DocVQA
by CVC Barcelona · free · Last verified 2026-03-01
DocVQA is a large-scale dataset and benchmark for Visual Question Answering on document images. It challenges models to answer questions by reading and interpreting text, understanding layouts, and reasoning about information within complex documents like forms, invoices, and reports. It serves as a standard for evaluating document intelligence systems.
https://www.docvqa.org ↗B
B—Above Average
Adoption: B+Quality: AFreshness: B+Citations: B+Engagement: F
Specifications
- License
- Apache-2.0
- Pricing
- free
- Capabilities
- Benchmarking multimodal model performance, Evaluating visual question answering on documents, Assessing text extraction (OCR) in context, Testing comprehension of complex document layouts, Measuring reasoning over structured and unstructured text, Standardized evaluation for comparing document AI models, Providing a large-scale dataset of document images and QA pairs
- Integrations
- Use Cases
- [object Object], [object Object], [object Object], [object Object]
- API Available
- No
- Evaluated Models
- claude-4, gpt-5, gemini-2.5-pro
- Metrics
- anls-score, accuracy
- Methodology
- Questions about scanned document images requiring text extraction and layout understanding. Evaluated using Average Normalized Levenshtein Similarity (ANLS) metric.
- Last Run
- 2026-02-10
- Tags
- benchmark, dataset, document-ai, document-understanding, evaluation, information-extraction, multimodal, nlp, ocr, vqa
- Added
- 2026-03-17
- Completeness
- 0.7%
Index Score
63.1Adoption
72
Quality
84
Freshness
78
Citations
70
Engagement
0