GPT-5 Nano (Medium): benchmark results
GPT-5 Nano evaluated at the medium reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.
Aggregate Rating 1554 ± 16, rank #857 of 2168 rated models, from 96 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA LiveCodeBench | 76.3 | Pass@1 (%) | 87 |
| IKP - T1 | 100 | Accuracy (%) | 86.1 |
| AA IFBench | 65.92 | Accuracy (%) | 77.4 |
| AA-Omniscience Hallucination Rate | 52.62 | Hallucination Rate (%) | 76.7 |
| AA AIME 2025 | 78.33 | Accuracy (%) | 72.9 |
| IKP - T2 | 99.5 | Accuracy (%) | 72.2 |
| LisanBench | 0.08 | Mean Path Length / Current Maximum | 71.9 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 33.64 | Accuracy (%) | 69.1 |
| LLM Chess (Saplin) | 413.3 | Native anchored-opponent Elo (signed points) | 65.5 |
| AI for Education Visual Maths - Algebra | 84.62 | Accuracy (%) | 64.9 |
| AA-Omniscience Index - Business | -14.4 | Omniscience Index | 63.6 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 20 | Accuracy (%) | 63 |
Interactive version: theaggregate.ai/model?slug=gpt-5-nano-medium · How It Works · Data refreshed daily, snapshot 2026-10-20.