News
Cohere releases Parse 5 document parsing model
Cohere has released the Parse 5 multimodal document parsing model with 2.3 billion parameters, capable of extracting text, tables, lists, forms, images, and descriptive text.
Cohere introduces the Parse 5 multimodal document parsing model.This model can extract text, tables, lists, forms, images, and descriptive text in the order of reading and output Markdown or HTML.
Parse 5 boasts 2.3 billion parameters and 8K contexts, and is now available through the Parse API, Microsoft Foundry, AWS SageMaker, and Model Vault.
refer to:Cohere Update Log