This message was deleted.

@Aaron Pham could you take a look?

import bentoml import pickle from transformers import DistilBertForSequenceClassification,DistilBertTokenizer tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased') #Loading model to compare the results model = pickle.load(open('model.pkl','rb')) print(model) #Save model saved_model=bentoml.transformers.save_model('ml_model',model) print(f"Model saved: {saved_model}") I am using this code to save my model.

what version of bentoml are you using?

1.0.15

PreTrained support starts from https://github.com/bentoml/BentoML/releases/tag/v1.0.17