Our mission is to democratize computer vision
Explore dozens of open tools to help you leverage computer vision faster and with less friction.
- GitHub
150+
Open source projects
- GitHub
93k+
GitHub stars
- PyPI
1M+
Monthly pip downloads
Featured projects
The libraries and models our community uses most.
- Vision utilities
Supervision
A range of utilities to help integrate computer vision into your application, covering functions from annotation to object tracking.
51,180GitHub starsMITPython
- Detection model
RF-DETR
A real-time, transformer-based object detection model architecture developed by Roboflow and released under the Apache 2.0 license.
9,768GitHub starsApache-2.0Python
- Inference server
Inference
An easy-to-use, production-ready inference server for computer vision supporting deployment of many popular model architectures and fine-tuned models.
2,476GitHub starsPython
- Object tracking
Trackers
A unified library offering clean-room re-implementations of leading multi-object tracking algorithms.
3,940GitHub starsApache-2.0Python
- Tutorials
Notebooks
A collection of open source Jupyter notebooks showing how to train and work with the latest state-of-the-art computer vision models.
9,706GitHub starsJupyter Notebook
- Auto-labeling
Autodistill
A framework for creating computer vision models without hand-labeling images. Uses big slow foundation models to train small fast supervised models.
2,789GitHub starsApache-2.0Python
- Sports analytics
Sports
Applications of computer vision in sport analytics, from ball tracking to reading jersey numbers.
5,414GitHub starsMITPython
- VLM fine-tuning
Maestro
Fine-tune vision-language models like Florence-2, PaliGemma 2 and Qwen2.5-VL, with ready-made recipes that handle configuration, data loading and training.
2,695GitHub starsApache-2.0Python
Roboflow Python Package
634GitHub stars
The official Roboflow Python package. Manage your datasets, models, and deployments.
Computer Vision Skills
44GitHub stars
Skills and MCP configuration that let coding agents like Claude Code, Codex and Cursor manage data, train, evaluate and deploy with Roboflow.
Model Leaderboard
110GitHub stars
Object detection models compared on the COCO dataset across a range of metrics, computed with Supervision.
RF100-VL
144GitHub stars
Code from the Roboflow100-VL paper, a multi-domain object detection benchmark for vision-language models.
PolygonZone
87GitHub stars
A web utility to draw polygons and retrieve their coordinates for computer vision applications.
OpenAI Vision Experiments
1,698GitHub stars
A resource for anyone who wants to experiment with and build on the OpenAI vision API.
Universe
The world's largest collection of open source computer vision datasets and APIs: 200M+ images, 200k+ datasets and 50k+ fine-tuned models.