Image Generation
The image generation feature lets your agent create images from text descriptions right within the conversation.
The agent automatically detects whether a message is a request to generate an image and routes it to the selected image model. If the message isn't an image generation request, the agent simply replies with a regular text message.
Image generation works when interacting with the agent through the external chat or an embedded widget.
Pricing depends on the selected model and is calculated per token. Input tokens (the prompt) and output tokens (the generated image) are billed separately.
Available Models Copy link
When you enable image generation, you can choose from the following models:
-
Gemini 3.1 Flash Image Preview: A fast model for generating illustrations, sketches, and drafts. Best when speed matters more than top-tier quality.
-
Gemini 3 Pro Image Preview: An advanced model with better detail and quality. Suited for production use cases and final illustrations.
-
GPT Image 2: An advanced model that delivers high detail and closely follows instructions. Ideal for final illustrations and more complex visuals.
You can switch models at any time in the agent's settings.
Enabling and Disabling Image Generation Copy link
You can turn image generation on or off either while creating an agent or from an existing agent's settings.
When Creating an Agent Copy link
On the Pricing step, enable the Image Generation option and select your preferred model.
In an Existing Agent's Settings Copy link
-
Go to AI Services → Agents and click the agent you want to configure.
-
Open the Management tab.
-
Click Modify in the Image Generation block.
-
Toggle the option on or off.
-
If enabling it, select an image generation model from the dropdown.
-
Click Save.
