OpenAI GPT-6.1 Terra Launch - Terminal-Bench 4.0: leaderboard
Official OpenAI October 26, 2026 GPT-6.1 Terra launch effort sweep; Score (%). Explicit version/effort/model settings are preserved in the board and row names. Shared Astra-card safety/health rows retain the unchanged GPT-6.1 Sol anchor only; changed older comparison checkpoints are not overwritten. Metric: Score (%).
Metric: Score (%). Source: openai.com. Saturation forecast: Around September 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 5.5 (Max) | 70.6 | #13 (Claude Sonnet 5.5) |
| 2 | Claude Sonnet 5.5 (xHigh) | 61.5 | #13 (Claude Sonnet 5.5) |
| 3 | GPT-6 Astra (Max) | 58.2 | #5 (GPT-6 Astra) |
| 4 | GPT-6 Astra (High) | 58.1 | #5 (GPT-6 Astra) |
| 5 | GPT-6 Astra (xHigh) | 57.9 | #5 (GPT-6 Astra) |
| 6 | GPT-6.1 Sol (Max) | 55.9 | #6 (GPT-6.1 Sol) |
| 7 | GPT-6 Astra (Medium) | 54.4 | #5 (GPT-6 Astra) |
| 8 | GPT-6.1 Sol (xHigh) | 53.4 | #6 (GPT-6.1 Sol) |
| 9 | GPT-6.1 Sol (High) | 53 | #6 (GPT-6.1 Sol) |
| 10 | GPT-6.1 Sol (Medium) | 52.7 | #6 (GPT-6.1 Sol) |
| 11 | GPT-6 Astra (Low) | 50.6 | #5 (GPT-6 Astra) |
| 12 | GPT-6.1 Terra (Max) | 50.3 | #10 (GPT-6.1 Terra) |
| 13 | GPT-6.1 Sol (Low) | 47.1 | #6 (GPT-6.1 Sol) |
| 14 | GPT-6.1 Terra (xHigh) | 46.4 | #10 (GPT-6.1 Terra) |
| 15 | Claude Sonnet 5.5 (High) | 43 | #13 (Claude Sonnet 5.5) |
No result here: #3 Claude Opus 5.5, #8 Gemini 4 Argon, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-6-1-terra-launch-terminal-bench-4-0 · How It Works · Data refreshed daily, snapshot 2026-10-27.