Inconsistency between SDK and API

[mbrutrifork]

I have previously created predictions using the Python SDK like this:

prediction = PredictionValue(
    model_version="test", score=0.99, result=[result]
)
client.predictions.create(task=task.id, **prediction.model_dump())

I am now trying to create annotations through the Python SDK like this:

annotation = AnnotationValue(result=[result])
client.annotations.create(id=task.id, **annotation.model_dump())

Unfortunately, AnnotationValue requires the field completed_by, not required by the API, which makes me get this error:

pydantic_core._pydantic_core.ValidationError: 1 validation error for AnnotationValue
completed_by

and AnnotationValue includes fields like result_count which is not expected from the API, which makes me get this error:

TypeError: AnnotationsClient.create() got an unexpected keyword argument 'result_count'

Initially, I thought this could be an issue with missmatch of versions, since I was using label-studio-sdk==1.0.8 and label-studio:1.15.0 (docker image). I saw in the release notes that the last time the SDK version was bumped was for release 1.13.1 to SDK version 1.05, so I changed to label-studio-sdk==1.0.5, but it still doesn't work.

Questions

Are PredictionValue and AnnotationValue supposed to be used not in this context? Did I do something wrong?