# Equivalent of \`inputs\_embeds\` for \`FlaxGPT2Model\`

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fequivalent-of-inputs-embeds-for-flaxgpt2model%2F9175%3E%5C
**Category:** 🤗Transformers\
**Created:** [August 12, 2021, 4:16am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fequivalent-of-inputs-embeds-for-flaxgpt2model%2F9175 "2021-08-12T04:16:38Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ethanabrook](https://avatars.discourse-cdn.com/v4/letter/e/9de053/32.png) [@ethanabrook](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fethanabrook)%5C
**Post date:** [August 12, 2021, 4:16am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fequivalent-of-inputs-embeds-for-flaxgpt2model%2F9175%2F1 "2021-08-12T04:16:38Z")

</div>

Hi,  
The `forward` method of `GPT2Model` allows the user to pass the embedded representation of tokens directly to GPT2 using the `inputs_embeds` keyword. However I notice that the ` __call__ ` method of `FlaxGPT2Model` does not seem to accept this keyword. Is there a workaround to achieve the same functionality? Thanks.