MedDialBench - Fabricating: leaderboard
Metric: Diagnostic accuracy (%) over 85 OSCE cases (each cell counts whole cases out of 85) after a free-form consultation with a patient who fabricates symptoms (logic, extreme), the final diagnosis judged correct or not by Qwen3-Max; higher is better. Source: arxiv.org. Saturation forecast: Around December 2027. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (Non-reasoning) | 63.5 |
| 2 | Gemini 3.1 Pro (Preview) | 60 |
| 3 | Claude Opus 4.6 | 55.3 |
| 4 | DeepSeek V3.2 (Non-reasoning) | 54.1 |
| 5 | Qwen 3.5 Plus (Non-reasoning) | 49.4 |
Interactive version: theaggregate.ai/benchmark?slug=meddialbench-fabricating · How It Works · Data refreshed daily, snapshot 2026-10-07.