OpenAI MentalHealthBench - Behavior - Context Seeking: leaderboard
Behavior axis of OpenAI's MentalHealthBench (launch post and paper, 2026-09-23): identifying and using the context needed, asking follow-up questions when necessary and not when context is already sufficient, scored on the 1,124 rubric criteria tagged with this axis. The post's axis score is the signed rubric score on this axis, normalized to its possible range (0-100, higher is better); GPT-5.6 Sol grades, models run at API default reasoning effort. The post's 15 models only: the paper's behavior figure uses a different scale.
Metric: Axis score, normalized to its possible range (%). Source: openai.com. Saturation forecast: Around June 2028. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 56.88 | #3 |
| 2 | GPT-6 Astra | 52.83 | #5 |
| 3 | Muse Spark 1.3 | 52.26 | #31 |
| 4 | Claude Fable 5.1 | 48.62 | #10 |
| 5 | GPT-6 Sol | 47.43 | #17 |
| 6 | Claude Sonnet 5 | 46.11 | #67 |
| 7 | GPT-5.6 Sol (August 2026) | 36.48 | |
| 8 | GPT-6 Luna | 36.24 | #62 |
| 9 | Claude Haiku 4.5 | 35.91 | #291 |
| 10 | GPT-5.6 Luna (August 2026) | 34.16 | |
| 11 | Gemini 3.8 Flash | 25.39 | #30 |
| 12 | Grok 4.7 | 24.31 | #26 |
| 13 | Gemini 3.1 Pro (Preview) | 22.11 | #66 |
| 14 | GPT-4o (Mar 2025) | 18.72 | #339 |
| 15 | Gemini 2.5 Pro | 15.32 | #161 |
No result here: #6 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=openai-mentalhealthbench-behavior-context-seeking · How It Works · Data refreshed daily, snapshot 2026-11-07.