GPT-5.6 Luna (Non-reasoning) — benchmark results
Provider: OpenAI. Released 2026-07-09. Access: API.
Unified ELO 1673 ± 29, rank #311 of 1744 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Natural Intelligence | 45.85 | NatInt Score | 89.3 |
| AA Omniscience - Software Engineering (SWE) - Go | 52 | Accuracy (%) | 88 |
| AA Omniscience - Software Engineering (SWE) - Python | 54 | Accuracy (%) | 87.7 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 48 | Accuracy (%) | 87 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 58.18 | Accuracy (%) | 86.6 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 54.44 | Accuracy (%) | 85.4 |
| AA Omniscience - Software Engineering (SWE) | 49.1 | Accuracy (%) | 85 |
| AA Omniscience - Software Engineering (SWE) - Julia | 36 | Accuracy (%) | 85 |
| AA Omniscience - Software Engineering (SWE) - HTML | 54 | Accuracy (%) | 84.1 |
| AA Omniscience - Software Engineering (SWE) - PHP | 48 | Accuracy (%) | 83.4 |
| AA Omniscience - Software Engineering (SWE) - Java | 33 | Accuracy (%) | 83.3 |
| AA Omniscience - Software Engineering (SWE) - C | 62 | Accuracy (%) | 81.2 |
Interactive version: aibenchmarks.dev/model?slug=gpt-5-6-luna-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-10.