Multi-turn psychiatric consultation on Psychiatric Consultation Human Eval (test)
4.1Professional ReliabilityMIND
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MIND2026.03 | 4.1 | 3.95 | 82 | |
| DeepSeek-V32026.03 | 3.25 | 3.6 | 58.5 | |
| Baichuan-M22026.03 | 2.75 | 3.35 | 43 | |
| Qwen3-8B2026.03 | 2.5 | 3.05 | 35.5 |