Granite Workshop¶
Welcome to our workshop! In this workshop we'll be using the InstructLab CLI to install and run a quantized Granite model before adding our own knowledge. After this we will show the simplified contribution workflow throught the InstructLab UI.
By the end of this workshop, you will be able to:
- Understand what Granite is and its general use cases.
- Understand what InstructLab is and its general use cases.
- Walk away with a local AI Co-Pilot for free, secure and air gapped!
- Create and test your own knowledge and skill submissions locally using the CLI and UI
- Learn a suggested workflow on how to fine tune a model with new knowledge and skills!
About this workshop¶
The introductory page of the workshop is broken down into the following sections:
Agenda¶
| Lab 0: Pre-work | Pre-work for the project |
| Lab 1: Installing a local AI | Overview of building a local Open Source AI |
| Lab 2: Using a local AI | Using a local Open Source AI |
| Lab 3: Setting up InstructLab | Setting up InstructLab |
| Lab 4: Tuning your local AI | Learning how to fine tune your local model |
| Lab 5: Using the InstructLab UI for AI Tuning | Leveraging the UI instead of the CLI for InstructLab |
Compatibility¶
This workshop has been tested on the following platforms:
- MacOS: version 14.5
- Windows: version 11
- Linux: Fedora 40