Document Understanding Available for Google Gemini 2.5 in OCI Generative AI

The Google Vertex AI Platform for OCI Generative AI now supports document understanding for the pretrained Gemini 2.5 Pro, Gemini 2.5 Flash, and Gemini 2.5 Flash-Lite models.

Gemini 2.5 models can process documents in PDF format using native vision capabilities to comprehend entire document contexts. This feature extends beyond basic text extraction, enabling the model to analyze and interpret content such as text, images, diagrams, charts, and tables, even in lengthy documents up to 1,000 pages. The document understanding feature supports extracting information into structured formats, summarizing key points, answering questions based on both visual and textual elements, and transcribing content (for example, to HTML) while preserving layouts and formatting for downstream applications.

For information about these models, see the Generative AI documentation.