GPT-6 Astra (Medium): benchmark results
Provider: OpenAI. Released 2026-09-03. Access: API.
Unified ELO 1960 ± 11, rank #20 of 2072 rated models, from 101 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 97.78 | Omniscience Index | 100 |
| Computer Anthology Terminal Tasks (Terminus-2) | 67 | pass@1 (%) | 100 |
| DrivingBench | 100 | Course progress (%, best of up to 3 attempts in one chat) | 100 |
| DrivingBench (First Attempt) | 48.82 | Course progress (%, first attempt) | 100 |
| Maze-Bench | 94.99 | Mean Score | 100 |
| Taiwan Exams - TVE Joint Entrance Exam (ROC 115) | 98.41 | Points Scored (%) | 100 |
| AA-Omniscience Index - Software Engineering (SWE) - Go | 90 | Omniscience Index | 99.5 |
| AA-Omniscience Index - Business | 33.2 | Omniscience Index | 99.1 |
| AA GDP.pdf - Healthcare - Criterion Pass Rate | 92.51 | Criterion Pass Rate (%) | 99 |
| AA-Omniscience Index - Humanities & Social Sciences | 40.7 | Omniscience Index | 98.9 |
| AA-Omniscience Index - Software Engineering (SWE) | 83 | Omniscience Index | 98.8 |
| AA-Omniscience Index - Software Engineering (SWE) - HTML | 88 | Omniscience Index | 98.8 |
Interactive version: theaggregate.ai/model?slug=gpt-6-astra-medium · How It Works · Data refreshed daily, snapshot 2026-10-02.