RunwayML is a platform for artists to use machine learning tools in intuitive ways without any coding experience for media ranging from video, audio, to text. On RunwayML, users can create and publish pre-trained machine learning models for applications such as as generating photorealistic images or image captions. Users also can train models of their own and import new models directly from GitHub.
RunwayML offers a gamut of models for a variety of purposes but can be categorized into three broad types: models that identify objects and people, models that transform content, and models that generate new media from the training data. Creators can input their data directly into RunwayML, in which it’s machine learning algorithm finds patterns in the dataset to create a model that can perform specific tasks. Results can be easily integrated in other software such as Unity and Adobe Photoshop.
RunwayML runs the model in a remote, high-computing GPU instead of overloading the users’ local hardware. Praised for lowering the barrier of entry for creating art with AI, RunwayML requires no coding experience and is able to run powerful models without extensive hardware requirements.