GPT-6 Astra Benchmark Data
A reusable, source-labeled data table compiled from OpenAI's launch materials. Filter it on-page or download the CSV.
Data status:
Vendor-reported, not independently reproduced. Last source verification: September 5, 2026.
Interactive benchmark table
| Benchmark | Category | GPT-6 Astra | GPT-5.6 Sol | Claude Opus 5 |
|---|---|---|---|---|
| Agents' Last Exam | Computer use | 59.3% | 53.6% | 55.5% |
| OSWorld 2.0 | Computer use | 72.6% | 65.7% | 70.2% |
| ScreenSpot-Pro | Computer use | 92.7% | 76.9% | — |
| AutomationBench | Professional | 41.4% | 18.1% | 26.9% |
| BenchCAD | Professional | 95.9% | 83.3% | 82.1% |
| BrowseComp | Professional | 91.5% | 90.4% | 90.8% |
| MRCR v2 256K–512K | Long context | 100.0% | 91.5% | — |
| MRCR v2 512K–1M | Long context | 96.3% | 73.8% | — |
| ARC-AGI-3 | Reasoning | 99.9% | 7.8% | 30.2% |
| ARC-AGI-2 | Reasoning | 95.0% | 92.5% | 90.4% |
| ARC-AGI-1 | Reasoning | 98.5% | 97.5% | 97.5% |
| ExploitBench | Cybersecurity | 100.0% | 78.5% | — |
| ExploitGym | Cybersecurity | 42.4% | 30.3% | — |
Vendor-reported figures from OpenAI's GPT-6 Astra launch materials. “—” means the launch table did not publish a directly comparable value in that row.
How to cite this dataset
When reusing these numbers, attribute the underlying results to OpenAI and describe this page as a compiled reference table. Do not describe the figures as independent tests conducted by GPT-6 Astra Guide.
Primary sources
Specifications and prices can change. We prefer first-party documentation and record the date each page was reviewed.