Skip to content

Dataset · getomni-ai

OCR benchmark for multimodal LLMs

getomni-ai/ocr-benchmark

Evaluation set for comparing OCR and text/JSON extraction by multimodal LLMs like gpt-4o and gemini-2.0.

What's inside

The benchmark compares OCR and data extraction capabilities of multimodal LLMs, including gpt-4o and gemini-2.0. Text and JSON extraction accuracy is evaluated. Modalities are image and text, size per metadata is 1K–10K; language is not specified in the card.

What you can do with it

  • Check OCR quality of a multimodal LLM
  • Evaluate JSON extraction from images
  • Compare gpt-4o and gemini-2.0 on OCR
Size
1K–10K rows
Downloads this month
1.7K
Likes
75

How to load it

from datasets import load_dataset

ds = load_dataset("getomni-ai/ocr-benchmark")