NaturalBench

NaturalBench is a benchmark for handling several input types together (images, documents, text). The official description says: "NaturalBench, benchmark for evaluating vision-language models on natural yes/no image questions". Below we describe in simple words what benchmarks of this type generally examine.

NaturalBench measures handling several input types together (images, documents, text). In detail, it checks:

• questions that mix text with images or document pages
• reading and extracting information from screenshots and scans
• understanding layout, titles, logos and figures in real documents
• staying accurate when the text content alone is not enough

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: NaturalBench, benchmark for evaluating vision-language models on natural yes/no image questions. Values reported as percentage. Source referenced on the LiquidAI/d1-2B model card.

What is the NaturalBench benchmark?

NaturalBench is a benchmark for handling several input types together (images, documents, text). The official description says: 'NaturalBench, benchmark for evaluating vision-language models on natural yes/no image questions'. Below we describe in simple words what benchmarks of this type generally examine.

What does NaturalBench test?

NaturalBench mainly measures handling several input types together (images, documents, text). Concretely, it checks: questions that mix text with images or document pages; reading and extracting information from screenshots and scans; understanding layout, titles, logos and figures in real documents.

How is the NaturalBench score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get NaturalBench data and scores?

The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/naturalbench/scores) or pull it directly from the Open LLM Radar API with any HTTP client.