This study systematically assesses how well Large Language Models (LLMs) perform on professional psychological counseling tasks, revealing their potential and limitations in mental health applications.

• Models were evaluated using 1,096 questions from China's National Counselor Examination
• Results show promising capabilities but highlight specific areas where LLMs still fall short
• Different models demonstrated varying levels of proficiency in counseling skills
• Findings establish benchmarks for measuring AI readiness in mental health contexts

This research is critical for the medical sector as it provides evidence-based insights on whether AI could safely supplement mental health services amid growing global demand and resource constraints.

Psychological Counseling Ability of Large Language Models