FLUX-2-max

Pricing

Input: We charge $0.03 for each megapixel on the input (i.e. reference images for editing)

Output: The first megapixel is charged $0.07 and then each subsequent MP will be charged $0.03

FLUX.2 Max

This is a high-quality model that scales up to highest volumes - FLUX.2 [max] now sets the cherry on top 🍒, improving over FLUX.2 [pro] in all dimensions. It’s easier and more intuitive to prompt and allows for even more consistent editing of characters, object, styles and backgrounds.

If you’re seeking to give your professional content the final touch before publishing at competitive price, FLUX.2 [max] is definitely your model.

For which use cases can I use FLUX.2 [max]

Use FLUX.2 [max] for the following and more

• Use FLUX.2 [max] in use cases where FLUX.2 [pro] lacks consistency in outputs. It will both give you higher hit/miss ratio for very complex edits , involving multiple references as well as better adherence to prompts for image generation.
• Create top quality product imagery from 0 to 1, cutting down the time-to-online from days to minutes.
• Transform low-quality images of your product into high-quality shots that are ready to publish. Get them from your cell phone to your product website in no time.
• Generate highest quality key frames with Motion-Picture-quality to animate static content while preserving not only characters but also their emotions, saving thousands of dollars at pre-visualization.
• Create systematic color variations of your products, precisely directed by hex color codes specified in your prompt.
• New 3D views of a scene from a single or a small set of images: Unlock virtual tours through real or imagined environments.
• Generate images grounded in real-time information - visualize trending products, current events, or the latest styles without manually sourcing reference material.

Why it matters

• Highest precision image editing. FLUX.2 [max] constitutes the highest consistency image editing in FLUX.2 models to date: It preserves colors, lighting, faces, written characters, and objects with all their details making it the default choice for high-quality use cases like product marketing, e-commerce, product variations, interior design, 3D reconstruction, food imagery, video production and more.
• The model that really understands what you want. Maximum prompt following for prompts long and short. FLUX.2 [max] listens to text prompts and edit instruction much better then all FLUX models before.
• Multi-reference editing with state-of-the-art character consistency. FLUX.2 will keep identities, products and styles consistent. Create 50 ad variants using the same face, create product mockups in any context, and dynamic fashion editorials where the model stays consistent.
  • up to 8 reference images possible on API
  • up to 10 reference images on the Playground
• FLUX.2 [max] is our most creative model to date. Generate multiple images based on the same text prompt and none will be similar to the others, while still following the text prompt. Use FLUX.2 [max] for maximum creativity for ideation, stimulating new thoughts as you prompt it.
• Generate and edit images in the established FLUX-speed. While FLUX.2 [max] improves significantly over FLUX.2 [pro] in image quality, prompt adherence editing precision, it still creates new content nearly as fast, making it up to three times as fast as competing models of similar quality.