GPT-5.5 (Low): benchmark results
GPT-5.5 evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2026-04-23. Access: API.
Unified ELO 1680 ± 1, rank #155 of 1761 rated models, from 38 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Natural Intelligence | 71.39 | NatInt Score | 98.2 |
| AA Terminal-Bench Hard | 52.27 | Accuracy (%) | 95.7 |
| AA Omniscience - Software Engineering (SWE) | 82.9 | Accuracy (%) | 95.2 |
| AA Omniscience - Business | 46.8 | Accuracy (%) | 95 |
| AA-Omniscience Accuracy | 54.85 | Accuracy (%) | 94.6 |
| AA Omniscience - Health | 46.8 | Accuracy (%) | 94 |
| AA Omniscience - Humanities & Social Sciences | 51.3 | Accuracy (%) | 93 |
| AA Long Context Reasoning | 81 | Accuracy (%) | 92.6 |
| AA Omniscience - Law | 51.1 | Accuracy (%) | 92.4 |
| AA Omniscience - Science, Engineering & Mathematics | 50.2 | Accuracy (%) | 92.4 |
| UGI - Writing | 58.69 | Writing Score | 92.2 |
| AA GPQA Diamond | 91.01 | Accuracy (%) | 91.3 |
Interactive version: theaggregate.ai/model?slug=gpt-5-5-low · How It Works · Data refreshed daily, snapshot 2026-09-05.