Remix
Create variations and remixes of existing images using AI — blend styles, apply artistic transformations, and generate entirely new interpretations of your original content.
What Is Remix?
Remix is an AI-powered image transformation tool that takes an existing image as input and produces a creative variation based on your text prompt. Unlike image generation from scratch, Remix preserves the composition and key elements of the original while applying a new artistic style, color palette, or visual theme.
Use Remix to explore alternative versions of your work, experiment with artistic movements, or create a series of related images that share a consistent visual language.
Core idea
Your source image provides the structure. Your text prompt provides the style. The AI combines both to produce something new that belongs to you.
Key Features
Style Transfer
Apply artistic movements — oil painting, watercolor, cyberpunk, Art Nouveau — to any photograph or digital image.
Prompt-Guided Transformation
Describe exactly what you want with natural language. The more specific your prompt, the closer the result matches your vision.
Negative Prompts
Exclude unwanted visual elements — artifacts, text, watermarks, or specific styles — from the output.
Drag-and-Drop Upload
Drop any JPG, PNG, or WebP image directly onto the editor to start remixing in seconds.
Reference Image Search
Browse the built-in image library to find a style reference image instead of writing a prompt from scratch.
Strength Control
Adjust how strongly the AI departs from the original — from a subtle tint to a complete stylistic reimagining.
Quick Start
- Upload your source image — drag and drop onto the Remix panel or click Upload. Alternatively use the search bar to select an image from the gallery.
- Write a style prompt — describe the artistic direction (see "Writing Effective Prompts" below).
- Set a negative prompt — list elements you do not want in the output (blurry, low quality, watermark, etc.).
- Click Remix — processing takes roughly 30–60 seconds depending on image size.
- Download or continue editing — save the result or open it in the editor to refine further.
Writing Effective Prompts
A well-structured prompt guides the AI toward the exact result you want. Include as many of these elements as relevant:
| Element | What to write | Example |
|---|---|---|
| Artistic movement | Name a recognized art style or movement | "Impressionist oil painting" |
| Color palette | Dominant hues or temperature | "Warm amber and terracotta tones" |
| Mood / atmosphere | Emotional quality of the scene | "Serene and melancholic" |
| Lighting | Light source and quality | "Golden hour backlight, soft shadows" |
| Technical quality | Resolution and detail level | "Ultra-detailed, 4K, sharp edges" |
Prompt examples
- Retro poster: "1980s magazine cover, neon geometric shapes, vintage halftone texture, bold sans-serif typography, high contrast"
- Van Gogh style: "Post-impressionist oil painting, swirling brushstrokes, cobalt blue and cadmium yellow, starry night atmosphere"
- Cyberpunk: "Neon-lit rainy cityscape, electric blue and magenta, volumetric fog, reflective wet pavement, cinematic"
Negative Prompts
Negative prompts tell the AI what to exclude. They are especially useful for removing recurring artifacts.
| Category | Suggested negative prompt terms |
|---|---|
| Quality issues | blurry, pixelated, low resolution, noisy, grainy |
| Unwanted overlays | text, watermark, logo, signature, border, frame |
| Style conflicts | photorealistic (when targeting painterly), cartoon (when targeting realism) |
| Compression artifacts | jpeg artifacts, compression noise, oversaturated |
Style Categories
Classical Art
Renaissance, Baroque, Impressionism, Post-Impressionism — great for portraits and landscapes.
Digital & Contemporary
Cyberpunk, vaporwave, synthwave, glitch art — ideal for modern compositions and branding.
Vintage & Retro
1950s poster, film noir, sepia photography, hand-drawn illustration — adds nostalgia and character.
Fantasy & Concept Art
Epic fantasy, sci-fi concept, dark surrealism — transforms ordinary scenes into imaginative worlds.
Technical Specifications
| Attribute | Value |
|---|---|
| Supported input formats | JPG, PNG, WebP |
| Maximum file size | 10 MB |
| Maximum input dimensions | 4096 × 4096 px |
| Typical processing time | 30–60 seconds |
| Output format | PNG (lossless) |
| AI model | Advanced style-transfer and image-to-image diffusion |