GPT-5.5 (Non-reasoning) — benchmark results
Provider: OpenAI. Released 2026-04-23. Access: API.
Unified ELO 1845 ± 21, rank #110 of 1839 rated models, from 43 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) - TypeScript | 80 | Accuracy (%) | 96.7 |
| UGI - Natural Intelligence | 62.57 | NatInt Score | 95.5 |
| AA Omniscience - Software Engineering (SWE) - Julia | 68 | Accuracy (%) | 95 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 68 | Accuracy (%) | 94.8 |
| AA Omniscience - Software Engineering (SWE) - PHP | 76 | Accuracy (%) | 94.6 |
| AA Terminal-Bench Hard | 49.24 | Accuracy (%) | 94.5 |
| AA Omniscience - Software Engineering (SWE) - C | 83 | Accuracy (%) | 94.2 |
| AA Omniscience - Software Engineering (SWE) - Python | 75 | Accuracy (%) | 93.9 |
| AA Omniscience - Software Engineering (SWE) | 69.5 | Accuracy (%) | 93.5 |
| AA Omniscience - Software Engineering (SWE) - R | 60 | Accuracy (%) | 93.5 |
| AA Omniscience - Business | 38.8 | Accuracy (%) | 93.2 |
| AA Omniscience - Law | 38.7 | Accuracy (%) | 92.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-5-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-08-05.