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.
Aggregate Rating 1797 ± 11, rank #144 of 2168 rated models, from 143 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| IFM K2-Horizon-375B Card - MCPMark | 74 | Score (%) | 100 |
| IFM K2-Horizon-375B Card - WildClawBench (English Text-Only Subset) | 60 | Score (%) | 100 |
| PM-LLM-Benchmark | 38.4 | Score | 98.2 |
| AA Terminal-Bench Hard | 57.58 | Accuracy (%) | 97.3 |
| Korean CSAT 2026 (Easy Mode) - Physics I | 50 | Points (out of 50) | 95.9 |
| Aikido CVE Rediscovery (pass@3) | 88.5 | Recall (%) | 93.5 |
| Korean CSAT 2026 (Easy Mode) - Korean | 100 | Points (out of 100) | 92.6 |
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 64.44 | Omniscience Index | 91.1 |
| AA CritPt | 22.86 | Accuracy (%) | 91 |
| AA-Omniscience Index - Software Engineering (SWE) - Python | 60 | Omniscience Index | 90.4 |
| AA-Omniscience Index - Software Engineering (SWE) - JavaScript | 58.18 | Omniscience Index | 88.9 |
| WeirdML | 78.27 | Average Score | 88.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-terra-high · How It Works · Data refreshed daily, snapshot 2026-10-20.