GPT-5.6 Luna (xHigh): benchmark results
GPT-5.6 Luna evaluated at the xhigh reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Aggregate Rating 1788 ± 10, rank #162 of 2168 rated models, from 193 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenRouter WideSearch (Search) | 22 | Best-configuration Exact Table Match (%) | 100 |
| AA Long Context Reasoning | 81.67 | Accuracy (%) | 91.1 |
| AA GDP.pdf - Manufacturing/Supply Chains - Criterion Pass Rate | 77.15 | Criterion Pass Rate (%) | 90.2 |
| AA CritPt | 20.57 | Accuracy (%) | 89.8 |
| Chatbot Arena (Text - Mathematical) | 1485 | Arena Score | 89.4 |
| Chatbot Arena (Text - German) | 1476 | Arena Score | 89.1 |
| Chatbot Arena (Text - Math) | 1477 | Arena Score | 89 |
| RuBench (Fresh-11) | 75.8 | Pass@1 (%; the 11 tasks with fix commits after 2026-06-17, e | 88.9 |
| AA GPQA Diamond | 89.49 | Accuracy (%) | 87.6 |
| LLM Chess (Saplin) | 1185.3 | Native anchored-opponent Elo (signed points) | 87.6 |
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 55.56 | Omniscience Index | 87.2 |
| Artificial Analysis Intelligence Index | 34.56 | Intelligence Index | 87 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-luna-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-20.