Skip to main content
Image modelsBlack Forest Labs

FLUX.2 [klein] 4B: model overview

FLUX.2 [klein] 4B generates images from text and supports multi-reference editing. Black Forest Labs releases this four-billion-parameter model under Apache-2.0. It is distinct from klein 9B variants and FP8 quantizations.

External model reference page. This page does not confirm availability in Neiron.

Content updated:

Capabilities

Generate a new image from a written brief.

Edit a source image following an instruction.

Use multiple visual references.

Use cases

Use FLUX.2 [klein] 4B when a dedicated preparation or analysis step is needed in a media pipeline.Compare FLUX.2 [klein] 4B with related FLUX.2 models if the task allows a different input or output format.Before starting, confirm the input format and expected result (a structured operation result).

Reviewed facts

Developer
Black Forest Labs
Purpose
Image generation and editing
Input
Text, Images
Output
Images
Verified
2026-09-15

Benchmarks

No comparable benchmark is recorded in this card.

Sources

What to verify before use

The model may distort lettering or miss instructions; verify text and object shape separately.
An image is not evidence of a real event. Scores for other FLUX variants do not replace a separate 4B evaluation.
Weights license: Apache-2.0. Check the developer repository for the applicable terms.

Prompts

These are example briefs for your own evaluation, not test results. A reference page does not imply Neiron access to the model.

Prompt example

Product scene: a matte green bottle on light stone at the left, a soft shadow, and empty space on the right. No labels or lettering.

Prompt example

Replace the lamp photo’s background with a gray studio. Add no objects; the lamp shape will be compared with the source afterward.

Prompt example

Use the first image for the cup’s shape and the second for the palette. Create a new wooden-shelf scene without copying either source composition.

FAQ

What can FLUX.2 [klein] 4B be used for?

FLUX.2 [klein] 4B generates images from text and supports multi-reference editing. Black Forest Labs releases this four-billion-parameter model under Apache-2.0. It is distinct from klein 9B variants and FP8 quantizations.

How do I distinguish variants and capabilities?

Do not mix 4B with klein 9B: verify licensing and evaluations for the chosen model. Available weights do not imply free generation in a third-party service.

How should I compare outputs on my own task?

Use the same source material and prompt. Write down which facts or details must be preserved. The examples on this page are evaluation briefs, not results of a Neiron test.