17.8K ·GitHub
What is this?
A collection of notebooks, code samples and sample apps for building generative AI on Google Cloud with Gemini and its Agent Platform, the evolution of Vertex AI.
A collection of notebooks, code samples and sample apps for building generative AI on Google Cloud with Gemini and its Agent Platform, the evolution of Vertex AI.Is this a library I add to my project, or something else?Neither — it's a repo of runnable Jupyter notebooks and sample code you copy or execute, organized into topic folders rather than an installable package.
Neither — it's a repo of runnable Jupyter notebooks and sample code you copy or execute, organized into topic folders rather than an installable package.How do I get the notebooks running?Start with the setup-env/ folder — it walks you through setting up a Google Cloud project, the Gen AI Python SDK, and a notebook environment on Google Colab or Workbench.
Start with the setup-env/ folder — it walks you through setting up a Google Cloud project, the Gen AI Python SDK, and a notebook environment on Google Colab or Workbench.Do I need a Google Cloud account to run these?Yes — you need a Google Cloud project with the Gen AI Python SDK, which setup-env/ helps you configure. The code itself is Apache-2.0 free, but it calls Google Cloud services like the Gemini API under your own project.
Yes — you need a Google Cloud project with the Gen AI Python SDK, which setup-env/ helps you configure. The code itself is Apache-2.0 free, but it calls Google Cloud services like the Gemini API under your own project.What will I find inside — only Gemini text examples?No — besides the gemini/ notebooks there's search/ for Vertex AI Search, rag-grounding/ for RAG, vision/ built on Imagen and Veo, and audio/ built on Chirp speech models.
No — besides the gemini/ notebooks there's search/ for Vertex AI Search, rag-grounding/ for RAG, vision/ built on Imagen and Veo, and audio/ built on Chirp speech models.Is this code safe to take straight to production?Not as-is — the disclaimer says the repository is not an officially supported Google product and the code is for demonstrative purposes only.
Not as-is — the disclaimer says the repository is not an officially supported Google product and the code is for demonstrative purposes only.Where should a beginner start?In the gemini/ folder — it has starter notebooks for discovering Gemini, plus use cases, function calling and sample apps.
In the gemini/ folder — it has starter notebooks for discovering Gemini, plus use cases, function calling and sample apps.Are there ready-made agents, or do I build from scratch?The README points to companion repos: ADK Samples offers ready-to-use agents from simple conversational bots to multi-agent workflows, and the Agent Starter Pack ships production-ready Gen AI agent templates for Google Cloud.
The README points to companion repos: ADK Samples offers ready-to-use agents from simple conversational bots to multi-agent workflows, and the Agent Starter Pack ships production-ready Gen AI agent templates for Google Cloud.Does it keep up with new Gemini releases?Yes — the README leads with the latest Gemini models (Gemini 3.8 Flash) and the newly released Gemini Enterprise Agent Platform, the evolution of Vertex AI.
Yes — the README leads with the latest Gemini models (Gemini 3.8 Flash) and the newly released Gemini Enterprise Agent Platform, the evolution of Vertex AI.What are similar projects?cookbook, gemini-skills, adk-python, agents-cli, gemini-fullstack-langgraph-quickstart, storyboard-ai, Hands-On-Large-Language-Models, agentic-rag-for-dummies
cookbook, gemini-skills, adk-python, agents-cli, gemini-fullstack-langgraph-quickstart, storyboard-ai, Hands-On-Large-Language-Models, agentic-rag-for-dummiesWhich topics is it in?vertex-ai, samples, gemini-api, google-cloud, function-calling, gcp, langchain, speech-recognition, google, jupyter notebook, multimodal, video-generation
vertex-ai, samples, gemini-api, google-cloud, function-calling, gcp, langchain, speech-recognition, google, jupyter notebook, multimodal, video-generation