Whenever you run a model, you're creating a prediction.
A prediction is an object that represents a single result from running a model, including the inputs that you provided, the outputs that the model returned, as well as other metadata like the model version, the user who created it, the status of the prediction, and timestamps.
Here's an example of a prediction object:
In the example above, the prediction is generated from running a simple model called replicate/hello-world that takes a single string input and returns a single string as output.
The prediction object includes the following fields:
id
is a unique identifier for the prediction.input
is the input that you provided to the model.output
is the output that the model returned.status
is the status of the prediction. See prediction lifecycle for more details.metrics
includes metrics for the prediction, like the time it took to run.