wan-2-7-video

Quick example

curl https://reapi.ai/api/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.7-video",
    "prompt": "a paper boat drifting down a rain-soaked street at dusk, neon reflections, slow dolly-in, cinematic",
    "resolution": "1080P",
    "duration": 5
  }'

import requests

resp = requests.post(
    "https://reapi.ai/api/v1/videos/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "model": "wan2.7-video",
        "prompt": "a paper boat drifting down a rain-soaked street at dusk",
        "resolution": "1080P",
        "duration": 5,
    },
    timeout=30,
)
print(resp.json())

const r = await fetch("https://reapi.ai/api/v1/videos/generations", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "wan2.7-video",
    prompt: "a paper boat drifting down a rain-soaked street at dusk",
    resolution: "1080P",
    duration: 5,
  }),
});
console.log(await r.json());

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)

func main() {
    body, _ := json.Marshal(map[string]any{
        "model":      "wan2.7-video",
        "prompt":     "a paper boat drifting down a rain-soaked street at dusk",
        "resolution": "1080P",
        "duration":   5,
    })
    req, _ := http.NewRequest("POST",
        "https://reapi.ai/api/v1/videos/generations", bytes.NewReader(body))
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", "application/json")

    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
    out, _ := io.ReadAll(resp.Body)
    fmt.Println(string(out))
}

Submit response

{
  "id": "task_018f5a3a1b6e7d9f8c2b4d6e8f0a2c4e",
  "model": "wan2.7-video",
  "status": "processing",
  "created_at": 1735000000
}

Poll GET /api/v1/tasks/{id} (see the Tasks reference) until status === "completed". The completed payload's output.video_urls holds the generated MP4 URLs.

Authentication

Every call needs a Bearer token. Generate keys at reapi.ai/settings/apikeys.

Authorization: Bearer YOUR_API_KEY

Endpoint

POST /api/v1/videos/generations
GET  /api/v1/tasks/{id}

Submission is async. The POST returns immediately with a task_id; the task endpoint returns the same envelope until completion. Polling does not consume credits.

Modes

There is no mode field. The umbrella model wan2.7-video infers the mode from which inputs you send:

The fields valid in each mode are mutually exclusive — sending, e.g., a reference field together with a video_url edit source is rejected as invalid.

Parameters

Duration limits by mode

Response envelope

Submit and poll share the same shape — only status and output fill in over time.

{
  "id": "task_018f5a3a1b6e7d9f8c2b4d6e8f0a2c4e",
  "model": "wan2.7-video",
  "status": "completed",
  "created_at": 1735000000,
  "output": {
    "video_urls": ["https://cdn.reapi.ai/media/tasks/018f5a3a1b6e7d9f8c2b4d6e8f0a2c4e/0.mp4"]
  },
  "error": null
}

Pricing

Per-second × resolution. The mode does not change the rate — only the resolution does. 1080P costs more per second than 720P.

For the input-billed modes (reference-to-video and video editing) reApi probes the uploaded clip server-side and folds input + output into the billable second count, rather than trusting a client value. Base text/image modes bill the requested output duration.

Bill formula (1 credit = $0.001):

billable_seconds = ceil(server_probed_or_output_seconds)
bill_usd         = per_second_usd(resolution) × billable_seconds
credits          = ceil(per_second_usd × billable_seconds × 1000)

See current per-second rates for 720P and 1080P on the model page. Failed jobs refund automatically. A probe failure on an input-billed mode returns 400 PRICING_UNAVAILABLE (code 30002) with no charge.

Validation errors

All cases below return HTTP 400. Pattern-match on code, not message — message strings carry request-specific context and are not a stable contract.

The full envelope is { "error": { "code", "message", "request_id" } } — see Errors catalog for the wire format and request_id correlation tips.

Recipes

Text-to-video

{
  "model": "wan2.7-video",
  "prompt": "a paper boat drifting down a rain-soaked street at dusk",
  "resolution": "1080P",
  "duration": 5
}

Image-to-video (animate a still)

{
  "model": "wan2.7-video",
  "prompt": "the subject turns to camera and smiles softly",
  "image_urls": ["https://your-cdn.com/portrait.png"],
  "resolution": "1080P",
  "duration": 5
}

Reference-to-video

{
  "model": "wan2.7-video",
  "prompt": "apply the reference choreography to the new character",
  "reference_image_urls": ["https://your-cdn.com/subject.png"],
  "reference_video_urls": ["https://your-cdn.com/dance-6s.mp4"],
  "resolution": "1080P",
  "duration": 6
}

Video editing

{
  "model": "wan2.7-video",
  "prompt": "restyle the clip with warm cinematic color grading",
  "video_url": "https://your-cdn.com/source-8s.mp4",
  "audio_setting": "origin",
  "duration": 0
}

Tips

• Let the inputs pick the mode. There is no mode flag — send a prompt for text-to-video, add image_urls for image-to-video, reference_* for reference-to-video, or video_url to edit. Don't mix fields across modes.
• Match duration to the mode. Base modes allow 2–15s; reference-with-video and editing cap at 10s, and editing accepts 0 for the full source length.
• Resolution drives cost. 720P is the cheaper run; switch to 1080P for the final take. The mode does not change the per-second rate.
• Keep reference_voice_urls aligned. Their count must equal the number of reference_image_urls, one voice per subject.
• Prompt the motion, not just the scene. Describing the action ("turns and waves, smooth and continuous") sharpens text-to-video and image-to-video results.

Related

• Errors catalog
• Authentication
• Quickstart