by oripress · Agent Tool · ★ 107
Website | Paper How good are language models at coming up with new algorithms? To try to answer this, we built a benchmark, AlgoTune, comprised of 154 widely used math, physics, and computer science functions. For each function, the goal is to write code that produces the same outputs as the original function, while being faster. In addition to the benchmark, we also provide an agent, AlgoTuner, which allows language models to easily optimize code. ✨ New: AlgoTune can now be easily run on AWS with just an OpenRouter API key and AWS credentials.
| Stars | 107 |
| Forks | 13 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 69.238284222667/100 |
| Open Issues | 2 |
| Last Updated | 2026-06-24 |
| Created | 2025-07-02 |
| Platforms | python |
| Est. Tokens | ~15k |
These tools work well together with AlgoTune for enhanced workflows:
Explore other popular agent tool tools:
AlgoTune is AlgoTune is a NeurIPS 2025 benchmark made up of 154 math, physics, and computer science problems. The goal is write code that solves each problem, and is faster than existing implementations.. It is categorized as a Agent Tool with 107 GitHub stars.
AlgoTune is primarily written in Python. It covers topics such as code-agent, code-generation, code-optimization.
You can find installation instructions and usage details in the AlgoTune GitHub repository at github.com/oripress/AlgoTune. The project has 107 stars and 13 forks, indicating an active community.
AlgoTune is released under the MIT license, making it free to use and modify according to the license terms.