# Incomplete Output of 4o through API

**URL:** <https://community.openai.com/t/incomplete-output-of-4o-through-api/764250>\
**Category:** API\
**Tags:** gpt-4o\
**Created:** [May 19, 2024, 7:32am UTC](https://community.openai.com/t/incomplete-output-of-4o-through-api/764250 "2024-05-19T07:32:23Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![\_j](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/_j/32/766292_2.png) [@\_j](https://community.openai.com/u/_j)\
**Post date:** [May 19, 2024, 8:02am UTC](https://community.openai.com/t/incomplete-output-of-4o-through-api/764250/2 "2024-05-19T08:02:39Z")

</div>

![image](https://us1.discourse-cdn.com/openai1/original/4X/0/9/d/09d9c39e3bbbf1fe5cdfa8660f4f83f79d1a17a5.jpeg)

 ![image](https://us1.discourse-cdn.com/openai1/original/4X/b/4/a/b4a3d962dc514edb4a93e68893800ef1057bea82.png)

First, you can press the ‘view code’ button to ensure you have the correct parameters. Then despite those, crank the max\_tokens value you can receive as output way up. You can see my truncation at 128 tokens.

Then look at the finish\_reason in the response. If it is “stop”, the AI just decided to stop writing.

Another oddity about this model is that specifying temperature and top\_p values too low is not just a setting to zero, it a setting to “useless”. The parsing of negative exponents is screwed up, perhaps, or math is done with reduced bits that overflow. So a setting like 0.0001 that still allows variance is needed to have moderately replicable outputs.

---

_[View the full topic](https://community.openai.com/t/incomplete-output-of-4o-through-api/764250)._