Midjourney Text To Image

Midjourney Text To Image

Playground

Create high-quality, artistic images from text prompts using Midjourney’s renowned creative interpretation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Midjourney Text-to-Image

Midjourney Text-to-Image generates high-quality images from natural-language prompts with flexible aspect ratios, style controls, reference image guidance, and support for multiple Midjourney model versions. It is suitable for concept art, stylized visuals, cinematic imagery, anime-inspired art, and other prompt-driven image generation workflows.

Why Choose This?

• Flexible Midjourney model versions Choose from 6, 6.1, 7, or 8.1 depending on your preferred look and behavior.

• Strong style control Fine-tune the result with stylize, chaos, and weird to balance coherence, creativity, and experimental output.

• Reference image support Use sref to guide the generated image toward a desired visual direction.

• Aspect ratio control Supports square, portrait, landscape, and cinematic aspect ratios.

• Optional HD generation Version 8.1 supports optional native 2k HD generation.

• Anime-focused mode Use niji settings when you want anime or Eastern-aesthetic results.

Parameters

How to Use

1. Write your prompt — describe the subject, style, lighting, mood, and composition you want.
2. Choose aspect ratio — select the output framing that fits your use case.
3. Set style controls (optional) — adjust stylize, chaos, and weird for more artistic or experimental results.
4. Choose version — use the Midjourney version that best matches your workflow.
5. Enable HD (optional) — turn on hd if you are using version 8.1 and want higher-resolution output.
6. Add a reference image (optional) — use sref when you want stronger style guidance.
7. Set seed (optional) — use -1 for random output, or a fixed seed for more reproducible results.
8. Submit — run the model and download the generated image.

Example Prompt

A cinematic sci-fi city street at night, neon reflections on wet pavement, dramatic fog, ultra-detailed architecture, moody lighting, dynamic composition

Pricing

Pricing is fixed per image, with an additional charge for HD generation on version 8.1.

Billing Rules

• Base price is $0.10 per image
• Enabling hd adds a surcharge only when version = 8.1
• version = 8.1 with hd=true costs $0.15 per image
• aspect_ratio, quality, stylize, chaos, weird, niji, seed, and sref do not affect pricing

Best Use Cases

• Concept art — Explore visual directions for environments, characters, and products.
• Stylized illustration — Generate artistic, cinematic, or graphic compositions from text prompts.
• Anime-inspired imagery — Use niji settings for anime and Eastern-aesthetic outputs.
• Marketing creatives — Produce eye-catching visuals for campaigns, ads, and presentations.
• Reference-guided ideation — Use sref to push the output closer to a target style or mood.

Pro Tips

• Use stylize when you want a stronger artistic interpretation.
• Increase chaos when you want more variation and less predictability.
• Use weird sparingly when you want unusual or surreal results.
• Only enable hd when you need the extra resolution, since it increases cost.
• Use sref when style consistency matters more than purely prompt-based generation.
• Keep seed fixed when comparing prompt or parameter changes across runs.

Notes

• prompt is required.
• hd only applies to version 8.1.
• seed = -1 means random generation.
• niji is intended for anime and Eastern-style visual output.
• Pricing is fixed per image, with a higher price only for 8.1 HD generation.

Related Models

• Other Midjourney image generation workflows — Useful when you want different generation modes or prompt setups.
• Reference-guided image generation workflows — Useful when style steering from an image matters more heavily.
• Anime-focused image generation workflows — Useful when stylized illustration and character art are the main priority.

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/midjourney/text-to-image" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "aspect_ratio": "1:1",
    "quality": 1,
    "stylize": 0,
    "chaos": 0,
    "weird": 0,
    "version": "8.1",
    "hd": false,
    "niji": "close",
    "seed": -1,
    "enable_base64_output": false
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"