Black Forest Labs' FLUX.2 variant that exposes step count and guidance scale, tuned for text rendering and fine detail. It sits between Pro and Max in price and supports text-to-image plus multi-reference editing.
Pick it if
you want to set steps and guidance yourself
legible text and small details matter more than latency
you edit from several reference images in one call
Skip it if
default settings suffice: FLUX.2 Pro costs less per image
you need open weights: Klein and Dev are the open FLUX.2 models
FLUX.2 [klein] is Black Forest Labs' fastest FLUX.2 tier: 4-step distilled open-weight models in 4B and 9B sizes, which BFL says generate in under a second on consumer GPUs. It is the cheapest FLUX.2 face here, below Pro, Flex and Max.
Pick it if
you want the lowest-cost FLUX.2 image in the lineup
you generate high volumes; BFL positions klein for real-time use
you want a model whose weights you can also run yourself
Skip it if
you need maximum FLUX.2 fidelity; FLUX.2 Max targets that
you need fine control over steps and guidance; Flex exposes those
FAQ
Which is cheaper, FLUX.2 Flex or FLUX.2 Klein?
FLUX.2 Klein is cheaper — $0.03 / image vs $0.1 / image.
How do I use FLUX.2 Flex or FLUX.2 Klein?
In Octomind Cloud on a paid plan, ask the agent to generate with either model by name; it runs the job and returns the file.
Try either model
Switch between models inside the same Octomind session.