Jerry Liu Pledges Open Benchmarks for Document Parsing

Jerry Liu, co-founder and CEO of LlamaIndex, posted that unlike other players in the OCR space the company does not block competitors from its platform. He stated a commitment to open reproducible benchmarks for document parsing and extraction on ParseBench and ExtractBench. The linked ParseBench site describes evaluation across tables charts content faithfulness semantic formatting and visual grounding. The ExtractBench site describes a schema-guided extraction benchmark for enterprise documents where every system fills the same JSON schema from the same document.