java.lang.IllegalArgumentException in SageMaker

I'm unable to invoke the my SageMaker endpoint. I'm seeing this error in the endpoint logs

java.lang.IllegalArgumentException: reasonPhrase contains one of the following prohibited characters: \r\n: tokenizers>=0.10.1,<0.11 is required for a normal functioning of this module, but found tokenizers==0.11.2.

Try: pip install transformers -U or pip install -e '.[dev]' if you're working with git

My Sagemaker endpoint is invoked through a lambda function. The code that calls the sagemaker endpoint is:

SM_ENDPOINT_NAME = "pytorch-inference-2021-xx-xx"
sm_runtime= boto3.client('runtime.sagemaker')
txt = "Canon SELPHY CP1300 Compact Photo Printer"
response = sm_runtime.invoke_endpoint(EndpointName=SM_ENDPOINT_NAME, ContentType='text/plain', Body=txt)

The response is supposed to contain a vector.

It's been working fine previously but I started seeing this exception today.

Is this a bug in SageMaker? If not, how do I fix it?