VisualWebBench
multimodal
+
+
+
+
About
VisualWebBench is a comprehensive multimodal benchmark assessing AI models' understanding and grounding capabilities in web scenarios, featuring 1,500 human-curated instances from 139 real websites across 87 sub-domains. This evaluation tests seven web-related tasks including captioning, webpage QA, OCR, element grounding, and action prediction, measuring multimodal models' proficiency in web interface understanding.
+
+
+
+
Evaluation Stats
Total Models2
Organizations1
Verified Results0
Self-Reported2
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
2 models
Top Score
79.7%
Average Score
78.7%
High Performers (80%+)
0Top Organizations
#1Amazon
2 models
78.7%
+
+
+
+
Leaderboard
2 models ranked by performance on VisualWebBench
| License | Links | ||||
|---|---|---|---|---|---|
| Nov 20, 2024 | Proprietary | 79.7% | |||
| Nov 20, 2024 | Proprietary | 77.7% |
