ClinHallu: leaderboard
ClinHallu (2026) tests medical multimodal models on radiology, pathology and clinical image questions while tracing where hallucinations arise in their reasoning. The score is LLM-judged answer accuracy averaged over VQA-RAD, PathVQA, MedFrameQA and MedXpertQA, in percent; higher is better.
Metric: Answer accuracy (%), macro-averaged over the ClinHallu test items of VQA-RAD, PathVQA, MedFrameQA and MedXpertQA (medical VQA with validated reference traces of visual recognition, knowledge recall and reasoning integration); the model writes a structured trace and answer (temperature 0.7) and a Qwen3.5-27B judge grades them (temperature 0.01); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 80.1 | #164 |
| 2 | Qwen 3.5 9B | 69.1 | #803 |
| 3 | Qwen 3.5 4B | 64.3 | #1175 |
| 4 | MedGemma-4B-IT | 53.2 | #2716 |
| 5 | Qwen 2.5 VL 7B Instruct | 42.7 | #1833 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=clinhallu · How It Works · Data refreshed daily, snapshot 2026-11-13.