MeiGen lets you generate images using multiple AI models from a unified interface. Open the generation sidebar from the right side of the screen to get started.
Prompt
The prompt editor is where you describe what you want to generate. Tips for writing effective prompts:
- Be specific — describe composition, lighting, colors, textures, and perspective
- Short prompts work too — use Prompt Enhance to expand brief descriptions into detailed prompts
- Language — most models accept English prompts. Niji 7 will auto-translate non-English prompts
You can drag an image from the gallery into the prompt area to copy its original prompt into the editor — a quick way to use an existing prompt as your starting point.
Reference Images
Upload up to 5 reference images to guide the generation. The AI will use these as visual context to influence the output.
- Supported formats: JPG, PNG, WebP
- Auto-compression: Images are automatically compressed to max 2MB / 2048px
- Upload method: Drag-and-drop or click to browse
The maximum number of reference images varies by model. See the Models page for details. Reference Types (Niji 7 only)
When using Midjourney Niji 7 with a reference image, you can choose between two reference types:
| Type | Description |
|---|
| Content Reference | The AI uses the image as subject matter — it tries to recreate the content. Control influence with the Image Weight (iw) parameter (0–3). |
| Style Reference | The AI extracts the visual style (colors, mood, technique) without copying the content. Control with Style Weight (sw, 0–1000). |
Generation Options
Count
Choose how many images to generate per request (1–4). Free users can generate 1 image at a time; paid users can generate up to 4.
Aspect Ratio
Select from a wide range of aspect ratios. Available ratios depend on the selected model:
| Ratio | Orientation | Common Use |
|---|
| 1:1 | Square | Social media profile, icons |
| 3:4, 4:5, 2:3 | Portrait | Phone wallpaper, posters |
| 9:16 | Tall portrait | Stories, Reels |
| 1:4, 1:8 | Ultra tall | Scrolling banners |
| 4:3, 3:2 | Landscape | Desktop wallpaper |
| 16:9, 21:9 | Wide | Cinematic, ultrawide displays |
| 4:1, 8:1 | Ultra wide | Panoramic banners |
Not all models support all ratios. The selector automatically filters to show only ratios supported by the current model.
Resolution
Choose the output resolution when available:
| Resolution | Description |
|---|
| 2K | Standard quality, available to all users |
| 3K | Higher detail (some models only) |
| 4K | Highest detail, paid users only |
Niji 7 Advanced Options
When using Midjourney Niji 7, additional parameters are available:
| Parameter | Range | Default | Description |
|---|
| Stylize | 0–1000 | 150 | Controls how strongly Midjourney’s aesthetic is applied. Lower = more literal, higher = more artistic. |
| Chaos | 0–100 | 0 | Adds variation between runs. Higher values produce more unexpected results. |
| Weird | 0–3000 | 0 | Introduces unusual, experimental qualities. |
| Raw Mode | on/off | on | Produces images with a raw, less processed aesthetic. |
| Image Weight | 0–3 | 1.5 | How much influence the content reference image has. Only available with a content reference. |
| Style Reference | URL/text | — | Provide a URL to an image whose style you want to mimic. |
| Style Weight | 0–1000 | 100 | How strongly the style reference is applied. |
| Style Version | 1–4 | 4 | Which style algorithm version to use. |
Reverse Prompt
See an image you like but don’t know how to describe it? Reverse Prompt uses AI vision to analyze an image and generate a detailed text prompt that captures its content, style, and composition.
How to use
There are two ways to trigger Reverse Prompt:
- Image detail dialog — open any image’s detail view and click the Reverse Prompt button. The AI will analyze the image and generate a descriptive prompt.
- Upload an image — drag and drop any external image onto the Reverse Prompt dropzone above the prompt editor.
What it generates
The AI examines the image and produces a prompt describing:
- Subject matter and scene composition
- Art style and technique (photorealistic, illustration, watercolor, etc.)
- Lighting, color palette, and mood
- Camera angle and perspective
You can use the generated prompt as-is, or edit it to create variations of the original image.
Reverse Prompt works well as a starting point. Try combining it with Prompt Enhance — first reverse-prompt an image you like, then enhance the result for even more detail. Background Removal (Cutout)
After generating an image, hover over it and click the Cutout button to remove the background. This creates a transparent PNG — useful for logos, stickers, or compositing.
Generation Status
After submitting, your image appears as a pending card with a progress indicator. Generation times vary by model:
| Model | Typical Time |
|---|
| Z Image Turbo | ~5 seconds |
| Nanobanana 2, Seedream | ~15 seconds |
| Nanobanana Pro | ~15 seconds |
| GPT image1.5 | ~20 seconds |
| Niji 7 | ~60 seconds |
| Veo 3.1 (video) | 2–6 minutes |
