# Using GPT's BPE tokenizer for BERT?

**URL:** <https://discuss.huggingface.co/t/using-gpts-bpe-tokenizer-for-bert/9998>\
**Category:** 🤗Transformers\
**Created:** [September 14, 2021, 9:41pm UTC](https://discuss.huggingface.co/t/using-gpts-bpe-tokenizer-for-bert/9998 "2021-09-14T21:41:38Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![dchang](https://avatars.discourse-cdn.com/v4/letter/d/48db29/32.png) [@dchang](https://discuss.huggingface.co/u/dchang)\
**Post date:** [September 14, 2021, 9:41pm UTC](https://discuss.huggingface.co/t/using-gpts-bpe-tokenizer-for-bert/9998/1 "2021-09-14T21:41:38Z")

</div>

Hello,

I’m training a custom vocab to train a BERT from scratch, and I was wondering if it would make sense to train a GPT-style BPE tokenizer and use a BertModel.

Has anyone done this kind of training with mismatched tokenizer and model types?  
I’d appreciate any insights!