Notebook experiments to model assets
Published Updated
This is the Learning Atlas from Orbit. Get your invite
Notebook environmentsHostinginfrastructuredatabasesLocal models and inference
-
Step 1: Jupyter
Why this matters here
Jupyter starts the path because notebooks are the common workspace for exploratory code, data checks, and model experiments.
In short
An open-source web application that allows you to create and share documents that contain live code, equations, visualizations, and narrative text.
-
Step 2: Google Colab
Why this matters here
Colab follows because it gives the same notebook pattern hosted compute and easy sharing.
In short
A free, hosted Jupyter notebook service provided by Google that requires no setup and provides free access to computing resources including GPUs and TPUs.
-
Step 3: Hugging Face
Why this matters here
Hugging Face belongs after hosted notebooks because learners often move from experiments to finding or publishing model assets there.
In short
The central collaborative hub for the machine learning community, hosting hundreds of thousands of open-weight models, datasets, and AI applications.
-
Step 4: GGUF
Why this matters here
GGUF matters once a model asset is selected, especially if the learner wants a local runnable artifact rather than only a hosted demo.
In short
A highly optimized file format specifically designed for storing and loading quantized large language models rapidly on consumer hardware.
-
Step 5: LM Studio
Why this matters here
LM Studio closes the path by giving the downloaded model file a desktop environment for testing and serving locally.
In short
A desktop graphical user interface (GUI) application that allows users to easily search, download, and chat with open-weight models locally.