Overview

UNI-1 Max is the quality-first model in the UNI-1 family, built for both image creation and precision image editing.

It uses the same core workflows as UNI-1, including prompt-based generation, source-image modification, multi-reference guidance, and seed-based iteration, but is positioned for higher-quality output when image polish and final visual quality matter more than using the default variant.

Strengths

Higher-Quality Family Variant

UNI-1 Max is the premium-output member of the UNI-1 family. It is the stronger fit when the same prompt and editing workflows need more refined final results.

Create and Modify in One Model

Like UNI-1, this variant supports both creating brand-new images and modifying an existing source image while preserving the parts of the frame that should remain stable.

Multi-Reference Control

The model supports up to 9 reference images and responds best when each reference is assigned a clear role such as character, style, composition, lighting, color palette, texture, or mood.

Reasoning-Informed Image Editing

UNI-1 Max is built for edits that require more than surface-level stylization. It works well when prompts need spatial consistency, coherent object relationships, or controlled scene logic while still producing a polished result.

Strong Directed Editing

In Modify mode, UNI-1 Max is well suited to premium editing workflows where source fidelity matters and the output needs to feel deliberate, coherent, and finished.

Seeded Iteration and Aspect Ratio Range

The model supports seed control for reproducibility and offers a wide aspect-ratio range for different use cases, including square, portrait, widescreen, panoramic, and ultra-tall formats.

Capabilities

Text-to-Image

UNI-1 Max supports prompt-driven image generation for new compositions, with optional references to guide identity, style, composition, and other visual layers.

Image-to-Image

UNI-1 Max supports precision editing of an existing source image, preserving structure unless the prompt explicitly changes it.

Multi-Reference Generation

The model can use multiple references in Create mode and benefits from explicit role assignment for each reference.

Multi-Turn Refinement

UNI-1 Max is designed for iterative refinement workflows where prompts, references, and seeds are adjusted deliberately over several generations.

Input and Output

• AIR ID: luma:uni@1-max
• Input: text prompts with optional reference images, or a source image plus edit instructions
• Output: generated or modified images
• Reference images: up to 9
• Seed support: supported

Best Fit

• Quality-first image generation
• Premium image editing and refinement
• Multi-reference art direction
• Character consistency workflows
• Final-pass visual polish