roboflow/notebooks

A collection of tutorials on state-of-the-art computer vision models and techniques. Explore everything from foundational architectures like ResNet to cutting-edge models like RF-DETR, YOLO11, SAM 3, and Qwen3-VL.

What is roboflow/notebooks?

roboflow/notebooks is a GitHub repository that hosts a growing collection of computer vision tutorials in the form of Jupyter notebooks. These tutorials teach you how to use state-of-the-art (SOTA) models for a variety of vision tasks.

What does it contain?

The repository currently lists 61 model tutorials (as of the README). Each tutorial is a notebook that demonstrates a specific model or technique. Examples from the table include:

Fine-tuning YOLO26 for object detection or instance segmentation
Object detection with Google Gemini 3.5 Flash
OCR (optical character recognition) with GLM-OCR
Segmenting images or videos with SAM3
Zero-shot object detection with Qwen2.5-VL or YOLOE
Fine-tuning RF-DETR for detection or segmentation
Data extraction (e.g., JSON) with models like PaliGemma2 or Qwen2.5-VL

Each notebook entry includes:

A link to the notebook file on GitHub
Buttons to open it directly in Google Colab, Kaggle, or SageMaker Studio Lab (so you can run it without local setup)
Optional links to complementary materials like blog posts and YouTube videos
Links to the underlying model repository or research paper (e.g., GitHub, arXiv)

How do you use it?

You can browse the table in the README, pick a tutorial that interests you, and click one of the “open in” badges to launch the notebook in a cloud environment. The notebook itself will guide you through the steps—loading data, running the model, and often fine-tuning it on a custom dataset.

Who is it for?

It’s aimed at anyone who wants to learn or apply modern computer vision techniques, from object detection and segmentation to pose estimation and OCR. The tutorials assume some familiarity with Python and machine learning, but they are designed to be hands-on and practical.

What else is in the README?

The README also links to other Roboflow projects (inference, autodistill, RF-DETR) and to Roboflow’s social channels (YouTube, blog, forum, etc.), but the core content is the notebook collection itself.

Note: This explanation is based solely on the provided README. No external features or facts have been added.