Description of the feature request:

This feature request proposes adding runnable Jupyter/Colab notebooks to the Gemini Cookbook, specifically to demonstrate the LearnLM model examples.
These notebooks would provide practical, executable code implementations for the scenarios currently described in the LearnLM documentation, such as 'Test prep', 'Teach a concept', 'Releveling', 'Guide a student through a learning activity', and 'Homework help'.
Having runnable notebooks would allow developers to quickly test, experiment with, and adapt the provided system instructions and prompts for their learning-focused applications using the Gemini API.

What problem are you trying to solve with this feature?

Currently, the LearnLM documentation provides excellent textual examples of system instructions and user prompts. However, developers need to manually copy, paste, and potentially adapt these into AI Studio or their code environment to see them in action via the API.
This adds a step and potential friction point for developers wanting to evaluate and integrate LearnLM quickly. Providing ready-to-run notebooks would significantly lower the barrier to entry for hands-on experimentation, making it faster and easier for developers to understand LearnLM's capabilities and start building with it. It bridges the gap between reading the concepts and running the code.

Any other information you'd like to share?

We believe executable examples are a highly effective way to learn and adopt new APIs and models. Adding notebooks for LearnLM would align with the cookbook's practical, hands-on nature and greatly benefit developers exploring LearnLM's potential for educational tools and applications. Ideally, each notebook would demonstrate setting up the LearnLM model, applying the specific system instruction, and running the example prompt(s). Thanks for considering!