Use Case: Marketing Image Generation at Scale

The Problem

Your marketing team needs dozens of image variations for campaigns — different sizes, styles, and messaging. Creating each manually is slow and expensive.

The Solution

Marketing brief (text)
  → AI generates image variations
    → Designer reviews and selects
      → Approved images → exported for campaigns

How to Build It

Step 1: Browse Content Generation

Go to /content-gen to see available generation workflows:

Step 2: Choose Your Generation Path

Need	Path	How
Quick social media images	AI Provider	OpenAI DALL-E or Cloudflare
High-quality product shots	MIFY Queue	ComfyUI SDXL/Flux workflows
Custom model styles	ComfyUI Local	Your own models + LoRA

Step 3: Build a Batch Workflow

For scale, create a workflow that generates multiple variations:

CSV with product names
  → Loop Map (iterate each product)
    → Chat Node (generate image prompt from product name)
      → Content Gen Queue (generate image)
        → Collect results

Step 4: Review and Approve

Use a Pause node for human review:

• Designer sees generated images in the Inbox
• Approve good ones, reject and regenerate others
• Approved images available in the run history

Step 5: Train Custom Models (Advanced)

For brand-consistent styles:

1. Upload training images to a dataset at /training/new
2. Select a training preset (LoRA fine-tuning)
3. Train a custom model
4. Use the trained model in future generation workflows

Results

• 10x faster than manual design for variations
• Consistent brand style with custom-trained models
• Human quality control via approval gates
• Scale — generate hundreds of variations from a single brief