OpenAI · model
GPT-5 mini benchmark scores
As of 2026-11-22, GPT-5 mini (OpenAI) has measured scores on 1 benchmark and estimated scores on 11 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-SciCode | 47.6% | estimated ± 3.8 pp, high confidence |
| AA Coding Index | 46.8% | estimated ± 6.9 pp, medium confidence |
| SWE-bench Verified | 73.9% | estimated ± 4.5 pp, high confidence |
| SWE-bench Pro | 54.1% | estimated ± 2.9 pp, medium confidence |
| Vals SWE-bench | 70.8% | estimated ± 7.1 pp, medium confidence |
| PostTrainBench v1.1 | 19.9% | estimated ± 10.5 pp, low confidence |
| Vibe Code Bench | 14.2% | measured |
| PostTrainBench v1 | 19.0% | estimated ± 3.1 pp, low confidence |
| AA LiveCodeBench | 85.7% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 61.7% | estimated ± 3.0 pp, low confidence |
| SWE-Rebench | 7.3% | estimated ± 6.5 pp, low confidence |
| PostTrainBench | 17.0% | estimated ± 8.3 pp, low confidence |