Stable Diffusion Prompt Generator

Stable Diffusion responds to weighted keywords, a negative prompt and sampler settings. The generator emphasises your subject with (subject:1.2) and suggests an SDXL-friendly resolution for the aspect ratio.

Every prompt is built from the same modular parts — role, task, context (your audience and details) and output format — then rendered in the style each AI follows best. Change the AI in the drop-down and the same request is rewritten instantly.

• Weight important terms with (term:1.2).
• Always fill in the negative prompt.
• Start around 25–30 steps and CFG 5–7, then adjust.

Choose a prompt type for your task; each comes with an expert role, context rules and an output format:

• Photorealistic image — for example “an elderly fisherman mending nets on a wooden pier”.
• Illustration / concept art — for example “a floating city above the clouds at sunset”.

Example output for Stable Diffusion:

Prompt: (an elderly fisherman mending nets on a wooden pier:1.2), a natural environment that suits the subject, photorealistic, high detail, shot on a 50mm lens, soft natural light, rule of thirds, shallow depth of field, authentic and calm, early-morning mist, muted blue tones, highly detailed, sharp focus

Negative prompt: blur, distortion, extra fingers, text, watermark, lowres, blurry, jpeg artifacts, bad anatomy, watermark

Settings: 1216×832 (3:2), Steps 30, CFG 6, Sampler DPM++ 2M Karras