Available now

FLUX.2 [klein] 4B Base: image editing

By Black Forest Labs

Black Forest Labs' FLUX.2 [klein] 4B Base, the small base model of the FLUX.2 family, creates images from text and edits up to four images you upload.

Model ID
Maker
Black Forest Labs

Input

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

Example prompt. Edit it or write your own.

If not provided, uses the input image size.

A maximum of 4 images are allowed.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB · 0/4

Additional settings

Negative prompt for classifier-free guidance. Describes what to avoid in the image.

Guidance scale for classifier-free guidance.

If not provided, a random seed will be used.

The number of inference steps to perform.

The acceleration level to use for image generation.

Price$0.02 / megapixel

You sign up first. Your settings are kept.

About this model

Overview

FLUX.2 [klein] 4B Base gives one to four images per run in square, portrait or landscape presets. It takes a negative prompt, a guidance scale and a step count from 4 to 50, so you can tune results by hand.

The edit mode takes one to four images and an instruction. If you leave the size empty, the result follows the size of your input.

What you can build

  • Drafts and variations while you shape a prompt
  • Product photos with a changed background or color
  • A scene built from objects in several photos
  • Simple labels and signs with short text

Prompting tips

  • Use the negative prompt for things that keep appearing, such as text, watermarks or extra objects.
  • Raise the step count for finer detail; lower it while you test ideas.
  • In the edit mode, describe the change and name what must stay the same.

What to check

Check exclusions against the result; a negative prompt does not guarantee an element is absent.

Inputs

  • Prompt (required): The prompt to edit the image.
  • Negative Prompt (optional): Negative prompt for classifier-free guidance. Describes what to avoid in the image.
  • Guidance Scale (optional): Guidance scale for classifier-free guidance. From 0 to 20.
  • Seed (optional): The seed to use for the generation. If not provided, a random seed will be used.
  • Number of Inference Steps (optional): The number of inference steps to perform. From 4 to 50.
  • Image Size (optional): The size of the generated image. If not provided, uses the input image size. Options: Square HD, Square, Portrait 4:3, Portrait 16:9, Landscape 4:3, Landscape 16:9.
  • Acceleration (optional): The acceleration level to use for image generation. Options: none, regular, high.
  • Output Format (optional): The format of the generated image. Options: jpeg, png, webp.
  • Images (required): The URLs of the images for editing. A maximum of 4 images are allowed.

Output and limits

  • Image sizes: 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, WEBP.
  • 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: $0.02 / megapixel