GPT-6.1 Terra (Low): benchmark results

Provider: OpenAI. Released 2026-10-26. Access: API.

Aggregate Rating 1836 ± 20, rank #94 of 2169 rated models, from 66 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA GDP.pdf - Engineering - Criterion Pass Rate83.87Criterion Pass Rate (%)99.6
AA-Omniscience Index - Software Engineering (SWE) - Go78Omniscience Index95.7
AA-Omniscience Index - Software Engineering (SWE) - JavaScript77.27Omniscience Index95.5
AA-Omniscience Index - Software Engineering (SWE) - PHP76Omniscience Index94.9
AA-Omniscience Index - Software Engineering (SWE) - Python78Omniscience Index94.5
AA-Omniscience Index - Software Engineering (SWE)68.3Omniscience Index93.5
AA-Omniscience Index - Software Engineering (SWE) - Kotlin70Omniscience Index93.4
AA-Omniscience Index - Software Engineering (SWE) - C80Omniscience Index93.2
AA-Omniscience Index - Software Engineering (SWE) - Julia60Omniscience Index93.2
AA-Omniscience Index - Software Engineering (SWE) - Java38Omniscience Index92.5
AA-Omniscience Index - Software Engineering (SWE) - TypeScript75.56Omniscience Index92.5
AA GDP.pdf - Real Estate - Criterion Pass Rate74.78Criterion Pass Rate (%)92.3

Interactive version: theaggregate.ai/model?slug=gpt-6-1-terra-low · How It Works · Data refreshed daily, snapshot 2026-10-27.