AlgoTune
Code · 2025-07-19
AlgoTune evaluates models on optimizing algorithmic Python code for speed while preserving correctness. Scores are reported as speedup factors, so higher values indicate stronger optimization performance.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.2 | 2.0 |
| Gemini 3.1 Pro Preview | 2.0 |
| GPT-5.4 | 1.9 |
| Gemini 3 Pro Preview | 1.8 |
| Opus 4.5 | 1.8 |
| O4 Mini | 1.7 |
| GPT-5 | 1.7 |
| Sonnet 4.5 | 1.5 |
| GLM-4.5 | 1.5 |
| Gemini 2.5 Pro | 1.5 |
| Opus 4.6 | 1.5 |
| gpt-oss-120b | 1.4 |
| GPT-5 Mini | 1.4 |
| Opus 4.1 | 1.3 |
| Opus 4 | 1.3 |