Documentation: Generate Image with AI Node¶

Overview¶

The Generate Image with AI Node is a node from the AI Models family that creates an image from a text description (prompt) using an image-generation model via OpenRouter. The node's output is a URL where the generated image can be viewed, ready to be used in subsequent nodes.

In IoT and industrial environments it is useful for automatically generating visual material: illustrations for incident reports, signage or warning graphics, or supporting images for notifications, without depending on a designer.

When to use this node?¶

Use this node when you need to:

Generate a descriptive or illustrative image from text within a flow.
Create supporting visual material for reports, notifications, or dashboards.
Produce signage or graphics dynamically based on the context of an event.

Node Configuration¶

The form is organized into three selectable sections: Model, Prompt, and Advanced. It also has the JSON Editor tab.

Model section of the Generate Image with AI node

Section: Model¶

1. API Key *Required¶

The OpenRouter credential that authenticates calls to the model. It is managed in a centralized and secure way (the key is not written in the node, but referenced from the selected credential).

2. Model *Required¶

The image-generation model to use. Clicking the selector opens a search panel with available models (filtered to those that produce images), along with their provider, context, and price.

Section: Prompt¶

1. System Prompt¶

Optional system instructions for the model (define the role or general style).

2. Prompt *Required¶

The description of the image to be generated. The more specific, the better the result. The node's output will be a URL where the image can be viewed.

Configured Prompt section of the Generate Image with AI node

Section: Advanced¶

Temperature: controls creative variance (0 = deterministic, 2 = very creative). Leave empty to use the default value.
Max Tokens: maximum token budget for the model's response. Leave empty to use the default value.

JSON Editor View¶

JSON Editor view of the Generate Image with AI node

JSON Structure (Input Parameters)¶

{
  "api_key": "",
  "model_id": "",
  "prompt": "A high-contrast warning sign illustration for a restricted industrial area, red and yellow colors, no text.",
  "system_prompt": "You are an assistant that creates clear industrial safety visuals."
}

JSON Fields¶

Field	Type	Description
`api_key`	string	Reference to the OpenRouter credential (managed securely).
`model_id`	string	ID of the selected image-generation model.
`prompt`	string	Description of the image to generate.
`system_prompt`	string	(Optional) System instructions for the model.
`temperature`	number	(Advanced) Creative variance (0–2).
`max_tokens`	number	(Advanced) Maximum token budget.

Output: Where the node's data comes from¶

The node returns a URL where the generated image can be viewed. This output can be used in subsequent nodes with {{node_key}} (for example, to attach it to an email, display it in a notification, or save it).

Usage Examples¶

Example 1: Generate a warning sign for a restricted area¶

Use case: When unauthorized access is detected, a warning sign illustration is generated to include in the incident report or notification.

Prompt: A high-contrast warning sign illustration for a restricted industrial area, red and yellow colors, no text.
System Prompt: You are an assistant that creates clear industrial safety visuals.

(see JSON structure above)

Example 2: Supporting image for a report¶

Use case: Generate an illustrative image of the type of equipment involved in an incident to enrich an automatic report.

Prompt: A clean technical illustration of an industrial water pump, isometric view, neutral background.

Validation and Errors¶

Condition	Common cause / fix
`api_key is required`	The OpenRouter credential was not selected. Select or create a valid credential.
Image is not generated	Verify that the selected model supports image generation and that the credential has balance/permissions.
Unexpected result	Refine the prompt with more detail (style, colors, composition).

Best Practices¶

Use centralized credentials: do not write the API Key in the node; select an OpenRouter credential managed securely.
Specific prompts: describe style, colors, and composition to obtain consistent results.
Choose the right model: compare context and price in the model selector according to your needs.
Reuse the output URL: chain the output with notifications, emails, or reports to make use of the generated image.