MLFlow supports models deployed with MLFlow.

HTTP API Input Schema Requirements’s model inference UI generates predictions by sending requests to the model with the application/json; format=pandas-split Content-Type. This is the default format used by MLFlow’s model server. Your model must support this format.

Here is an example of the expected input schema:

curl \
-H 'Content-Type: application/json; format=pandas-split' \
-d '{
    "columns": ["a", "b", "c"],
    "data": [[1, 2, 3], [4, 5, 6]]