if you’re an openai api customer, i’d love your feedback. what annoys you the most? what features do you want us to add? if you looked at using the api and didn’t, why not?

Aug 20, 2022 · 4:10 PM UTC

253
27
6
537
Load newest
Replying to @sama
While I understand that DALL-E is built for images, it would be nice if it understood letters and words better. love what you do! Charlie
2
Replying to @sama
I would like to improve your onboarding, starting from the email that I got the invitation and ending with my Trial Period is ending.
1
Replying to @sama
It repeats input text too much. Our customers frequently complain. Frequency/presence penalties don’t fix it.
2
1
9
Replying to @sama
Token generation limits for live apps
1
Replying to @sama
Cost was not competitive with midjourney
1
Replying to @sama
Unlimited storage for DALL-E images. I imagine the cost is trivial compared to compute, annoying not to be able to access old images.
1
1
Replying to @sama
Embedding endpoints are just prohibitively expensive for a lot of use cases I'm interested in
1
Replying to @sama
I’d love to see more editing tools in dalle lab. built in !Pad function so i don’t have to upload and download to discord, Style controls and some adjustable parameters for prompt accuracy vs creativity especially when using reference images got-3 prompt formula generator
1
1
Replying to @sama
I'm just a tester not a tech guy. It might be an amazing user experience if .audio or .mp3 or any other music file can be attached and Ai will interpret the song with several images. Or maybe, a voice command instead of typing a prompt.🤷‍♂️
1
Replying to @sama
In playground, copying and pasting sometimes doesn't work properly on the first try. Also is there an option to use saved prompts in the API?
3
Load more