MiniMax's first text-to-image model, available on its api platform since February 2025. it renders one to nine images per request at fixed aspect ratios or a custom size between 512 and 2048 pixels.
pick it if
you want several images from one prompt, up to nine per request
you need a fixed aspect ratio from 9:16 to 21:9
you want to set an exact width and height for the output
skip it if
you need to edit or reference an existing image; this lane takes a text prompt only
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, MiniMax Image-01 or FLUX.2 Klein?
MiniMax Image-01 is cheaper — $0.004 / image vs $0.03 / image.
how do i use MiniMax Image-01 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.