Core Concepts

AI deployment made unbelievably easy.

Function streamlines AI deployment by removing the need to manage Docker containers, Kubernetes clusters, and other infrastructure.


With Function, you can execute AI prediction functions from anywhere. These predictors are just Python functions that consume and produce arbitrary data.

# Every predictor is literally just a `predict` function in a Jupyter notebook 😉
def predict (radius: int):
Return the area of a circle given its radius.
return 3.14 * radius * radius

You can also explore predictors on Function, without bringing your own.


Values are any data consumed or produced by a predictor. Following from the above example, I can make a prediction with a radius value:

# Make a prediction with a `radius` value of 2.5:

With this construct, using AI in your apps boils down to sending your input values to a predictor and processing the output values.

Welcome to Function