DocAttriBench - VisualWebBench: leaderboard

Metric: Overall answer correctness (%; grounded answer generation on VisualWebBench (233 test questions): an answer counts only when its text matches the reference and its evidence box is correct; zero-shot, with model-specific prompts and box parsers; element-level evidence boxes from the MAPPET attribution pipeline over Docling layout regions, a box counting when IoU >= 0.5). Source: arxiv.org. Saturation forecast: Around January 2028. 16 models tracked.

Top models

#ModelScore
1Qwen 2.5 VL 32B Instruct19.3
2Qwen 3 VL 8B8.6
3Qwen 2.5 VL 7B Instruct3
4InternVL3-38B1.7
5InternVL3-8B1.3
6InternVL2.5-2B0.4

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