Available now

Sana 1.0: text to image

By NVIDIA

Sana 1.0 makes large images from a written prompt, at 3840x2160 by default, with nine style presets to choose from.

Model ID
Maker
NVIDIA

Input

Fields marked * are required. Review the price before generating.

Example prompt. Edit it or write your own.

Additional settings

The number of inference steps to perform.

The same seed and the same prompt given to the same version of the model will output the same image every time.

Price$0.01 / megapixel

You sign up first. Your settings are kept.

About this model

Overview

Sana 1.0 renders one to four images per run, by default at 3840x2160 or in six smaller square, portrait or landscape sizes, as JPEG or PNG. Steps run from 1 to 50 with 18 by default.

Choose a style preset such as Cinematic, Photographic, Manga, Neonpunk or 3D Model, or leave it on no style, and use the negative prompt for details you want left out.

What you can build

  • Desktop and TV wallpapers
  • Large backgrounds for slides and websites
  • Neonpunk, manga or 3D-style scenes
  • Many quick variations at small sizes

Prompting tips

  • Use a smaller preset size while you test a prompt, then switch back to 3840x2160.
  • Let the style preset carry the look and keep the prompt about the scene.
  • Raise the guidance scale when the image drifts from the prompt.

What to check

Check the main subject, repeated details and requested text at full size.

Inputs

  • Prompt (required): The prompt to generate an image from.
  • Negative Prompt (optional): The negative prompt to use. Use it to address details that you don't want in the image. This could be colors, objects, scenery and even the small details (e.g. moustache, blurry, low resolution).
  • Image Size (optional): The size of the generated image. Options: 3840 × 2160, Square HD, Square, Portrait 4:3, Portrait 16:9, Landscape 4:3, Landscape 16:9.
  • Number of Inference Steps (optional): The number of inference steps to perform. From 1 to 50.
  • Seed (optional): The same seed and the same prompt given to the same version of the model will output the same image every time.
  • Guidance scale (CFG) (optional): The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you. From 0 to 20.
  • Number of Images (optional): The number of images to generate. From 1 to 4.
  • Output Format (optional): The format of the generated image. Options: jpeg, png.
  • Style Name (optional): The style to generate the image in. Options: (No style), Cinematic, Photographic, Anime, Manga, Digital Art, Pixel art, Fantasy art, Neonpunk, 3D Model.

Output and limits

  • Image sizes: 3840 × 2160, Square HD (1024×1024), Square (512×512), Portrait 4:3 (768×1024), Portrait 16:9 (576×1024), Landscape 4:3 (1024×768), Landscape 16:9 (1024×576).
  • File formats: JPEG, PNG.
  • Up to 4 images per run.
  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • 1 megapixel: $0.01 / megapixel