MidJourney SREF Codes

What They Are and How to Use Them in the Web Interface

What SREF Codes Are

SREF means Style Reference. An SREF code is a numeric identifier that represents a complete visual style. Instead of describing a look with many words, you attach a single code to your prompt and MidJourney applies that aesthetic automatically.

Think of an SREF code as a style preset:

It controls color palette, contrast, texture, mood, and overall visual language.
It does not define the subject, only how the image looks.
The same code can be reused across many different prompts.

SREF codes allow you to separate what you generate from how it looks.

Why SREF Codes Matter

Describing style with text alone is unreliable. Words like “cinematic”, “moody”, or “painterly” are vague and interpreted differently every time.

SREF codes solve this problem by:

Providing style consistency across multiple images.
Reducing prompt length and complexity.
Allowing fast exploration of different aesthetics.
Making visual output repeatable and controllable.

If you are creating a series, a brand identity, or a coherent visual project, SREF codes are essential.

How SREF Codes Work

When you add an SREF code to a prompt, MidJourney uses that code as a style reference layer. Your prompt is interpreted in two parts:

Subject description: What the image is about.
Style reference: How the image should look.

The SREF code influences lighting, color, texture, and mood, while your text defines the content.

How to Use SREF Codes in the Web Interface

This guide refers only to the MidJourney web interface.

Basic Usage

Start a new image prompt in the web interface.
Write your subject description.
Add –sref followed by the numeric code.
Generate the image.

Example Prompt portrait of a woman standing in the rain –sref 2213253170

The subject is defined by the text. The style is defined by the SREF code.

Reusing the Same Style

To keep a consistent look across multiple images, reuse the same SREF code.

Consistency Example portrait of a woman standing in the rain –sref 2213253170

portrait of a woman indoors by a window –sref 2213253170

portrait of a woman walking at night –sref 2213253170

Different scenes. Same aesthetic. This is how you build visual continuity.

Using Multiple SREF Codes Together

You can combine more than one SREF code in the same prompt by listing them after –sref.

Combined Styles futuristic city skyline at dusk –sref 2583392999 4114158294

This blends multiple styles into a single result.

Weighted Style Blending

You can control how much each style influences the image using weights. Higher numbers give that style more influence.

Weighted Example futuristic city skyline at dusk –sref 2583392999::2 4114158294::1

How SREF Codes Fit Into a Workflow

SREF codes work best when treated as style assets, not experiments. A solid workflow looks like this:

Choose one SREF code as your base aesthetic.
Test it across several subjects.
Keep the code fixed.
Change only the subject and composition.

This allows you to maintain stylistic coherence, iterate quickly, and avoid rewriting long style descriptions.

Common Mistakes to Avoid

Changing SREF codes randomly while expecting consistency.
Using too many codes at once without intent.
Treating SREF codes as subjects instead of styles.
Over-describing style in text while also using SREF.

            Pro Tip: Let the SREF code do the style work. Use text to describe content, not aesthetics.
        

When to Use SREF Codes

SREF codes are ideal when you want:

A coherent art series.
A unified brand look.
Consistent editorial or cinematic visuals.
Fast exploration of different aesthetics.
Repeatable results across generations.

They are less useful if you only want one-off experiments with no continuity.

Final Step: Explore the Library

To start using SREF codes effectively, you need access to a wide, well-organized collection. The GenArtCentral SREF library contains 700+ curated SREF codes, ready to copy and paste.

Visit SREF Library →