miromind-ai/MiroThinker-32B-DPO-v0.1

Introduction

MiroThinker is an open-source agentic model series built on top of Qwen3. Designed for deep research and 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, making it suitable for a wide range of real-world applications.

We have released the MiroThinker-v0.1 series, including both SFT and DPO variants at parameter scales of 8B, 14B, and 32B. Notably, MiroThinker v0.1 achieves state-of-the-art performance among open-source models on the GAIA benchmark, a rigorous evaluation suite for advanced agentic capabilities, demonstrating its strength in long-context, decision-intensive, and real-world task scenarios.

Online Demo

Welcome to try out our online demo here. In this demo, we have deployed our MiroThinker-32B-DPO-v0.1 along with commercial tools (you can find more details in our GitHub), aiming to deliver a better experience.

Performance

GAIA Benchmark

Method	Text-103 Best Pass@1	Text-103 Pass@1 (Avg@8)	Val-165 Best Pass@1	Val-165 Pass@1 (Avg@8)
Search-o1-7B	17.5	-	-	-
R1-Searcher-7B	20.4	-	-	-
WebDancer-7B	31.0	-	-	-
WebSailor-7B	37.9	-	-	-
CK-Pro-8B	40.3	-	32.7	-
MiroThinker-8B-SFT-v0.1	44.7	40.1	34.6	31.8
+ Commercial Tools	46.6	42.1	37.6	33.9
MiroThinker-8B-DPO-v0.1	46.6	44.8	37.0	35.4
+ Commercial Tools	50.5	46.7	38.2	35.9
Search-o1-32B	28.2	-	-	-
WebThinker-32B-RL	48.5	-	-	-
WebDancer-QwQ-32B	51.5	-	-	-
WebSailor-32B	53.2	-	-	-
WebShaper-QwQ-32B	53.3	-	-	-
WebShaper-72B	60.1	-	-	-
MiroThinker-14B-SFT-v0.1	47.6	44.4	37.0	34.4
+ Commercial Tools	49.5	47.5	41.8	39.8
MiroThinker-14B-DPO-v0.1	48.5	46.6	42.4	39.2
+ Commercial Tools	52.4	48.5	45.5	42.0
MiroThinker-32B-SFT-v0.1	55.3	51.3	44.9	42.7
+ Commercial Tools	58.3	54.2	48.5	45.8
MiroThinker-32B-DPO-v0.1	57.3	54.1	48.5	45.9
+ Commercial Tools	60.2	57.9	50.9	48.9

1. Following the practices of WebThinker, WebAgents, and CognitiveKernel, we report the Best Pass@1, the highest score across three runs, which often reflects stronger performance, though it may exhibit some variability. To provide a more stable measure, we additionally report Pass@1 (Avg@8), which offers greater consistency at the cost of slightly lower scores.

2. For consistency with prior open-source works, we evaluate GAIA-Text-103 using the WebAgents LLM-as-judge template, and report results on GAIA-Val-165 using the official GAIA scorer script.

3. By default, we use open-source tools wherever possible, except for the code tool E2B and the Google search tool Serper. We use Whisper, Qwen2.5-VL-72B-Instruct, and Qwen3-235B-A22B-Thinking-2507 in our implementation. The framework can be easily extended to other open-source tools of your choice.

4. Commercial tools were mainly used for multimodal capabilities and certain complex reasoning subtasks. The majority of tasks, including planning, browsing, refinement, navigation, and more, were handled by our models.

More Benchmarks

Coming soon

Quick Start

MiroThinker-v0.1 is trained on our large-scale, high-quality trajectory and preference datasets MiroVerse-v0.1, 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-v0.1 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].