- Text, up to 4 images
Inputs
- 0.25 to 4 megapixels
Output size
- Guidance, steps, negative prompt, seed
Controls
- PNG, JPG, WebP
Formats
About FLUX.2 Klein 4B Base
FLUX.2 [klein] 4B Base is the smaller of the two Klein base models from Black Forest Labs, which Masonry runs through fal. It is an un-distilled base model, so you control guidance and inference steps yourself instead of getting one fixed recipe. Black Forest Labs positions the 4B size for fine-tuning and lighter deployments; in Masonry you use the hosted base model as is.
In Masonry it makes images from a prompt, and edits or combines up to 4 input images in the same model. You choose an output size from 0.25 to 4 megapixels or one of 10 preset ratios, set guidance scale, inference steps, a negative prompt, and a seed, turn on the fast path for quicker drafts, and export PNG, JPG, or WebP. It takes the same inputs as FLUX.2 Klein 9B Base and is billed at a lower rate.
It suits quick, repeatable test rounds where you change one setting at a time, and high-volume drafts where cost per image matters more than the last bit of detail. Move to Klein 9B Base or FLUX 3 Image for finals.
Why teams choose FLUX.2 Klein 4B Base
Choose FLUX.2 Klein 4B Base when you want the Klein controls (guidance, steps, negative prompt, seed) at a lower cost per image, for drafts and test rounds. For finals, step up to Klein 9B Base or FLUX 3 Image, and if you would rather not tune settings, try GPT Image 2.5 or Nano Banana 2.
What FLUX.2 Klein 4B Base can do
The capabilities that set FLUX.2 Klein 4B Base apart and earn its place in a brief
Smaller, Lower-Cost Klein
The 4B model takes the same inputs as Klein 9B Base and costs fewer credits per image, which fits drafts and large test batches.
Guidance and Steps You Set
As an un-distilled base model it uses classifier-free guidance. Set guidance scale and inference steps to trade speed for detail.
Edits From Up to 4 Images
Add up to 4 input images to edit a photo or combine a product with a scene, and describe the change in the prompt.
0.25 to 4 Megapixels
Pick 0.25, 0.5, 1, 2, or 4 megapixels, or one of 10 preset ratios from square to an ultra-wide banner.
Negative Prompt and Seed
List what to avoid, and set a seed so a rerun changes only the setting you are testing.
Fast Path for Drafts
Turn on the fast path for quicker drafts, then turn it off for the higher-quality pass.
Where teams reach for FLUX.2 Klein 4B Base
- Low-cost concept and layout drafts
- Repeatable setting tests with a fixed seed
- Quick product-in-scene edits from up to 4 images
- Small social images at 0.25 to 1 megapixel
- Comparing against Klein 9B Base before a final run
Frequently asked questions
What teams need to know about creating with FLUX.2 Klein 4B Base in Masonry
What is FLUX.2 Klein 4B Base?
FLUX.2 [klein] 4B Base is the smaller un-distilled base model in Black Forest Labs' FLUX.2 [klein] family. Masonry runs it through fal.
How is it different from FLUX.2 Klein 9B Base?
Both take the same inputs and controls in Masonry. 4B is the smaller model and costs fewer credits per image; 9B is the larger sibling to compare for finals.
Can I fine-tune FLUX.2 Klein 4B in Masonry?
No. Black Forest Labs positions the 4B size for fine-tuning, but Masonry runs the hosted base model as is.
Can it edit images?
Yes. Add up to 4 input images and describe the change, such as placing a product into a scene.
What sizes does it output?
0.25, 0.5, 1, 2, or 4 megapixels, or one of 10 preset ratios from square to an ultra-wide banner.
Can I run FLUX.2 Klein 4B Base from the command line?
Yes. Install the Masonry CLI with npm install -g @masonryai/cli, sign in with masonry login, then run masonry image with your prompt and --model flux-2-klein-4b-base.
How does my team use FLUX.2 Klein 4B Base in Masonry?
Open a Masonry canvas, pick FLUX.2 Klein 4B Base from the model selector, and describe the image you need: a product shot, an ad creative, a social post. Masonry generates it, then you refine, edit, and combine FLUX.2 Klein 4B Base with other models in one workspace.
Is FLUX.2 Klein 4B Base free to try?
New Masonry accounts get 50 free credits to try GPT Image 2.5 Flare. FLUX.2 Klein 4B Base is included in every paid plan, and the editor shows the exact credit cost before you generate.
How do I write good prompts for FLUX.2 Klein 4B Base?
Fix a seed first, then change one setting at a time (guidance, steps, or the negative prompt) so you can see what each does. Draft at 0.5 or 1 megapixel with the fast path on, then rerun the winner at a larger size. See the prompt gallery on this page for real FLUX.2 Klein 4B Base prompts you can copy and adapt.
Who makes FLUX.2 Klein 4B Base?
FLUX.2 Klein 4B Base is built by Black Forest Labs. Inside Masonry it runs alongside 50+ image and video models, so your team can pick the right one for each brief without switching tools.
Can I see examples made with FLUX.2 Klein 4B Base?
Yes, the prompt gallery on this page shows real images teams have generated with FLUX.2 Klein 4B Base in Masonry, each paired with the exact prompt you can copy and adapt for your own brand.
Start creating with FLUX.2 Klein 4B Base
Generate, edit, and compare across 50+ models in one workspace.
Guides for FLUX.2 Klein 4B Base
Prompt walkthroughs and examples from the Masonry blog
Explore more AI models
Compare FLUX.2 Klein 4B Base with other models teams run in Masonry
