Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

Can you pass pedantic models between flows?

Furthermore, if you can, can you pass models that allow abstract types

hey <@U0310QYGU2F> - yes, locally you can pass around pretty much anything that python functions can return / accept
```In [14]: class Anything(BaseModel):
    ...:     foo: str
    ...:

In [15]: @flow
    ...: def emits_anything() -&gt; Anything:
    ...:     return Anything(foo="something")
    ...:

In [16]: @flow(log_prints=True)
    ...: def accepts_anything(anything: Anything):
    ...:     print(anything)
    ...:

In [17]: anything = emits_anything()
15:12:33.619 | INFO    | prefect.engine - Created flow run 'quartz-iguana' for flow 'emits-anything'
15:12:34.441 | INFO    | Flow run 'quartz-iguana' - Finished in state Completed()

In [18]: accepts_anything(anything)
15:12:44.867 | INFO    | prefect.engine - Created flow run 'magnetic-copperhead' for flow 'accepts-anything'
15:12:45.372 | INFO    | Flow run 'magnetic-copperhead' - foo='something'
15:12:45.495 | INFO    | Flow run 'magnetic-copperhead' - Finished in state Completed()```


but if you're passing in flow run params in network calls, then you'll have to pass a JSON-serializable representation of those models, which the flow run will coerce into the expected type according to the flow run type annotations

Got it. So as long as its json serializable, its prefect cloud safe? Awesome!