wan2.7-r2v-2026-06-12

wan2.7-r2v-2026-06-12 offers asynchronous video creation through the UniAll API. One to five ordered image/video references, optional first frame and reference voices; 2-15 seconds with images or 2-10 with videos, at 720P/1080P. Use the model-specific input fields and keep ordered references in their intended roles. Create the task once, then query its public task ID until it succeeds or fails. A task ID is a tracking identifier, while the completed video URL is the actual media result. Download completed videos within 24 hours. Prices use the original Alibaba Cloud Beijing list rates converted to USD at CNY 6.8 per dollar. Output and separately capped reference-video seconds are charged; input images are free. Existing group discounts apply.

A paper collage combines a woman in a vermilion coat, a cream staircase and a black horse.
Model ID
wan2.7-r2v-2026-06-12
Provider
Alibaba
API endpoint
/v1/videos
Pricing
0.147058823529411765
Last updated
2026-10-27

Capabilities

  • One to five ordered image/video references, optional first frame and reference voices; 2-15 seconds with images or 2-10 with videos, at 720P/1080P.
  • Unified and native asynchronous API formats with the same API key.
  • Locally measured output duration with failure refunds and request-time price freezing.

Limitations

  • Completed media URLs expire after 24 hours; save the result promptly.
  • Generation is probabilistic; a seed does not guarantee identical media.
  • No audio boolean is exposed for this model. Use a supported Flash or Wan 3.0 model when an explicit silence switch is required.

Use cases

  • Cinematic advertising and branded storytelling.
  • Visual storyboards and reference-directed creative production.
  • Automated video creation with task polling and USD usage tracking.

Pricing and billing examples

  • 1080P billable reference video | 0.147058823529411765 | input_video_seconds | {"resolution_tier":"1080p"}
  • 720P billable reference video | 0.088235294117647059 | input_video_seconds | {"resolution_tier":"720p"}
  • 1080P output video | 0.147058823529411765 | output_video_seconds | {"resolution_tier":"1080p"}
  • 720P output video | 0.088235294117647059 | output_video_seconds | {"resolution_tier":"720p"}

Request parameters

ParameterTypeRequiredDefaultOptions
modelstringYes
prompttextareaYes
imagesasset_listNo
videosasset_listNo
durationintegerNo5
resolutionselectNo1080P720P, 1080P
seedintegerNo
prompt_extendbooleanNotrue
watermarkbooleanNofalse
aspect_ratioselectNo16:916:9, 4:3, 1:1, 3:4, 9:16
negative_prompttextareaNo

API examples

cURL

curl -X POST 'https://api.uniall.ai/v1/videos' \
  -H 'Authorization: Bearer $UNIALL_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"aspect_ratio":"16:9","duration":5,"model":"wan2.7-r2v-2026-06-12","prompt":"Describe the output you want","prompt_extend":true,"resolution":"1080P","watermark":false}'

Python

import json
import requests

payload = json.loads(r'''{"aspect_ratio":"16:9","duration":5,"model":"wan2.7-r2v-2026-06-12","prompt":"Describe the output you want","prompt_extend":true,"resolution":"1080P","watermark":false}''')
response = requests.post(
    "https://api.uniall.ai/v1/videos",
    headers={"Authorization": "Bearer " + UNIALL_API_KEY},
    json=payload,
)
response.raise_for_status()
result = response.json()

JavaScript

const response = await fetch("https://api.uniall.ai/v1/videos", {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.UNIALL_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({"aspect_ratio":"16:9","duration":5,"model":"wan2.7-r2v-2026-06-12","prompt":"Describe the output you want","prompt_extend":true,"resolution":"1080P","watermark":false}),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
const result = await response.json();

Frequently asked questions

Which APIs are available?

Create with POST /v1/videos or /v1/videos/generations and poll GET /v1/videos/{task_id}. Native creation uses POST /api/v1/services/aigc/video-generation/video-synthesis, X-DashScope-Async: enable and GET /api/v1/tasks/{task_id}.

Which inputs and durations are supported?

One to five ordered image/video references, optional first frame and reference voices; 2-15 seconds with images or 2-10 with videos, at 720P/1080P.

Can I generate a silent video?

No audio boolean is exposed for this model. Use a supported Flash or Wan 3.0 model when an explicit silence switch is required.

What are the USD prices?

720P $0.088235294118/s; 1080P $0.147058823529/s. Output and separately capped reference-video seconds are charged; input images are free. Original list prices divided by 6.8; temporary promotions are excluded. Group discounts apply.

Is a task ID the final result?

No. Poll the public task ID; only a successful task returns its video URL. Save the video within 24 hours.

Can I retry an uncertain creation request?

Do not automatically resend an uncertain paid submission. Poll the returned task ID when available; inspect an unknown submission before creating another task.

How is reference-video input duration charged?

Input sharing counts videos only, excluding reference images and the first frame. Apply the cap separately to each video: 1/2/3/4/5 items use 5/2.5/1.65/1.25/1 seconds per video; sum the capped values.