GPT-5 Nano (High): benchmark results
GPT-5 Nano evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.
Aggregate Rating 1559 ± 13, rank #834 of 2168 rated models, from 152 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA LiveCodeBench | 78.94 | Pass@1 (%) | 91.1 |
| SLR-Bench - Basic | 100 | Accuracy (%) | 86.4 |
| AA IFBench | 67.55 | Accuracy (%) | 81 |
| AA AIME 2025 | 83.67 | Accuracy (%) | 80.4 |
| ObviousBench | 97.22 | Answer pass³ (%) | 75.1 |
| AA-Omniscience Hallucination Rate | 59.01 | Hallucination Rate (%) | 72.5 |
| LLM Chess (Saplin) | 682.96 | Native anchored-opponent Elo (signed points) | 72.3 |
| AA Omniscience - Software Engineering (SWE) - Swift | 40 | Accuracy (%) | 70.1 |
| AA Omniscience - Software Engineering (SWE) - Dart | 22 | Accuracy (%) | 66.2 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 20 | Accuracy (%) | 63 |
| AA-Omniscience Index - Business | -17.8 | Omniscience Index | 59.9 |
| AA MMLU-Pro | 78.03 | Accuracy (%) | 59.2 |
Interactive version: theaggregate.ai/model?slug=gpt-5-nano-high · How It Works · Data refreshed daily, snapshot 2026-10-20.