Sword Health Launches MindEval, the First Multi-Turn Mental Health Benchmark for Evaluating Large Language Models in Realistic Therapeutic Dialogue
Sword Health Launches MindEval, the First Multi-Turn Mental Health Benchmark for Evaluating Large Language Models in Realistic Therapeutic Dialogue MindEval reveals that 12 state-of-the-art AI models struggle with realistic therapy-style conversations, especially in severe cases–and neither model size nor reasoning capabilities reliably improve therapeutic behavior GlobeNewswire December 09, 2025 New York, NY, Dec. 09, 2025 […]