Text to Image

text2image

AI rendering features can be disabled per source in the Imgix dashboard. See Advanced Settings → Render Settings for details.

URL Signing Required

Use of this feature requires URL signing to prevent unauthorized usage.

Enabling Capability

This parameter can be disabled/enabled in the Advanced Settings of your Source.

All rendering capabilities are enabled by default.

The text2image parameter uses the Generative AI Endpoint (/ai?text2image=true&prompt=your_prompt) to create AI-rendered images using a text based prompt.

text2image accepts a string prompt as a suggestion for the type of image to generate using our AI model. The generated image can then leverage all the parameters under the Rendering API to transform or resize the image.

https://assets3.imgix.net/ai?text2image=true&seed=170221&prompt=A%20sunset%20over%20an%20ocean%20as%20a%20graphic%20for%20a%20t-shirt&auto=format%2Ccompress&fit=crop&w=1946&h=1900&s=8d9d6ab3818b79a38e30becf2c537a3c
sandbox-demo
https://assets3.imgix.net/ai?text2image=true&seed=16001&prompt=a%20moon%20over%20a%20mountain%20as%20a%20graphic&auto=format%2Ccompress&fit=crop&w=1946&h=1900&s=c5f218fac42a1b1817e0e08768f5f0ad
sandbox-demo

Limits

Guarantee of keyword output

While text2image suggests what the output should look like, it does not guarantee that the prompt keywords will be used in the final fill.

Character limits

There is no direct character limit for prompt values, though we recommend keeping your image URLs under 1024 characters.