AI Models · flux

FLUX.2 Pro

High-quality image generation and editing with up to eight reference images.

10 CR / RUN · flux · 7.8M RUNS

FLUX.2 Pro

About

flux-2-pro is a high-quality image model from Black Forest Labs built for both generation and editing. It supports prompts, adjustable dimensions, aspect ratios, and output settings for flexible creative control.

A standout feature is support for up to eight reference images, making it a strong fit for workflows that need visual guidance across multiple inputs. With millions of runs on Replicate, it is a widely used option for image creation and refinement.

Strengths

  • 01High-quality image generation and editing
  • 02Supports up to eight reference images
  • 03Flexible control over size and aspect ratio
  • 04Adjustable output format and quality
  • 05Seed support for repeatable results

Example prompts

Create a high-end fashion editorial portrait in soft studio lighting, clean background, realistic skin texture, 4:5 aspect ratio.

Try this prompt

Edit this product image into a premium ecommerce hero shot with balanced lighting, subtle shadows, and a minimal modern backdrop.

Try this prompt

Generate a cinematic sci-fi city street at night, reflective pavement, neon signage, atmospheric fog, ultra-detailed composition.

Try this prompt

Using the reference images, create a cohesive character concept sheet with consistent face, outfit, and color palette across poses.

Try this prompt

Use cases

Reference-guided concept art

Use up to eight reference images to steer style, subject, or visual direction for more controlled concept development.

Image editing workflows

Refine existing visuals with prompt-based edits while keeping control over output quality, format, and composition settings.

Marketing and product visuals

Create polished campaign, ecommerce, or brand imagery with customizable dimensions and aspect ratios for different channels.

Repeatable creative iterations

Use seeds and structured settings to revisit ideas, compare variations, and maintain consistency across multiple generations.

FAQ

Related models