OpenAI
GPT-6 Astra
OpenAI's strongest public model until GPT-6.2 Astra (17 November 2026). Scores 169.2 on the Epoch Capabilities Index. The first OpenAI model rated Critical for cybersecurity under its Preparedness Framework.
- Announced
- Access
- Public
Benchmark scores
| Test | Score | Best on the tracker | How it was measured |
|---|---|---|---|
| Epoch Capabilities Indexmany tests combined into one score | 169 points | GPT-6.2 Astra 174 points | Independentlisted as "GPT-6 Astra", CI 166.19 to 172.51Run 9 Nov 2026 |
| AA Intelligence Index v5about ten tests in one score | 56.9 points | Claude Fable 5.2 64.6 points | Independentv5.0.1; listed as GPT-6 Astra (Max)Run 10 Nov 2026 |
| Remote Labor Indexpaid freelance projects | 21% | Claude Fable 5.2 41% | IndependentGPT 6 AstraRun 17 Sept 2026 |
| FrontierSWE V2very long engineering jobs | 66% | Claude Fable 5.2 70% | Independentproximus harness, max effort, mean@5 over 34 tasks, 20-hour budget (best@5 73.0) |
| Terminal-Bench 4.1long command-line tasks | 60% | Claude Opus 5.5 66% | IndependentCodex harness, max effort, 330 trials, ±2.82 (95% CI) |
| MirrorCoderebuilding whole programs | 47% | Claude Opus 5.5 79% | Independenthigh reasoning effort, MirrorCode (ML, +Private, 2L): 15 programs x 2 languages, 3 runs each, 10B-token / 7-day budget, ±11.9 SERun 30 Aug 2026 |
| APEX-Agentsprofessional office tasks | 65% | Gemini 4 Carbon 84% | Independentpass@1, unknown effort; listed as "GPT-6 Astra" |
| Humanity's Last Examexpert questions, no tools | 55% | Top score | Independentleaderboard entry "GPT 6 Astra" (gpt-6-astra_unknown) |
| ARC-AGI-3puzzle games, standard harnessTypical first-time human player: 100% | 63% | Claude Fable 5.2 81% | IndependentARC Prize config "GPT-6 Astra (Max)" (openai-gpt-6-astra-max), standard harness |
| FrontierMath T1–3hard maths problems | 94% | Claude Fable 5.2 95% | Independentmax reasoning effort (gpt-6-astra_max)Run 30 Aug 2026 |
| GPQA DiamondPhD-level science questionsPhD experts: 70%Saturated | 96% | Gemini 4 Argon 96% | Independentmax reasoning effort (gpt-6-astra_max)Run 30 Aug 2026 |
| FrontierMath T4research-level maths problemsSaturated | 98% | GPT-6.1 Sol 100% | Independenthigh reasoning effort (gpt-6-astra_high)Run 30 Aug 2026 |
| ARC-AGI-2visual puzzles, model aloneAverage person: 60%Saturated | 95% | Top score | IndependentARC Prize config "GPT-6 Astra (Max)" (openai-gpt-6-astra-max) |
| OTIS Mock AIMEcompetition mathsSaturated | 100% | Top score | Independentmax reasoning effort (gpt-6-astra_max)Run 30 Aug 2026 |
| METR time horizon (TH1.1 + TH2)task length an AI agent finishes half the time | 1.7 days | Top score | IndependentPooled TH1.1 + TH2 suites; 95% CI 791.5–158701.0 min |
Also on these charts
Sources
- GPT-6 Astra: A new generation of intelligenceOpenAI · Official · openai.com
- Safety overview: GPT-6 AstraOpenAI · Official · openai.com
- OpenAI API changelogOpenAI · Official · developers.openai.com
- OpenAI's GPT-6 Astra on ARC-AGI-3ARC Prize · Official · arcprize.org
- Epoch Capabilities IndexEpoch AI · Dataset · epoch.ai
- Epoch AI Benchmarking Hub (model release dates and runs)Epoch AI · Dataset · 9 Nov 2026 · epoch.ai