DALL-E 2 AI Image Generation

Project Showcase
1

The power of AI. Dall-e can make pictures from descriptions. And does it really well. See for yourself:

CleanShot 2022-04-07 at 07.08.37@2x
CleanShot 2022-04-07 at 07.08.37@2x1318Ă—826 129 KB

Imagine how great this is going to be in a future version of Glide. Read more:

2

Update:

3

OpenAPI and their GPT-3 models are doing incredible things. Thanks for sharing!

4

Yeah, both text and images are great for dummy content.

5

Guess what. DALL-E 2 is available!

If you haven’t yet checked out OpenAI’s’s DALL-E 2, prepare to have your mind blown!:exploding_head:

No more endless Google searching for the perfect game art. Have AI create it for you in SECONDS!

Check out my demo here:

6

Did you see this? https://twitter.com/nptacek/status/1548402120075800577

7

Midjourney is nice too!
http://www.midjourney.com/

8

I really wish I could have used this. I applied and got through but then found out OpenAI does not support my country, and there’s no way I can work around it (VPN doesn’t work)

9

Hi, I don’t know how to use API, but is this OpenAI API a way to integrate Dall-E with Glide?

10

Oh…just wait. There will soon be a way I’m sure :wink:

11

All the more than Dall-E has an API …OpenAI API

12

This is cool! Let me put that in Glide, LOL… give me 10 minutes

Screen Shot 2023-02-15 at 6.57.28 PM
Screen Shot 2023-02-15 at 6.57.28 PM1728Ă—1854 264 KB

13

WOW! What column did you use to do that?

14

google script

15

Soon you won’t even need Google Script… just wait!

16

Hahaha, I have so much fun with this… the script is working well!

Screen Shot 2023-02-15 at 8.46.12 PM
Screen Shot 2023-02-15 at 8.46.12 PM1920Ă—1845 150 KB

You can try it here:

go to the Dream Catcher tab

17

Well done, but 2 questions :

  • Google script, hum, not easy to address for a non-coder (“10 mn”, I am jealous !)
  • Did you use the API DallE that I mentionned above? If yes, won’t you be quickly limited in terms of nb of images, cf. credits?
18

yes, they give the curl…

curl https://api.openai.com/v1/images/generations \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "prompt": "a white siamese cat",
    "n": 1,
    "size": "1024x1024"
  }'

all you need to do is adjust the payload, fetch the result, fish out the URL, and place it into the image component.
I did not find any limitations other than 10 images per call.

19

I fixed the Glide entry component. When you hit enter instead of the Go button, DALL-E will return an error, so all of you who did that have never seen the image… now it will work no matter what you enter :wink:

20

I love the “all you need” for a non coder :slight_smile: