# DALL-E 3 Misspellings? What gives?

**URL:** <https://community.openai.com/t/dall-e-3-misspellings-what-gives/461366>\
**Category:** Bugs\
**Created:** [October 30, 2023, 11:58pm UTC](https://community.openai.com/t/dall-e-3-misspellings-what-gives/461366 "2023-10-30T23:58:21Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![brianward](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/brianward/32/229924_2.png) [@brianward](https://community.openai.com/u/brianward)\
**Post date:** [October 30, 2023, 11:58pm UTC](https://community.openai.com/t/dall-e-3-misspellings-what-gives/461366/1 "2023-10-30T23:58:21Z")

</div>

I’m using ChatGPT Plus and the beta version of DALL-E 3. It produces images quite well, but if I ask it to add text to the image if fails miserably, producing misspellings despite being given the correct spellings. Has anyone else experienced this?  
Thanks in advance!

---

<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:** [October 31, 2023, 1:32am UTC](https://community.openai.com/t/dall-e-3-misspellings-what-gives/461366/2 "2023-10-31T01:32:25Z")

</div>

Everyone has experienced this. It is 500% better at producing text than DALL-E 2:

 ![image](https://us1.discourse-cdn.com/openai1/original/3X/9/a/9a33e18c7f205f7e969d2f85f892c6f3609b8080.jpeg)
