# 400 error when trying to make predictions using HuggingFacePredictor

**URL:** <https://discuss.huggingface.co/t/400-error-when-trying-to-make-predictions-using-huggingfacepredictor/31651>\
**Category:** Beginners\
**Created:** [February 13, 2023, 6:58pm UTC](https://discuss.huggingface.co/t/400-error-when-trying-to-make-predictions-using-huggingfacepredictor/31651 "2023-02-13T18:58:52Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![peng0208](https://avatars.discourse-cdn.com/v4/letter/p/ee7513/32.png) [@peng0208](https://discuss.huggingface.co/u/peng0208)\
**Post date:** [February 13, 2023, 6:58pm UTC](https://discuss.huggingface.co/t/400-error-when-trying-to-make-predictions-using-huggingfacepredictor/31651/1 "2023-02-13T18:58:52Z")

</div>

![image](https://us1.discourse-cdn.com/hellohellohello/original/2X/a/a0d3e9c15759c0267b389404e5bd12b9c3be3144.jpeg)

using the sample code here to deploy a model and tried to make a prediction.But it returns the following error:  
An error occurred (ModelError) when calling the InvokeEndpoint operation: Received client error (400) from primary with message "{  
“code”: 400,  
“type”: “InternalServerException”,  
“message”: “T argument needs to be of type (SquadExample, dict)”  
}

Any suggestions?

Thank you.