Which AI tools are supported?

Compatible with FLUX.2, Midjourney V6, Google Gemini, NanoBanana, ComfyUI, Stable Diffusion XL, DALL-E 3, and more.

What does the negative prompt do?

Negative prompts tell the AI what to avoid. Use terms like "low quality, blurry" to improve output quality.

How to get better results?

Use descriptive adjectives, be specific about subject/style/lighting. Start with lower steps for quick tests.

Can I modify the prompt?

Absolutely! Mix and match keywords to create unique styles.

What if the result is not ideal?

Try adjusting guidance scale, changing aspect ratio, or iterating on keywords.

Sitting on a Modern Subway Bench - AI Generated Image

Reference Image

Text-to-Image

Pure Text Generation

This artwork was generated entirely by AI from text prompts, without any reference image

Sitting on a Modern Subway Bench

fashioninteriorminimalistnaturephotography

Prompt

Reference image: uploaded photo Face: keep the same face from the reference photo, calm and composed expression, eyes slightly hidden behind dark sunglasses. Preserve realistic proportions and lighting. Pose: seated pose on a modern subway bench, one arm resting on the leg, the other hanging down loosely. The body leans slightly forward with casual confidence, one shoulder lowered. Clothes: oversized houndstooth-patterned jacket over black turtleneck, wide dark jeans, chunky lace-up boots. Accessories: narrow black sunglasses, subtle rings, crossbody strap barely visible. Background: minimalistic urban metro interior with grey concrete walls, metallic accents, and modern geometric design. Lighting: diffused artificial indoor light, soft shadows, neutral color balance emphasizing grayscale tones. Atmosphere: moody street editorial — quiet confidence, understated power, cinematic stillness. Format: 2:3 Do not replace face; keep the same person with realistic proportions and lighting accuracy.