Craiyon, formerly DALL·E mini, is an AI model that can draw images from any text prompt!
-
-
Are you related to DALL·E mini?
-
Yes, Boris Dayma (who trained the current version of the AI model) and Pedro Cuenca (who worked on the backend) are both part of the craiyon team. You can find more details in the DALL·E mini model card!
-
-
How do you keep it free?
-
The model requires a lot of compute so we rely on ads and donations to pay for our servers.
-
-
Do you have any tips to create better images?
-
It’s always a good idea to be specific. Here are a few keywords that can be interesting to experiment with: “illustration”, “photorealistic”, “high definition”… We’ve seen so many cool tricks from the community so you should definitely check what others do for inspiration!
-
-
What to do when there is too much traffic?
-
We are increasing the number of servers as fast as we can so anybody can generate images with no issue. In the meantime, you may need to try a few times to generate images.
-
-
How does the AI model work?
-
The model used is called “DALLE mini”, specifically the larger version also known as “DALLE mega” and is trained using Google TRC. You can find more details in the W&B Project Report and the DALL·E mini model card. You can also watch The Story Behind DALL·E mini on Gradient Dissent with Boris Dayma.
-
-
What about limitations and biases?
-
While the capabilities of image generation models are impressive, they may also reinforce or exacerbate societal biases. Because the model was trained on unfiltered data from the Internet, it may generate images that contain harmful stereotypes. The extent and nature of the biases of the DALL·E mini model have yet to be fully documented. Work to analyze the nature and extent of these limitations is ongoing and being documented in more detail in the DALL·E mini model card.
-
-
Why are faces so weird?
-
It’s a limitation of our current image encoder. We are working on developing a better solution.
-
-
Can you make larger images?
-
Not at the moment but we plan to make it possible in the future.
-
-
Where does the logo come from?
-
We use the 🖍️ crayon emoji, which is displayed natively (inspired by 🤗 Hugging Face). Images of the emoji use Twemoji – Twitter’s open-source emoji collection (License: CC BY-SA 4.0). We just changed the color to orange.
-
-
Can I use the images generated through craiyon?
-
Yes, as long as you respect the Terms of Use, feel free to use them as you wish for personal use, whether you want to share them with your friends or print on a T-shirt.
Please credit craiyon.com for the images.
For commercial use, please refer to the “Commercial Licenses” section of the Terms of Use.
-
-
Do you have an app?
-
Yes, we have an app for Android devices on Google Play Store.
WARNING: USE THE LINK ABOVE AS THERE ARE MANY COPYCATS.
There is currently no app available for iPhone as it requires more development. For Androids, we were able to offer a simple PWA which replicates the website and only caches files for a faster experience.
