# Convert slow XLMRobertaTokenizer to fast one
**URL:**
**Author:**  [@nielsr](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fnielsr)%5C
**Post date:** [July 26, 2022, 8:47am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconvert-slow-xlmrobertatokenizer-to-fast-one%2F20876%2F1 "2022-07-26T08:47:08Z")
Posting a question I had which @SaulLu answered 😃 suppose you have a repo on the hub that only has slow tokenizer files, and you want to be able to load a fast tokenizer. Here’s how to do that:
```auto
!pip install -q transformers sentencepiece
model_name = "naver-clova-ix/donut-base-finetuned-docvqa"
from transformers import XLMRobertaTokenizerFast
tokenizer = XLMRobertaTokenizerFast.from_pretrained(model_name, from_slow=True)
tokenizer.save_pretrained("fast_tok", legacy_format=False)
```
---
**Author:**  [@ahmedlone123](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fahmedlone123)%5C
**Post date:** [January 24, 2023, 12:20pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconvert-slow-xlmrobertatokenizer-to-fast-one%2F20876%2F2 "2023-01-24T12:20:45Z")
hey @nielsr@SaulLu , Thank you for this but what if you have fast tokenizers and you want to conver them to slow tokenizers? I am thinking that somehow you could parse the .json (when saving a fast tokenizer) and create the sentencpeice model you need for slow tokenizers from the json. what do you think?
---
**Author:**  [@abdul-r17](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fabdul-r17)%5C
**Post date:** [June 15, 2024, 10:56am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconvert-slow-xlmrobertatokenizer-to-fast-one%2F20876%2F3 "2024-06-15T10:56:47Z")
Hi @nielsr@ahmedlone123 , I want to convert a slow tokenizer (LlamaTokenizer) to a fast one (LlamaTokenizerFast), without losing the legacy behaviour i.e, it should not append a dummy prefix “\_” in front of the sentence. It works fine with LlamaTokenizer, but when I load the same tokenizer using LlamaTokenizerFast, this behaviour is lost. How to retain the default behaviour of LlamaTokenizer (not adding a dummy prefix) in fast tokenizer?
---
**Author:**  [@itazap](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fitazap)%5C
**Post date:** [August 26, 2024, 5:36pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fconvert-slow-xlmrobertatokenizer-to-fast-one%2F20876%2F4 "2024-08-26T17:36:34Z")
Hi @abdul-r17 . Both slow and fast Llama add a dummy prefix. This can be controlled with the `add_prefix_space=False` parameter which you can set on the model before you save or when loading the tokenizer:
```auto
tokenizer = LlamaTokenizer.from_pretrained(model, add_prefix_space=False)
tokenizer.save_pretrained(temp_folder)
LlamaTokenizerFast.from_pretrained(temp_folder)
```
or when loading:
```auto
tokenizer = LlamaTokenizer.from_pretrained(model)
tokenizer.save_pretrained(temp_folder)
LlamaTokenizerFast.from_pretrained(temp_folder, add_prefix_space=False)
```
Hope that works in your use case! 😄