GPT-6 Astra
ModelActiveby OpenAI · family “gpt-astra”
GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon... (description from the OpenRouter listing)
in: filein: imagein: textout: textReasoningTool useStructured output
Epoch Capabilities Index
166.6 #1 of 268
90% CI 163.0 – 172.0
Released
3 Sep 2026
Context window
1.05M tokens
Max output
128K tokens
Input $ / 1M tokens
$10.0
Output $ / 1M tokens
$50.0
Cached input $ / 1M
$1.00
Each benchmark shown separately with its own source
Benchmark results (19)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| Furniture Assembly | multimodal | 80.0% ±4.9 | 96% | max | 10 Sep 2026 | Epoch ↗ |
| FrontierSWE | other | 65.5% | 100% | max | — | External ↗ |
| LMCA | agents | 63.8% | 94% | high | — | External ↗ |
| DTBench | reasoning | 96.5% | 98% | high | — | External ↗ |
| Mystery Game Puzzles | games | 84.0% ±3.7 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| EBR-bench | reasoning | 76.2% ±8.4 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| MirrorCode | coding | 46.7% ±11.9 | 60% | high | 30 Aug 2026 | Epoch ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 93.7% ±1.4 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| FrontierMath-Tier-4-v2-Private | math | 97.6% ±2.4 | 100% | high | 30 Aug 2026 | Epoch ↗ |
| FrontierCode | coding | 53.3% | 100% | max | — | External ↗ |
| DeepSWE | coding | 73.2% | 99% | high | — | External ↗ |
| Chess Puzzles | games | 72.0% ±4.5 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| SimpleQA Verified | knowledge | 75.6% ±1.4 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| ARC-AGI-2 | reasoning | 95.0% | 100% | max | — | External ↗ |
| WeirdML | coding | 93.3% ±0.0 | 100% | max | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 100.0% | 100% | max | 30 Aug 2026 | Epoch ↗ |
| Balrog | games | 68.3% | 100% | max | — | External ↗ |
| GPQA diamond | science | 95.8% ±1.4 | 100% | max | 30 Aug 2026 | Epoch ↗ |
| ARC-AGI | reasoning | 98.5% | 100% | high | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
Unchanged since first recorded on 27 Sep 2026 (OpenRouter listing); re-read on every data refresh, most recently 39m ago. Steps show when the price changed.
Events
- BenchmarkEpoch AI evaluates GPT-6 AstraEpoch AI Benchmarking Hub
- Model launchMajorOpenAI releases GPT-6 AstraEpoch AI Benchmarking Hub
- BenchmarkEpoch AI evaluates GPT-6 AstraEpoch AI Benchmarking Hub