MedRepBench - Value Recall: leaderboard

Metric: Field-level recall (%; measured value of each ground-truth lab item, matched by exact normalized item name within the model's top-K items, K = number of reference items; VLM on the report image only). Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.

Top models

#ModelScore
1Qwen 2.5 VL 7B73.17
2Llama 4 Maverick72.26
3InternVL3-8B71.29
4InternVL3-14B67.69
5Llama 4 Scout65.69
6InternVL3-38B64.83

Interactive version: theaggregate.ai/benchmark?slug=medrepbench-value-recall · How It Works · Data refreshed daily, snapshot 2026-09-26.