Our Mission is to Democratize Computer Vision

A range of utilities to help integrate computer vision into your application, covering functions from annotation to object tracking.

A collection of open source Jupyter notebooks showing how to train and work with the latest state-of-the-art computer vision models.

A framework for creating computer vision models without hand-labeling images. Uses big slow foundation models to train small fast supervised models.

An easy-to-use, production-ready inference server for computer vision supporting deployment of many popular model architectures and fine-tuned models.

Multimodal-Maestro gives you more control over large multimodal models, like GPT-4 Vision, LLaVA or CogVLM, to get the outputs you want.

The world's largest collection of open source computer vision datasets and APIs, with 200 million+ images, 200k+ datasets and 50k+ fine-tuned models.