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 1944 ± 32, rank #42 of 1776 rated models, from 46 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PM-LLM-Benchmark | 38.4 | Score | 98.1 |
| AA Terminal-Bench Hard | 57.58 | Accuracy (%) | 97.3 |
| AA CritPt | 22.86 | Accuracy (%) | 97.2 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 78.89 | Accuracy (%) | 96.3 |
| AA Long Context Reasoning | 72.33 | Accuracy (%) | 96.1 |
| AA Omniscience - Software Engineering (SWE) - Python | 78 | Accuracy (%) | 96.1 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 68 | Accuracy (%) | 95.8 |
| Artificial Analysis Intelligence Index | 48.95 | Intelligence Index | 95.8 |
| WeirdML | 78.27 | Average Score | 95.6 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 77.27 | Accuracy (%) | 95.5 |
| AA Omniscience - Health | 42.2 | Accuracy (%) | 95.1 |
| AA Omniscience - Software Engineering (SWE) - Go | 68 | Accuracy (%) | 94.5 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-terra-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.