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.
More from Models
- antirez backs low-bit DeepSeek 4.1 Flash over Qwen 27B dense on Apple M5 Macs — antirez · 2026-11-12
- Gemini re-renders images as text when vision is removed, dropping only 6% WSR — zacharynado · 2026-11-12
- Bare-harness coding benchmark: Fable 5.2 best overall, Gemini 4 Carbon best value — Semantic_meaning · 2026-11-12
- Google's Physical-Engineering's Last Exam: all 14 frontier models score below 50% — dl_weekly · 2026-11-12
- Reddit's 2026 LLM hall of fame: Opus 5.5 still unbeaten after 9 rival releases — G0dZylla · 2026-11-12
- Commit leak reveals unreleased Gemini 4 variants: Argon, Hydrogen and Helix (unconfirmed) — BlackHC · 2026-11-11