# How to create a half precision pretrained model

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-create-a-half-precision-pretrained-model%2F17904%3E%5C
**Category:** Beginners\
**Created:** [May 13, 2022, 5:53am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-create-a-half-precision-pretrained-model%2F17904 "2022-05-13T05:53:06Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Betacat](https://avatars.discourse-cdn.com/v4/letter/b/9f8e36/32.png) [@Betacat](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FBetacat)%5C
**Post date:** [May 13, 2022, 5:53am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-create-a-half-precision-pretrained-model%2F17904%2F1 "2022-05-13T05:53:06Z")

</div>

Hi, I am using roberta-base model to finetune my dataset. I want to create a fp16 parameter model. How could I do it?

```auto
from transformers import AutoModelForSequenceClassification
import torch.nn as nn
model = AutoModelForSequenceClassification.from_pretrained("roberta-base")

```