GPT-5.6 (Low) — benchmark results
Provider: OpenAI. Access: API.
Unified ELO 1965 ± 30, rank #31 of 1723 rated models, from 38 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) - JavaScript | 89.09 | Accuracy (%) | 99.6 |
| AA Omniscience - Health | 50.3 | Accuracy (%) | 99.5 |
| AA Omniscience - Software Engineering (SWE) - Go | 84 | Accuracy (%) | 99.2 |
| AA Omniscience - Software Engineering (SWE) - Rust | 86 | Accuracy (%) | 99.2 |
| AA Omniscience - Software Engineering (SWE) - C | 88 | Accuracy (%) | 99.1 |
| AA Omniscience - Software Engineering (SWE) - PHP | 92 | Accuracy (%) | 99.1 |
| AA Omniscience - Business | 49.1 | Accuracy (%) | 98.9 |
| AA Terminal-Bench Hard | 60.61 | Accuracy (%) | 98.7 |
| AA-Omniscience Accuracy | 56.43 | Accuracy (%) | 98.6 |
| AA Omniscience - Humanities & Social Sciences | 54.5 | Accuracy (%) | 98.1 |
| AA SciCode | 55.44 | Accuracy (%) | 98.1 |
| AA Omniscience - Software Engineering (SWE) - HTML | 84 | Accuracy (%) | 98 |
Interactive version: aibenchmarks.dev/model?slug=gpt-5-6-low · How the rankings work · Data refreshed daily, snapshot 2026-07-10.