# Updating model and tokenizers inside Trainer.train

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fupdating-model-and-tokenizers-inside-trainer-train%2F103667%3E%5C
**Category:** Models\
**Created:** [August 23, 2024, 8:25am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fupdating-model-and-tokenizers-inside-trainer-train%2F103667 "2024-08-23T08:25:19Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![gb0820](https://avatars.discourse-cdn.com/v4/letter/g/7feea3/32.png) [@gb0820](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fgb0820)%5C
**Post date:** [August 23, 2024, 8:25am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fupdating-model-and-tokenizers-inside-trainer-train%2F103667%2F1 "2024-08-23T08:25:19Z")

</div>

Hi,  
I am working on adapting vocabulary for LLMs. I want to know is there any way to update model and tokenizer inside the trainer.train function or do I need to write the standard pytorch train loop to fine-tune?

PS. This needs to be done along with LoRA as we are dealing with fine-tuning LLMs.  
Planning to do something similar to what is described in the [work](https://link.springer.com/article/10.1007/s10590-019-09243-8)