GPT-5 Nano (Minimal): benchmark results
GPT-5 Nano evaluated at the minimal reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.
Aggregate Rating 1449 ± 16, rank #1253 of 2168 rated models, from 78 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PRISM-VLM - Instruction Following | 83.7 | Per-axis mean score (%; mean compliance with 1-3 verifiable | 90.2 |
| AI for Education Visual Maths - Statistics and Probability | 28.57 | Accuracy (%) | 60.8 |
| PRISM-VLM - Sycophancy Resistance | 46.6 | Per-axis mean score (%; share of turn-0-correct answers held | 56.1 |
| AA LiveCodeBench | 46.98 | Pass@1 (%) | 53.8 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 16 | Accuracy (%) | 47.8 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 25.45 | Accuracy (%) | 42 |
| PRISM-VLM - Counting | 49.3 | Per-axis mean score (%; exact-integer accuracy on synthesize | 40.2 |
| AA Terminal-Bench Hard | 6.82 | Accuracy (%) | 37.1 |
| AA Omniscience - Software Engineering (SWE) - Dart | 14 | Accuracy (%) | 36.2 |
| AA Omniscience - Software Engineering (SWE) - PHP | 16 | Accuracy (%) | 35 |
| Tau2-Bench Telecom | 25.73 | Success Rate (%) | 30.7 |
| AA Omniscience - Software Engineering (SWE) - Julia | 4 | Accuracy (%) | 30.4 |
Interactive version: theaggregate.ai/model?slug=gpt-5-nano-minimal · How It Works · Data refreshed daily, snapshot 2026-10-20.