GPT-5.6 Luna (High): benchmark results
GPT-5.6 Luna evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Aggregate Rating 1763 ± 10, rank #210 of 2168 rated models, from 143 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PM-LLM-Benchmark | 37.2 | Score | 93.4 |
| AA-Omniscience Index - Software Engineering (SWE) - PHP | 64 | Omniscience Index | 91.5 |
| Korean CSAT 2026 (Easy Mode) - Physics I | 48 | Points (out of 50) | 90.9 |
| AA-Omniscience Index - Software Engineering (SWE) - Swift | 60 | Omniscience Index | 90.3 |
| AA Long Context Reasoning | 80.33 | Accuracy (%) | 87.5 |
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 55.56 | Omniscience Index | 87.2 |
| AA GPQA Diamond | 89.19 | Accuracy (%) | 86.8 |
| o11y-bench - Pass^3 | 66.67 | Tasks passed on all three attempts, Pass^3 (%) | 86.8 |
| AA CritPt | 16.57 | Accuracy (%) | 85.3 |
| o11y-bench - Pass@3 | 87.3 | Tasks passed on at least one of three attempts, Pass@3 (%) | 85.1 |
| AA-Omniscience Index - Software Engineering (SWE) - Python | 46 | Omniscience Index | 84 |
| AA-Omniscience Index - Software Engineering (SWE) - JavaScript | 48.18 | Omniscience Index | 83.9 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-luna-high · How It Works · Data refreshed daily, snapshot 2026-10-20.