Generate Image
What it is
Section titled “What it is”Native tool that lets the agent generate images from a text prompt during conversation.
When to use
Section titled “When to use”- On-demand marketing creatives in chat.
- Visual mockups from a natural-language description.
- Assistants that produce illustrations, avatars, or example images.
- Any case where the agent needs to return an image, not just text.
Configuration
Section titled “Configuration”- In the agent’s Tools tab, click + Add Tool.
- Under NATIVE TOOLS, pick Generate Image.
- In the form, select the image model to be used.
The model list comes from the image models active for your organization (configured under Settings → AI Models). See Image Models to understand the cost of each one.
Without a model selected, the tool stays not configured (yellow) and the agent can’t call it.
How the agent invokes
Section titled “How the agent invokes”The agent decides to call this tool when the user explicitly asks for an image (“draw a logo”, “generate a cover for my post”, “create an illustration of X”) or when the agent prompt instructs it to always accompany responses with visuals.
The generated image is attached to the agent’s reply automatically.
Limitations and notes
Section titled “Limitations and notes”- Model must be active in the org. Without an active image model in Settings → AI Models, the tool stays not configured and the agent can’t call it.
- Extra cost. Each call consumes credits per the chosen image model price (usually more than a text call). The cost lands on the org invoice along with normal agent usage. See Image Models for the full table.
- Latency. Image generation takes seconds to several seconds, depending on the model. The agent sends the image when ready.
- Quality depends on the prompt. The model receives the text the agent passes — the more specific, the better. Guide this in the agent prompt.