Problem Hunting with LLMs
This website documents attempts by advanced Large Language Models (LLMs) to solve open mathematical problems. The idea for this project is due to mathematician Paata Ivanisvili.
Most of the current attempts are by GPT Pro 5.2.
Problem Sources
Erdos Problems
For Erdos problems, we track problems from two key resources:
- Terry Tao's Erdos Problems Database - A collaborative database of Erdos problems with formal statements and progress tracking.
-
erdosproblems.com (by Thomas Bloom) -
For detailed problem statements, historical context, and relevant comments.
For each Erdos problem number X, the full problem statement can be found at
https://www.erdosproblems.com/X.
MathOverflow Problems
MathOverflow problems are sourced from MathOverflow, a question and answer site for professional mathematicians. Each problem links to its original MathOverflow discussion.
Acknowledgments
- Paata Ivanisvili - For the idea
- Terry Tao - For the Erdos Problems database
- Thomas Bloom - For the erdosproblems.com website
Disclaimer
Important: LLM output is not fully reliable. The attempts documented on this website represent exploratory work by frontier AI models and should not be considered verified mathematical proofs.
While only the most advanced LLMs with a proven track record of making progress on open mathematical problems are featured here (such as GPT Pro, GPT 5.2, Gemini Deep Think, and Opus 4.5), their claimed solutions are not final and require rigorous verification by human mathematicians.
Any "solved" status indicated on this website means that the LLM produced output claiming to solve the problem - it does not mean the solution has been verified or accepted by the mathematical community.
Contributing
Contributions are welcome. Please see our Contributing Guidelines for details on how to submit new LLM attempts or improvements.
Only contributions featuring the most advanced frontier LLMs are accepted:
- GPT Pro
- GPT 5.2
- Gemini Deep Think
- Opus 4.5
- Other comparable frontier models with demonstrated mathematical reasoning capabilities
License
This project is licensed under the Apache License 2.0.