Docs
Image generation
Create images from a text prompt with a choice of models, styles and templates.
Generate an image
- Open Image in the sidebar.
- Describe what you want to see. You can also start from one of the templates.
- Choose an image model and, if you like, a style.
- Generate. If the result is not quite right, adjust the prompt and try again.
You can also ask for an image in any chat, for example "Draw a watercolor of a lighthouse at dawn."
Write a good prompt
A useful prompt covers four things:
- Subject: what is in the picture. "A red fox sitting on a mossy log"
- Setting: where and when. "in a misty forest at sunrise"
- Style: the look you want. "soft watercolor illustration"
- Composition: framing and detail. "close-up, shallow depth of field"
Put together: A red fox sitting on a mossy log in a misty forest at sunrise, soft watercolor illustration, close-up.
Start simple and add detail one step at a time, so you can see what each change does.
Models and styles
Mochii offers several image models. Some are faster, others produce more detail or handle a particular look better. Styles apply a consistent visual treatment, such as photography or anime, without you having to describe it in the prompt.
Character avatars
The character editor uses the same image generation to create avatars. See AI Characters.
Credits
Image generation uses credits from your plan, like chat does. The cost depends on the image model. See Plans and credits.