midjourney-v8

Quick example

curl https://reapi.ai/api/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "midjourney",
    "prompt": "a serene Japanese garden with cherry blossoms, painterly",
    "version": "8.1",
    "size": "16:9",
    "speed": "fast"
  }'

import requests

resp = requests.post(
    "https://reapi.ai/api/v1/images/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "model": "midjourney",
        "prompt": "a serene Japanese garden with cherry blossoms, painterly",
        "version": "8.1",
        "size": "16:9",
        "speed": "fast",
    },
    timeout=30,
)
task_id = resp.json()["data"][0]["task_id"]

const resp = await fetch('https://reapi.ai/api/v1/images/generations', {
  method: 'POST',
  headers: {
    Authorization: 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    model: 'midjourney',
    prompt: 'a serene Japanese garden with cherry blossoms, painterly',
    version: '8.1',
    size: '16:9',
    speed: 'fast',
  }),
});
const { data } = await resp.json();

body := strings.NewReader(`{
  "model": "midjourney",
  "prompt": "a serene Japanese garden with cherry blossoms, painterly",
  "version": "8.1",
  "size": "16:9",
  "speed": "fast"
}`)
req, _ := http.NewRequest("POST", "https://reapi.ai/api/v1/images/generations", body)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
req.Header.Set("Content-Type", "application/json")
resp, _ := http.DefaultClient.Do(req)

Generation is asynchronous. The call returns a task_id; poll GET /api/v1/tasks/:task_id until status is completed, then read the image URLs. Generation returns 1–4 images (content review may filter some) and is billed once per request.

Endpoint

POST https://reapi.ai/api/v1/images/generations   # midjourney (image)
POST https://reapi.ai/api/v1/videos/generations    # midjourney-video
GET  https://reapi.ai/api/v1/tasks/:task_id

Operations — action

midjourney runs one operation per request, selected by action (default imagine). All share the images endpoint.

The parent-referencing actions take the reAPI task_id of a completed midjourney task plus index (1–4) to pick which grid image to act on. reAPI resolves your task id to the upstream reference automatically.

Generate — action: "imagine"

The structured parameters below are the JSON form of Midjourney's native --flags. You may also type the flags into prompt; a body field overrides the matching flag. Ranges mirror Midjourney exactly.

Online-verified versions: 8.1, 7, 6.1, 5.2, 5.1, plus niji 7 / niji 6 (set niji: true with version: "7" or "6"). The version does not change the price.

Follow-up operations

All follow-ups share the images endpoint and the async task_id lifecycle. Common fields: task_id (required, a completed midjourney task), index (1–4, which grid image), speed, custom_id (advanced — a button id from the source task, bypassing index auto-match).

blend takes image_urls (2–4) and an optional dimensions (SQUARE / PORTRAIT / LANDSCAPE). edits takes prompt + image_urls plus the same structured params as imagine.

Example — upscale tile 1 of a prior generation:

{
  "model": "midjourney",
  "action": "upscale",
  "task_id": "task_xxx",
  "index": 1
}

Image-to-video — midjourney-video

Animate a still into a short (~5 second) clip on the videos endpoint.

Pricing

midjourney bills per request (a 4-image grid for generation; one image for upscale and the like). The bill depends only on:

• Operation tier — generation vs every other image action.
• Speed — relax / fast / turbo. The model version never changes the price.

midjourney-video bills per resolution (480p / 720p) × batch_size.

Integer credits, 1 credit = $0.001:

credits = ceil(request_price_usd × 1000)        # midjourney
credits = ceil(clip_price_usd × batch_size × 1000)   # midjourney-video

See the model page for current rates.

Output

Poll GET /api/v1/tasks/:task_id. On completed:

{
  "id": "task_xxx",
  "status": "completed",
  "output": { "image_urls": ["https://cdn.reapi.ai/.../0.png", "https://cdn.reapi.ai/.../1.png"] }
}

output.image_urls holds 1–4 URLs (content review may filter some). midjourney-video returns output.video_urls. Save them promptly.

Errors

Customer-facing errors use the standard envelope { error: { code, message, request_id } }. Common cases:

See the full catalog at /docs/api/errors.

Tips

• Generate with imagine first, then chain upscale / variation / zoom / pan / remix_strong off the returned task_id — no re-upload needed.
• A body field overrides the matching --flag in the prompt, so pick one.
• relax is the cheapest and the default; fast / turbo only change queue priority and price, not the visual result.
• remix_strong / remix_subtle require a v8 / v8.1 parent; use the variation actions for older versions.
• All references (cref / sref / dref, blend inputs) must be public http(s) URLs.

Related

• Model page & live playground
• API quickstart
• Error codes