Write it. See it.
Text-to-image generation turns natural language descriptions into images. Describe the subject, style, lighting, and composition — the AI handles the rendering.
How it works
Prompt to image in seconds
- 01
1 · Write a detailed prompt
Describe the subject, setting, lighting, camera angle, and style. Example: "A product shot of a matte-black smartwatch on a marble surface, studio softbox lighting, shallow depth of field."
- 02
2 · Choose model and settings
Select Flux-2 Pro (photorealism), Imagen 4 Ultra (complex compositions), or Nano Banana (fast iteration). Pick aspect ratio and resolution.
- 03
3 · Download your image
The model generates your image in seconds. Download it, iterate on the prompt, or try a different model for comparison.
Prompt tips
Write prompts that get results
The quality of your output depends on the specificity of your prompt.
Be specific about the subject
"A cat" is vague. "A tabby cat sitting on a windowsill, afternoon light, soft focus background" gives the model enough to work with.
Include style and medium cues
Add terms like "oil painting", "35mm film", "flat vector illustration", or "anime cel shading" to steer the visual style.
Describe lighting and camera
"Golden hour", "studio softbox", "low angle", "shallow depth of field" — these details significantly affect the output.
Use negative prompts (Flux-2 Pro)
If you're using Flux-2 Pro, add negative prompts to exclude unwanted elements: "no text, no watermark, no blurry."
FAQ
Text to Image FAQ
Turn your first prompt into an image
Sign in, pick a model, and generate. 30 free credits, no card required.