Benchmark
InfoVQA Benchmark
What is InfoVQA?
InfoVQA dataset with 30,000 questions and 5,000 infographic images requiring joint reasoning over document layout, textual content, graphical elements, and…
- Released
- 2021-04-26
- Modality
- multimodal
- Evaluates
- multimodal, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 10
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/infovqa?top_n=500
InfoVQA results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek VL2 | DeepSeek | 0.781 | 2024-12-13 | evidence |
| DeepSeek VL2 Small | DeepSeek | 0.758 | 2024-12-13 | evidence |
| DeepSeek VL2 Tiny | DeepSeek | 0.661 | 2024-12-13 | evidence |
| Gemma 3 12B | 0.649 | 2025-03-12 | evidence | |
| Gemma 3 27B | 0.706 | 2025-03-12 | evidence | |
| Gemma 3 4B | 0.5 | 2025-03-12 | evidence | |
| North Micro Vision Instruct | Cohere | 0.652 | 2026-08-12 | evidence |
| Phi-4-multimodal-instruct | Microsoft | 0.727 | 2025-02-01 | evidence |
| Qwen2.5 VL 32B Instruct | Qwen | 0.834 | 2025-02-28 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.826 | 2025-01-26 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.