GPT-5.6 Terra — benchmark results
OpenAI's balanced GPT-5.6 tier for everyday reasoning and coding work. Provider: OpenAI. Released 2026-07-09. Access: API.
Unified ELO 1879 ± 16, rank #79 of 1776 rated models, from 104 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| WolfBench | 82 | Average Score (%) | 96.8 |
| Vals AI TaxEval v2 | 76.17 | Accuracy (%) | 96.5 |
| ZeroEval GPQA Diamond | 92.9 | GPQA Diamond Score | 96 |
| LLM Stats (MRCR v2 (8-needle)) | 89.6 | Score (%) | 95 |
| BenchLM | 72.6 | Overall Score | 94.9 |
| LLM Stats (BrowseComp) | 87.5 | Score (%) | 94.6 |
| Gert Labs Rankings | 67.02 | GScore (%) | 94.2 |
| Benchmarks.bio - SpatialBench-Long | 22.22 | Pass Rate (%) | 94.1 |
| Vals AI ProofBench | 71 | Accuracy (%) | 93.8 |
| Epoch AI - ECI | 158.36 | ECI Score | 93.4 |
| RuneBench | 5934 | Total Peak XP Rate (XP/min) | 92.1 |
| EQ-Bench Longform Writing | 78 | Writing Score (0-100) | 91.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-terra · How the rankings work · Data refreshed daily, snapshot 2026-07-22.