GPT-5.6 Luna (Medium): benchmark results
GPT-5.6 Luna evaluated at the medium reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Unified ELO 1659 ± 1, rank #214 of 1761 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) | 65.1 | Accuracy (%) | 87.3 |
| AA Omniscience - Science, Engineering & Mathematics | 44.6 | Accuracy (%) | 85.8 |
| AA-Omniscience Accuracy | 40.7 | Accuracy (%) | 83.8 |
| AA Omniscience - Humanities & Social Sciences | 37.3 | Accuracy (%) | 82.5 |
| AA Omniscience - Business | 32.1 | Accuracy (%) | 81.9 |
| AA Omniscience - Health | 36.2 | Accuracy (%) | 81.9 |
| Wolfram LLM Benchmarking Project | 56 | Correct Functionality (%) | 81.6 |
| Artificial Analysis Intelligence Index | 30.19 | Intelligence Index | 81.2 |
| AA GPQA Diamond | 85.86 | Accuracy (%) | 80.1 |
| AA Omniscience - Law | 28.9 | Accuracy (%) | 79.9 |
| LisanBench | 0.14 | Mean Path Length / Current Maximum | 79.7 |
| LLM Chess (Saplin) | 730.1 | ELO | 79.5 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-luna-medium · How It Works · Data refreshed daily, snapshot 2026-09-05.