SD Models

Stable Diffusion checkpoint models.

SD 1.x Models

Popular Checkpoints

Model	Description	Size
v1-5-pruned-emaonly	SD 1.5 base, recommended	2.5GB
v1-5-pruned	SD 1.5 full	4GB
v1-5	SD 1.5 with inference code	7GB

Fine-tuned Models

Model	Style	Description
dreamlike-photoreal-2.0	Photo	Realistic images
deliberate_v3	Versatile	High quality
realisticvision_v5.0	Photo	Photorealistic

SD 2.x Models

Model	Resolution	Description
768-v1.1	768x768	Higher resolution
512-base-ema	512x512	Faster

SDXL Models

Model	Description
sd-xlt-base-1.0	SDXL base
sd-xlt-refiner-1.0	SDXL refiner
sdxl-turbo	Fast generation

Download Locations

Civitai

• SD 1.5 Models
• SDXL Models

Hugging Face

• Stability AI
• CompVis

Installation

ComfyUI/models/checkpoints/
  v1-5-pruned-emaonly.safetensors
  realisticvision_v5.0.safetensors
  sd-xlt-base-1.0.safetensors

Model Selection Guide

Need	Recommended
General use	v1-5-pruned-emaonly
Photorealistic	realisticvision_v5.0
High quality	SDXL base
Fast generation	SDXL Turbo

Tips

1. Use safetensors: Safer and faster
2. Check size: Match to your VRAM
3. Refiner: Use for higher quality (SDXL)

Next Steps

• Flux Models - Flux models
• LoRA - Style models