Cohere Releases Parse 5 for Parsing Tables and Images in Documents

cohere · x · 2026-08-27

Cohere has released Parse 5, a document parsing model that recognizes text along with tables, forms, and images, turning them into clean, machine-readable documents with bounding boxes. It is designed for extracting structured data, building RAG systems, and automated document processing. It is available via Cohere API, AWS Sagemaker, Microsoft Foundry, and a free HuggingFace Space.

Related event: Cohere Launches Parse 5: Cost-Effective Document Parsing Model(6 posts)→

Original post →

More from Models

Models channel →