MiroThinker
Open-source agent research, data, and evaluation for research agent, agent, agent framework.
What does this agent do, and when should you use it?
The repository describes MiroThinker as: MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively. This profile is a source-based catalog entry; an independent FARS review is still pending.
MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively.
- Run a documented research, analysis, or evaluation workflow.
- Compare agent behavior with reproducible evidence.
- Adapt its datasets, environments, or analysis components.
What are this agent's strengths and limitations?
- Public source and README are available for inspection.
- Focused on research agent, agent, agent framework.
- Setup, model-provider support, and maturity must be confirmed against the current release.
- No independent FARS score has been assigned yet.
How do you install or deploy this agent?
Follow the current installation instructions in the [repository README](https://github.com/MiroMindAI/MiroThinker#readme). Requirements and provider setup vary by release.
How do you use this agent?
Start with the examples and quickstart in the [repository documentation](https://github.com/MiroMindAI/MiroThinker#readme), then test the workflow with limited permissions and non-sensitive data.