Gemini on Vertex AI

This folder contains comprehensive set of resources on how to use Gemini on Vertex AI in conjunction with other Vertex AI services for enterprise Generative AI use cases and solutions.

Requirements

To run the notebooks you'll need access to a Google Cloud project with the Vertex AI API enabled.

Using this repository

	Description	Contents
Gemini prompting recipes: prompting_recipes/	Code samples demonstrating prompting techniques for Gemini specific use cases	multimodal prompting, long context window, pdf processing
Gemini Evals Playbook: evals_playbook/	Tools built with Vertex AI services to streamline experimentation and evaluation of Generative AI applications used for pre-production performance testing, regression testing, or migration from other LLM APIs.	evaluating prompt iterations, sampling parameters, long context

[!TIP] Refer here for introduction to prompt design, and here for Vertex Gemini prompting strategies and best practices.

Getting Help

If you have any questions or find any problems, please report through GitHub issues.