pipeline creation for onnx model

[OriAlpha]

I have torch model used for nlp, which is converted to onnx. I am using updated bentoml version v1.0.13.
I saw a way to pass tokenizer for transformers but this doesnt work for onnx model.
So is there any way to create pipeline for onnx so that it carried tokenizer data for inference ??

tokenizer = AutoTokenizer.from_pretrained("distilgpt2")
model = AutoModelForCausalLM.from_pretrained("distilgpt2")
generator = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
bento_model = bentoml.transformers.save_model("text-generation-pipeline", generator)

I just noticied in onnx module to specfify external_modules, but any example to define it ??

external_modules (:code:`List[ModuleType]`, `optional`, default to :code:`None`):
user-defined additional python modules to be saved alongside the model or custom objects,
e.g. a tokenizer module, preprocessor module, model configuration module