If you rely on the Rust Tokenizer implementations (through bindings to [Tokenizers](https://github.com/huggingface/tokenizers)) and therefore always requires the tokenizer.json file.

Generated with:

```python
from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("aware-ai/whisper-tiny-german")
assert tokenizer.is_fast
tokenizer.save_pretrained("...")
```