InfoVQA
InfoVQA dataset with 30,000 questions and 5,000 infographic images requiring joint reasoning over document layout, textual content, graphical elements, and data visualizations with elementary reasoning and arithmetic skills
Progress Over Time
Interactive timeline showing model performance evolution on InfoVQA
State-of-the-art frontier
Open
Proprietary
InfoVQA Leaderboard
9 models
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 34B | — | — | ||
| 2 | Alibaba Cloud / Qwen Team | 8B | — | — | ||
| 3 |  DeepSeek | 27B | 129K | — | ||
| 4 | DeepSeek | 16B | — | — | ||
| 5 |  Microsoft | 6B | 128K | $0.05 / $0.10 | ||
| 6 |  Google | 27B | 131K | $0.10 / $0.20 | ||
| 7 | DeepSeek | 3B | — | — | ||
| 8 | Google | 12B | 131K | $0.05 / $0.10 | ||
| 9 | Google | 4B | 131K | $0.02 / $0.04 |
Notice missing or incorrect data?
FAQ
Common questions about InfoVQA
InfoVQA dataset with 30,000 questions and 5,000 infographic images requiring joint reasoning over document layout, textual content, graphical elements, and data visualizations with elementary reasoning and arithmetic skills
The InfoVQA paper is available at https://arxiv.org/abs/2104.12756. This paper provides detailed information about the benchmark methodology, dataset creation, and evaluation criteria.
The InfoVQA leaderboard ranks 9 AI models based on their performance on this benchmark. Currently, Qwen2.5 VL 32B Instruct by Alibaba Cloud / Qwen Team leads with a score of 0.834. The average score across all models is 0.716.
The highest InfoVQA score is 0.834, achieved by Qwen2.5 VL 32B Instruct from Alibaba Cloud / Qwen Team.
9 models have been evaluated on the InfoVQA benchmark, with 0 verified results and 9 self-reported results.
InfoVQA is categorized under multimodal and vision. The benchmark evaluates multimodal models.