Midjourney V8: AI Image Generator
Generate 4 photorealistic images per run with Midjourney V8. Improved coherence and detail over V7. Supports text-to-image and reference image guidance.
Generate 4 photorealistic images per run with Midjourney V8. Improved coherence and detail over V7. Supports text-to-image and reference image guidance.
About this model
Midjourney V8 is the latest photorealistic model, delivering improved coherence, sharper detail, and better prompt adherence over V7. Each run produces 4 high-quality images from your text prompt or reference image. At $0.10 per run it is the most capable option for demanding creative and commercial work.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | /bin/zsh.10 per run (4 images) | Latest V8 model with full parameter control and 4 images per generation. |
| Fal.ai | Not available | Midjourney is not available on Fal.ai. |
| Replicate | Not available | Midjourney is not available on Replicate. |
Latest V8 model with full parameter control and 4 images per generation.
Midjourney is not available on Fal.ai.
Midjourney is not available on Replicate.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text description of the image to generate. | A celestial cartographer mapping moving constellations inside a circular observatory suspended above waterfalls, rotating brass instruments casting shifting shadows, star reflections flowing across polished stone floors, elegant cinematic framing, highly detailed textures, dreamlike realism, sophisticated visual storytelling. |
| Reference Image URL | string | Optional reference image URL. Influences the style and content of the generation. | undefined |
| Aspect Ratio | Enum (6 options) | Output image aspect ratio. | 1:1 |
| Stylize | int | Controls how artistic the result is. Range: 0–1000. Lower = more literal, higher = more stylized. | 100 |
| Chaos | int | Controls variation between the 4 images. Range: 0–100. Higher = more diverse. | 0 |
| Weird | int | Adds unconventional aesthetics. Range: 0–3000. | 0 |
| Negative Prompt | string | Things to exclude from the image, e.g. "text, watermark". | text, watermark, blurry |
| Seed | int | Reproducibility seed. Range: 0–4294967295. Same seed + same prompt ≈ similar result. | 0 |
Text description of the image to generate.
A celestial cartographer mapping moving constellations inside a circular observatory suspended above waterfalls, rotating brass instruments casting shifting shadows, star reflections flowing across polished stone floors, elegant cinematic framing, highly detailed textures, dreamlike realism, sophisticated visual storytelling.Optional reference image URL. Influences the style and content of the generation.
undefinedOutput image aspect ratio.
1:1Controls how artistic the result is. Range: 0–1000. Lower = more literal, higher = more stylized.
100Controls variation between the 4 images. Range: 0–100. Higher = more diverse.
0Adds unconventional aesthetics. Range: 0–3000.
0Things to exclude from the image, e.g. "text, watermark".
text, watermark, blurryReproducibility seed. Range: 0–4294967295. Same seed + same prompt ≈ similar result.
0Developer documentation
Write a prompt: V8 follows prompts more closely than V7 — be precise about composition, lighting, and style.
Add a reference image (optional): Provide a to guide style or subject.
Tune the parameters:
Pick an aspect ratio: , , , , , or .
Poll for results: Each run returns 4 image URLs once complete (typically 30–90 seconds).
Frequently asked
V8 delivers better prompt adherence, improved anatomical coherence, sharper detail, and more consistent quality across all 4 generated images.
Every run generates 4 upscaled images returned in the output array.
Yes — provide a URL in . The reference influences the style and content of all 4 outputs.
Use V8 when you need maximum quality and prompt fidelity. V7 is a solid choice for faster iteration and experimental work.