Generate images from text descriptions using Stable Diffusion. Select up to 4 models to compare results side by side.
Pick 1-4 checkpoints. Each one has its own style — SDXL models are higher quality but need 1024px, SD 1.5 models are faster at 512px. Selecting multiple generates the same prompt with each for side-by-side comparison.
Describe what you want to see. Be specific — subject, style, lighting, colors, and quality words (e.g. "photorealistic", "cinematic", "8k detail") work better than vague descriptions. Comma-separated keywords are standard.