Remove processor class from tokenizer_config.json

#7
by Xenova HF Staff - opened
Google org
edited Mar 12

A bit of a nit, but this just means that when loading, you can use AutoProcessor or AutoTokenizer

This fixes issue https://github.com/vllm-project/vllm/issues/19575. Currently calling AutoProcessor for this model fails, which slows down vLLM

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment