# Default gpt-j output length

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdefault-gpt-j-output-length%2F17113%3E%5C
**Category:** Beginners\
**Created:** [April 23, 2022, 4:11pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdefault-gpt-j-output-length%2F17113 "2022-04-23T16:11:28Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![DLiebman](https://avatars.discourse-cdn.com/v4/letter/d/e9c0ed/32.png) [@DLiebman](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FDLiebman)%5C
**Post date:** [April 23, 2022, 4:11pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdefault-gpt-j-output-length%2F17113%2F1 "2022-04-23T16:11:28Z")

</div>

I am writing a app that uses gpt-j sized models. I have an api key from huggingface, but when I use the api I get only a single word as a response.

I believe the model hosted on huggingface only allows for a limited number of total output tokens. My app uses many tokens as part of the query. How might I use more tokens? Is there an environment variable that I can set for making a longer query?