MentalHealthBench: Frontier Models Improving but Gaps Remain in Context-Seeking

thekaransinghal · x · 2026-09-24

MentalHealthBench scores show steady progress in how frontier models respond to realistic mental health situations, with clear differences across models. Improvement is most needed in seeking context from users and supporting users in making their own decisions.

Related event: OpenAI Releases MentalHealthBench Built With 80+ Clinicians(11 posts)→

Original post →

More from Models

Models channel →