Hi there, I'm Niels from the Hugging Face team.
This PR improves the model card for the BitDance tokenizer by:
- Adding the `image-feature-extraction` pipeline tag to the metadata for better discoverability.
- Adding a table detailing the three available tokenizer variants (`ae_d16c32`, `ae_d32c128`, `ae_d32c256`) and their respective configurations.
- Linking to the official Hugging Face paper page and GitHub repository.
Best,
Niels
