Skip to main content
Text to Image

Flux.1 Dev

Generate high-quality images using FLUX.1 [dev] 12B parameter flow transformer from Black Forest Labs

View details
Try it in Ampere

Inputs

Loading input fields...
Execution Steps

Loading workflow structure...

Loading curated examples...

Overview

Flux.1 Dev turns a written prompt into one to four generated images. Use it for campaign concepts, ad or social visuals, product storytelling, landing-page art, and mood-board exploration when you want FLUX text-to-image generation with preset or custom sizing, png/jpeg output, guidance and inference controls, safety checking, and seed support.

Use cases

  • Generate visual directions for a campaign, ad set, or social post.
  • Create product storytelling art from a clear description of the scene and style.
  • Explore several image options from one prompt before choosing a direction.
  • Use seed, guidance, and inference controls when repeatable exploration matters.

Input tips

  • Use a prompt of 3-5,000 characters with the subject, setting, style, composition, and intended use.
  • Choose an image_size preset or custom dimensions when the output canvas matters.
  • Set output_format to png or jpeg based on the handoff need.
  • Adjust guidance_scale from 1-20 and num_inference_steps from 1-50 when you need more control.
  • Set num_images from 1-4 when you want multiple options.
  • Keep the safety checker enabled and use seed for repeatable generation.

Expected output

The AI Tool returns the prompt used, generated images with URLs, width, height, optional content type, timing metadata, safety flags, seed, and cost metadata. The FLUX output view renders a single image or gallery and shows prompt, seed, image count, timings, any NSFW detection notice, and the Flux.1 Dev model label.

Caveats

  • Prompt wording strongly affects output quality and what appears in the image.
  • Custom sizing can alter framing, composition, edges, or fine details.
  • Higher guidance or inference values can change prompt adherence and visual texture.
  • Safety checking can flag or filter some images, but outputs still need human review.
  • Generated text, logos, product details, faces, and image edges should be reviewed before use.

Related AI Tools