# Generating images (/docs/characters/images)
Once a character exists, you can drop them into any scene you can describe. The face comes from the character, so your prompt is only about the setting.
## Generate an image [#generate-an-image]
From the character's dashboard, click **Generate Image**.
1. **Choose an image model.** The picker shows the packs you have installed, then the models inside each one. Pick one, and SimpliGen remembers it for next time.
2. **Describe the scene.** Write where the character is and what is happening, for example "on a beach at sunset, warm golden light". You do not describe the character's face again. That comes from the character.
3. **Set the aspect ratio and how many to make**, then generate.
The results collect on the character's dashboard. The scene is whatever you asked for, but the face stays locked to the character.
## Describe the scene, not the person [#describe-the-scene-not-the-person]
This is the part that makes characters powerful. A normal prompt has to re-describe the person every time, and you get a different face every time. With a character, you only write the scene:
* Good: "sitting in a cafe by a rainy window, soft morning light"
* Good: "hiking a mountain trail, backpack, bright daylight"
* Skip: hair colour, eye colour, face shape. The character already holds those.
## Getting better results [#getting-better-results]
* **Be specific about setting, lighting, and pose.** Those are what your prompt controls.
* **Set the right [aspect ratio](/docs/generating/resolution-duration)** for where the image will be used.
* **Make a few at once** and pick the best.
* **Enhance the ones you love** for crisp, finished detail. See [Enhancing](/docs/characters/enhancing).
## Next steps [#next-steps]
Animate your character into a clip.
Upscale a result to a finished image.