Image

Configure image generation capabilities in IntraLLM AI. This page explains how image prompts, generation engines, workflows, and default image parameters are configured.

Image Settings

This page configures image generation capabilities in IntraLLM AI. Image generation allows models and workflows to create images from text prompts using a selected generation engine and workflow configuration.

Image Prompt Generation

Image prompt generation controls how text prompts are prepared and passed to the image generation engine. Prompts may originate from user input or be generated by models as part of a workflow.

Image Generation Engine

The Image Generation Engine can be selected from the available engines in the dropdown list.

ComfyUI

When ComfyUI is selected:

  • Image generation requests are sent to a connected ComfyUI service
  • The engine executes a predefined workflow to produce images

Engine Configuration

  • ComfyUI Base URL: specifies the address of the ComfyUI service
  • ComfyUI API Key: used to authenticate requests to the ComfyUI service

The selected engine must be reachable from the IntraLLM AI backend.

ComfyUI Workflow

Workflow Definition

The ComfyUI Workflow defines the image generation pipeline in JSON format. The workflow controls:

  • Model loading
  • Prompt encoding
  • Sampling configuration
  • Image decoding and output

You can:

  • Paste a workflow definition directly
  • Upload a workflow.json file exported from ComfyUI (API format)

Workflow Nodes

Workflow nodes map generation parameters to ComfyUI components.

Common node mappings include:

  • Prompt encoding nodes
  • Model loader nodes
  • Sampler configuration nodes
  • Latent image size nodes
  • Image decoding and output nodes

At least one prompt node ID is required for image generation to function.

Default Image Generation Settings

Default settings are used when a workflow or template does not explicitly override them.

  • Set Default Model: default diffusion model used for image generation
  • Set Image Size: default output resolution
  • Set Steps: default number of sampling steps

These values provide a baseline for image quality and performance.

How it works

When image generation is triggered:

  • An image prompt is prepared
  • The selected generation engine executes the configured workflow
  • Models and parameters defined in the workflow are applied
  • The generated image is returned and displayed in the platform

Usage considerations

  • Workflow configuration directly affects image quality and performance
  • Larger image sizes and higher step counts require more GPU resources
  • Ensure the selected engine and workflow are correctly configured before use

Quick checklist

  • Image generation engine selected
  • Engine connection configured correctly
  • Workflow loaded and validated
  • Prompt node IDs defined
  • Default model, image size, and steps configured
  • Sufficient GPU resources available