks.

Compared to v0.1, MiroThinker-v0.2 delivers consistent gains across benchmarks. For example, scores improved from 57.3 → 64.1 on GAIA-Text-103 and from 17.0 → 29.4 on BrowseComp-ZH, reflecting substantial advancements in the model’s general research agent capabilities.

Online Demo

Welcome to try out our online demo here.

Performance

Comparison with SOTA Research Agents

GAIA Benchmark

Quick Start

MiroThinker-v0.2 is trained on our large-scale, high-quality trajectory and preference datasets MiroVerse-v0.2, utilizing the efficient training framework MiroTrain, and enhanced with tool-use capabilities through our agentic framework MiroFlow.

To promote reproducibility and benefit the community, we decided to open-source the entire suite mentioned above. For more technical details, evaluation results, and usage tutorials, please visit our GitHub repository.

License

MiroThinker-v0.2 is licensed under Apache 2.0.

Contact Us

MiroThinker is developed by the MiroMind Foundation Model Team. If you would like to leave us a message, feel free to get in touch. In addition to GitHub, Discord, WeChat, and RedNote, you can also reach us via email at [email protected].