Since the release of the beta version 0.7.0, you can link multiple machine learning models together, allowing you to explore an even more extensive and comprehensive set of creative workflows.
What is Model Chaining?
Model chaining is an experimental feature inside of Runway that allows you to use the output of one model as the input to another one, effectively chaining them together. This becomes very helpful when you want to create a continuous pipeline of processing for a given input.
Things to keep in mind 🧠
- Please note that this is still an experimental feature! 🧪👩🏾🔬
- Not all models can be connected to one another. Only models with matching input/output types can be chained together. For example, models that expects text as input can only be the downstream chains of models that produce text-based output.
- Be sure all models in the chain are running before expecting the final output!
Learn More 👇🏽
To learn how to chain models inside the RunwayML app, read this tutorial.