ImajevBench

ImajevBench is a benchmark for handling several input types together (images, documents, text). The official description says: "ImajevBench v2.0-lite (mohit67890/imajev-bench on Hugging Face), image-based decision/judgment benchmark". Below we describe in simple words what benchmarks of this type generally examine.

ImajevBench 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: ImajevBench v2.0-lite (mohit67890/imajev-bench on Hugging Face), image-based decision/judgment benchmark. Values reported as percentage. Source: https://huggingface.co/datasets/mohit67890/imajev-bench

What is the ImajevBench benchmark?

ImajevBench is a benchmark for handling several input types together (images, documents, text). The official description says: 'ImajevBench v2.0-lite (mohit67890/imajev-bench on Hugging Face), image-based decision/judgment benchmark'. Below we describe in simple words what benchmarks of this type generally examine.

What does ImajevBench test?

ImajevBench 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 ImajevBench 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 ImajevBench data and scores?

The official source is https://huggingface.co/datasets/mohit67890/imajev-bench. 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/imajevbench/scores) or pull it directly from the Open LLM Radar API with any HTTP client.