Cohere’s Parse 5 Promises Efficient Multi-Modal Information Extraction From Complex Documents

Cohere has released Parse 5, a 2.3-billion-parameter multimodal model designed to convert complex enterprise documents into structured Markdown format. By providing bounding box coordinates, the system enables visual grounding for data extraction tasks in documents like PDFs. This tool is intended to help businesses automate the processing of dense, visually rich files that are often difficult for standard text-only models to interpret.