GPT-5.6 Terra (High) — benchmark results

Provider: OpenAI. Released 2026-07-09. Access: API.

Unified ELO 1951 ± 40, rank #39 of 1744 rated models, from 39 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
PM-LLM-Benchmark38.4Score98.7
AA CritPt22.86Accuracy (%)97.4
AA Terminal-Bench Hard57.58Accuracy (%)97.3
AA Long Context Reasoning72.33Accuracy (%)96.2
AA Omniscience - Software Engineering (SWE) - TypeScript78.89Accuracy (%)96.2
Artificial Analysis Intelligence Index48.95Intelligence Index96.2
AA Omniscience - Software Engineering (SWE) - Python78Accuracy (%)96
AA Omniscience - Software Engineering (SWE) - JavaScript77.27Accuracy (%)95.8
AA Omniscience - Software Engineering (SWE) - Kotlin68Accuracy (%)95.7
AA Humanity's Last Exam36.7Accuracy (%)95.1
AA Omniscience - Health42.2Accuracy (%)95
AA Omniscience - Software Engineering (SWE) - Go68Accuracy (%)94.4

Interactive version: aibenchmarks.dev/model?slug=gpt-5-6-terra-high · How the rankings work · Data refreshed daily, snapshot 2026-07-10.