Black Forest Labs: FLUX.2 Max

FLUX.2 Max is an image generation model from Black Forest Labs, the lab behind the FLUX family of image models. It sits at the top tier of that family, positioned by the provider as its highest level of image quality, prompt understanding and editing consistency. On api14 it appears under the model id black-forest-labs/flux.2-max and is listed in the image category.

The model accepts text and image inputs and produces image output. Its context window is 46,864 tokens, with a maximum output of 42,177. Generation is controlled through image parameters rather than a broad set of sampling options: aspect_ratio accepts one of 1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16, 21:9 or auto; output_format accepts png or jpeg; and n has a fixed range of 1 to 1, so each request returns a single image. The model also accepts up to 8 input references, which is what makes image-to-image work and editing against supplied images possible, and seed is exposed as a boolean parameter.

Those capabilities point to a fairly specific set of uses. Text to image generation covers the usual product, editorial and concept work where you describe a scene and pick an aspect ratio to match the placement. Because image inputs are supported alongside text, the model also fits editing and reference driven workflows: passing one or more of the up to 8 input references lets you steer a generation toward an existing image, restyle it, or keep a subject consistent across outputs. The aspect ratio list covers square, portrait, landscape, wide and ultra wide framing, so assets can be generated close to their final crop instead of being cut down afterwards. The choice between png and jpeg is useful when you need either lossless output or smaller files.

You call it through api14's OpenAI-compatible API using the model id black-forest-labs/flux.2-max. Requests follow the same shape as other models on the gateway, with the model id in the request and your prompt as input. The parameters the model supports are seed, plus the image parameters aspect_ratio, output_format, n and input_references. Aspect ratio and output format are enums with the values listed above, n is fixed at 1, input_references takes between 0 and 8 images, and seed is a boolean.

A few constraints are worth noting before you commit. Output is image only, so if you need text back from the same call, this is not the model for that. Each request returns exactly one image, so batches have to be issued as separate calls. If your task is text generation, classification or anything that does not end in an image, a text model on api14 will fit better.