# Long summarization

**URL:** <https://discuss.huggingface.co/t/long-summarization/21445>\
**Category:** 🤗Transformers\
**Created:** [August 9, 2022, 9:15pm UTC](https://discuss.huggingface.co/t/long-summarization/21445 "2022-08-09T21:15:31Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Samuel-Fipps](https://avatars.discourse-cdn.com/v4/letter/s/e56c9b/32.png) [@Samuel-Fipps](https://discuss.huggingface.co/u/Samuel-Fipps)\
**Post date:** [August 9, 2022, 9:15pm UTC](https://discuss.huggingface.co/t/long-summarization/21445/1 "2022-08-09T21:15:31Z")

</div>

Hello I was trying to use the t5 to summarize text that’s around 10,000 characters, and return about 20% of the original length. However it seems to only do about 1 sentences and the rest is either blank spaces or random words depending on what the no\_repeat\_ngram\_size is set too. Also if I set num\_return\_sequences to something more then 1, it will return “and…” as a sequence.

Should I be using a different model?