OmniGen2 Multimodal Image Editing ComfyUI Workflow - Windows One Click Installer

OmniGen2 is now available and can run on low VRAM devices! I’ve created a one-click Windows installer for ComfyUI, plus a workflow using the Detail Daemon node for extra detail.

What is OmniGen2?
OmniGen2 is a new open-source AI model that can both understand images and generate high-quality images from text prompts. It supports text-to-image, image editing by instructions, and can combine elements from multiple images. It’s fast, efficient, and works great even on consumer GPUs.

Model Details

• Text Generation Model Size: ~3 billion parameters

• Image Generation Model Size: ~4 billion parameters

• Speed: Under 1 minute on RTX 4090; even faster on enterprise GPUs

System Requirements

• Nvidia RTX 30XX, 40XX, or 50XX series GPU (FP16 support required; GTX 10XX/20XX not tested)

• CUDA-compatible GPU with at least 6GB VRAM

• Windows OS

• At least 30GB free storage

What’s Included

• Portable ComfyUI Windows Installer: Pre-configured for OmniGen2

• Custom Workflow: Supports text-to-image generation

• Automatic Downloads: All required nodes and models install automatically

Preloaded Models

• omnigen2_fp16.safetensors (ComfyUI\models\diffusion_models)

• qwen_2.5_vl_fp16.safetensors (ComfyUI\models\text_encoders)

• ae.safetensors (ComfyUI\models\vae)

Usage Notes

• OmniGen2 works best with detailed, instruction-based prompts.

• For best likeness, experiment with the Image CFG value (try 2.0–4.0; adjust as needed).

Support and More Information

• Project GitHub: For technical details, updates, and documentation, visit the project’s GitHub repository.

• Community Support: For troubleshooting or to connect with other users, join the Discord server.