license: mit size_categories: - 1K<n<10K
A comprehensive benchmark that compares OCR and data extraction capabilities of different multimodal LLMs such as gpt-4o and gemini-2.0, evaluating both text and JSON extraction accuracy.
Benchmark Results (Feb 2025) | Source Code
