Local Image Generation
Run Stable Diffusion locally with ComfyUI — generate unlimited images without cloud API costs.
Setup time
30–60 minutes
Min hardware
6 GB VRAM (RTX 3060 / RX 6600 XT)
Software
ComfyUI + Stable Diffusion
Recommended model
Stable Diffusion XL
Install ComfyUI
Clone ComfyUI from GitHub: git clone https://github.com/comfyanonymous/ComfyUI. Install dependencies: pip install -r requirements.txt.
Use Python 3.10+ and ensure CUDA drivers are up to date for NVIDIA GPUs
Download a base model
Get Stable Diffusion XL from HuggingFace. Place the .safetensors file in ComfyUI/models/checkpoints/.
For 6 GB VRAM, use SD 1.5 (~2 GB) instead of SDXL (~6.5 GB)
Start ComfyUI
Run: python main.py --listen. Open localhost:8188 in your browser.
Generate your first image
Load the default workflow in ComfyUI, enter a prompt in the positive text node, and click Queue Prompt. Your first image will generate in 10–60 seconds.
Explore LoRA models
Extend SDXL with style LoRAs from Civitai. Place .safetensors in ComfyUI/models/loras/.
Affiliate disclosure: links to Amazon may earn us a commission at no extra cost to you.
Compatible GPUs for this workflow
For smooth local AI inference with this workflow, these NVIDIA RTX GPUs deliver the best experience.
Prices and availability may change.