LingxiDiagBench (Static) - 4-Class Accuracy: leaderboard
Metric: Accuracy (times 100, 0-100) on four-way classification (pure depression, pure anxiety, mixed depression-anxiety, other); LingxiDiagBench-Static on the 1,000 test cases of LingxiDiag-16K (synthetic Chinese psychiatric consultation dialogues with EMRs matching the Shanghai Mental Health Center outpatient distribution): zero-shot diagnosis from the complete dialogue; higher is better. Source: arxiv.org. Saturation forecast: Around 2033. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.1 Fast | 47 | #208 |
| 2 | Claude Haiku 4.5 | 44.4 | #271 |
| 3 | DeepSeek V3.2 | 44.1 | #198 |
| 4 | Qwen 3 32B | 43.8 | #424 |
| 5 | GPT-5 Mini | 43.4 | #176 |
| 6 | Gemini 3 Flash | 42.2 | #93 |
| 7 | Kimi K2 (Thinking) | 40.9 | #236 (Kimi K2) |
| 8 | Qwen 3 8B | 40.8 | #667 |
| 9 | GPT-OSS-20B | 40.2 | #499 |
| 10 | Qwen 3 4B | 40.1 | #823 |
| 11 | Qwen 3 1.7B | 39.2 | #1186 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=lingxidiagbench-static-4-class-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-11.