Create Photorealistic Social Media Images with Flux Realism, LoRa Models & Airtable

Last edited 58 days ago

AI Image Generation Workflow for Social Media Content

Overview

This n8n workflow automates the creation of photorealistic AI-generated images for social media content. The workflow uses RunComfy (ComfyUI cloud service) combined with Airtable for data management to create high-quality images based on custom prompts and LoRa models.

Key Features

Automated Image Generation: Creates photorealistic images using Flux Realism model and custom LoRa models

Airtable Integration: Centrally manages content requests, model information, and image status

Cloud-based Processing: Utilizes RunComfy servers for powerful GPU processing without local hardware requirements

Status Tracking: Monitors generation process and automatically updates database entries

Telegram Notifications: Sends success notifications after image completion

Technical Workflow

  1. Server Initialization: Starts RunComfy server with configured specifications
  2. Data Retrieval: Fetches content requests from Airtable database
  3. Image Generation: Sends prompts to ComfyUI with Flux Realism + LoRa models
  4. Status Monitoring: Checks generation progress in 30-second intervals
  5. Download: Downloads completed images
  6. Database Update: Updates Airtable with image links and status
  7. Server Cleanup: Deletes RunComfy server for cost optimization

Prerequisites

  • RunComfy Membership with API access
  • Airtable Account with configured database
  • Telegram Bot for notifications
  • Flux Realism Workflow in RunComfy library
  • Uploaded LoRa Models in RunComfy

Airtable Schema

The database must contain these fields:

  • topic: Content description
  • pose_1: Detailed image prompt
  • LoRa Name Flux: LoRa model name
  • Model: Character name
  • pose_1_drive_fotolink: Link to generated image
  • Bilder erstellt: Generation status

Configuration Options

Image Resolution: Default 832x1216px (adjustable in ComfyUI parameters)

Generation Parameters: 35 steps, Euler sampler, Guidance 2.0

Server Size: "Large" for optimal performance (adjustable based on requirements)

Time Intervals: 30s status checks, 50s server initialization

This workflow is ideal for content creators who need regular, high-quality, character-consistent images for social media campaigns.

New to n8n?

Need help building new n8n workflows? Process automation for you or your company will save you time and money, and it's completely free!