GPT-5.6 Terra (High): benchmark results
GPT-5.6 Terra evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Unified ELO 1698 ± 1, rank #114 of 1761 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PM-LLM-Benchmark | 38.4 | Score | 98.2 |
| AA Terminal-Bench Hard | 57.58 | Accuracy (%) | 97.3 |
| AA CritPt | 22.86 | Accuracy (%) | 93.9 |
| LLM Chess (Saplin) | 1210.7 | ELO | 93.8 |
| Aikido CVE Rediscovery (pass@3) | 88.5 | Recall (%) | 93.5 |
| AA Omniscience - Health | 44 | Accuracy (%) | 92 |
| WeirdML | 78.27 | Average Score | 91.1 |
| Artificial Analysis Intelligence Index | 41.3 | Intelligence Index | 90.9 |
| AA Omniscience - Software Engineering (SWE) | 69.7 | Accuracy (%) | 90.8 |
| AA Humanity's Last Exam | 38.51 | Accuracy (%) | 89.7 |
| AA Omniscience - Science, Engineering & Mathematics | 45.6 | Accuracy (%) | 88.5 |
| AA GPQA Diamond | 89.6 | Accuracy (%) | 88.2 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-terra-high · How It Works · Data refreshed daily, snapshot 2026-09-05.