ParseBench: Document Parsing & OCR Benchmark
ParseBench is an open benchmark by LlamaIndex for evaluating document parsing and OCR systems on enterprise documents. It uses human-verified pages and deterministic test rules across five dimensions: tables, charts, content faithfulness, semantic formatting, and visual grounding. The benchmark provides a public dataset, reproducible evaluation code, a research paper, and an interactive leaderboard.
Dataset and evaluation
- Download the ParseBench dataset on Hugging Face
- Run the open-source ParseBench evaluation
- Research paper
