Tutorials

Overview This tutorial walks through a “Count Everything” application built with Falcon Perception, a vision model designed for open-vocabulary object detection. ...

Overview This tutorial walks you through the process of fine-tuning a Small Language Model (SLM) to work with tool calling data — datasets that enable the model to understand and execute calls to external tools within a conversation, thereby enhancing automation capabilities and accuracy. ...

Overview Not every task is about retrieving facts, sometimes you need the model to think, weigh options, and walk through a problem in multiple steps. That is where the new Falcon reasoning models come in. Built to deliver structured and logical outputs, they extend the capabilities of the Falcon family into domains like problem solving, mathematics, code logic, and multi stage decision making, while still being lightweight enough to run on laptops or compact devices. We have already gone through the series guiding you on how to work with the instruct model versions, including Falcon H1 Instruct and Falcon3 Instruct , and now it is time to start exploring reasoning models to see how they change the way we approach complex workloads. ...

Overview If you have ever wanted to run a powerful AI model directly on your laptop for quick experiments or personal projects, Falcon 3 makes it simple. You do not need expensive servers or constant internet access, just your device and the model running locally. It is built for speed, lightweight setup, and customization so you can focus on creating, not configuring. ...

Overview Sometimes your projects are about more than quick answers. You might need to read through hundreds of pages, keep a long conversation going, or work in multiple languages. Falcon H1 was designed for exactly that. By combining Transformer-based attention with State Space Models (SSM), it understands complex information while still running efficiently on everyday hardware. ...