Flux 2

Capabilities

What it costs in credits

Derived from the same pricing source the app charges against, so these figures and the cost preview on the generation screen can never disagree.

Credits come from your plan and from credit packs. See the pricing page for the plan ladder.

When to use Flux 2

Flux 2 from Black Forest Labs is the compositing model of the catalogue. It is edit-only: there is no text to image path, so it always starts from images you supply, and it takes a large reference stack in one call. That is precisely what it is for. When a shot has to combine a character, a garment, a product and a background that all came from somewhere else, this is the model that holds them together in one frame without you staging the whole thing in a prompt and hoping. It has a single fixed quality setting, so there is no tier to choose and the price is the same every time, which makes it easy to budget for in a batch. Reach for it when the wardrobe system has produced a garment and you need the character wearing it, when a product asset has to appear in a scene that already exists, or when you are trying to reconcile several references that a text-driven model keeps blending rather than placing. Its edit-only nature is also the main thing to plan around: you need a starting image, so pair it with a text-capable model. A common pattern is to generate the base scene on Nano Banana 2 or GPT Image 2, then bring it here with the extra references to compose the final frame. Where you would pick something else: Nano Banana 2 for plain language edits to a single image, Nano Banana Pro when the finished frame needs 4K detail, and Seedream 5.0 Pro when the challenge is physical logic rather than combining sources.

Best for

• Putting a generated garment onto a character
• Placing a product asset into an existing scene
• Reconciling several references a text driven model keeps blending

Other models to consider