OpenAI GPT-6.1 Terra System Card - MentalHealthBench (October 2026) - overall: leaderboard
Official OpenAI October 26, 2026 GPT-6.1 Terra system-card addendum; Mean task rubric score (%; four responses per task, max effort). 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: Mean task rubric score (%; four responses per task, max effort).
Metric: Mean task rubric score (%; four responses per task, max effort). Source: deploymentsafety.openai.com. Saturation forecast: Around 2029. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 58.7 | #5 (GPT-6 Astra) |
| 2 | GPT-6.1 Sol (Max) | 57.9 | #6 (GPT-6.1 Sol) |
| 3 | GPT-6.1 Terra (Max) | 55.5 | #10 (GPT-6.1 Terra) |
| 4 | GPT-6 Sol (Max) | 54.2 | #17 (GPT-6 Sol) |
| 5 | GPT-6 Luna (Max) | 51.7 | #60 (GPT-6 Luna) |
| 6 | GPT-5.6 Sol (Max) | 46.7 | #20 (GPT-5.6 Sol) |
| 7 | GPT-5.6 Terra (Max) | 46.2 | #39 (GPT-5.6 Terra) |
| 8 | GPT-5.6 Luna (Max) | 44.4 | #73 (GPT-5.6 Luna) |
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-system-card-mentalhealthbench-october-2026-overall · How It Works · Data refreshed daily, snapshot 2026-10-27.