MedDialBench - Fabricating + Incoherent: leaderboard
MedDialBench (2026) has doctor-agent LLMs interview a simulated patient (Claude Opus 4.5 following case-specific behaviour scripts) on 85 OSCE cases and name a diagnosis, which Qwen3-Max judges correct or not. This board uses a patient who fabricates symptoms and speaks incoherently. Diagnostic accuracy in percent; higher is better.
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 and is incoherent (combination C2), the final diagnosis judged correct or not by Qwen3-Max; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 52.9 | #102 | |
| 2 | GPT-5.4 | Non-reasoning | 51.8 | #138 |
| 3 | Claude Opus 4.6 | 41.2 | #114 | |
| 4 | Qwen 3.5 Plus | Non-reasoning | 30.6 | #250 |
| 5 | DeepSeek V3.2 | Non-reasoning | 28.2 | #425 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=meddialbench-fabricating-plus-incoherent · How It Works · Data refreshed daily, snapshot 2026-11-22.