GPT-5 Nano: benchmark results
OpenAI's smallest, cheapest GPT-5 tier with a 400K context, built for high-volume classification, extraction and autocomplete (August 2025). Provider: OpenAI. Released 2025-08-07. Access: API.
Unified ELO 1567 ± 1, rank #376 of 1392 rated models, from 847 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Safety | 98.6 | Mean score (self-reported) | 100 |
| PCB-Bench - QA Placement SBERT | 67.86 | SBERT similarity (%) | 100 |
| PCB-Bench - QA Routing SBERT | 63.47 | SBERT similarity (%) | 100 |
| StereoTales | 67 | Emissions (self-reported) | 100 |
| OpenVLM MathVista - GPS | 93.8 | Accuracy (%) | 99.7 |
| OpenVLM MMMU - Accounting | 86.7 | Accuracy (%) | 99.6 |
| OpenVLM MMMU - Math | 76.7 | Accuracy (%) | 99.5 |
| OpenVLM MathVista - SCI | 80.3 | Accuracy (%) | 99.3 |
| OpenVLM MathVista - ALG | 89.7 | Accuracy (%) | 99 |
| OpenVLM MathVista - GEO | 90.4 | Accuracy (%) | 99 |
| OpenVLM MMMU - Mechanical Engineering | 63.3 | Accuracy (%) | 98.9 |
| OpenVLM MMMU - Chemistry | 76.7 | Accuracy (%) | 98.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-nano · How It Works · Data refreshed daily, snapshot 2026-09-05.