Get a working starting point for building an app that grounds Gemini answers in your own documents using File Search.
Experiment with Gemini's LiveAPI for real-time multimodal interaction without reading the full reference docs.
Run a batch mode example to send many Gemini requests at a discount rate.
Learn how to generate images or music using the Gemini Nano-Banana or Lyria models.
| google-gemini/cookbook | stefan-jansen/machine-learning-for-trading | ufund-me/qbot | |
|---|---|---|---|
| Stars | 17,212 | 17,322 | 17,322 |
| Language | Jupyter Notebook | Jupyter Notebook | Jupyter Notebook |
| Setup difficulty | easy | hard | hard |
| Complexity | 2/5 | 4/5 | 4/5 |
| Audience | developer | data | data |
Figures from each repo's GitHub metadata at analysis time.
Each notebook has a Colab button so you can run it in the browser with no local setup needed.
This repository is the official cookbook for Google's Gemini API. The Gemini API is a service that lets developers send text, images, audio, video and other inputs to Google's Gemini family of AI models and get answers back. The cookbook does not contain the API itself, it is a collection of hands-on tutorials and runnable examples that show how to use each feature in practice. The full reference documentation lives at ai.google.dev. The material is organised into two main sections. Quick Starts are step-by-step guides covering one feature at a time, from the basic Get Started introduction through to specific capabilities like Webhooks for notifications about long-running jobs, Inference tiers for trading off speed against cost, File Search for grounding answers in your own documents, Grounding with Google Search, YouTube, URLs or Google Maps, Batch mode for sending many non-urgent requests at a discount, and the multimodal LiveAPI for real-time interaction. Examples then show how to combine several features into complete use cases. Recent additions cover the Nano-Banana 2 and Nano-Banana Pro image generation models, the Lyria 3 music generation model, the Veo 3.1 video generation model, a text-to-speech guide, and Gemini Robotics-ER 1.5 for spatial reasoning. Everything is delivered as Jupyter notebooks, with a Colab button on each one so you can open and run it in the browser. Someone would use this if they want a working starting point for building on top of Gemini, or to learn how a particular feature behaves without reading the full reference docs first. The tech stack is Python in Jupyter notebooks. The full README is longer than what was provided.
Official collection of runnable Jupyter notebook tutorials for Google's Gemini AI API covering text, image, audio, video, and real-time interaction, open any notebook in Colab with one click.
Mainly Jupyter Notebook. The stack also includes Python, Jupyter Notebook, Google Colab.
License not mentioned in the explanation.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.