Wan 2.7 image-to-video

Submit an asynchronous Ali wan2.7-i2v generation task.

The project supports two input styles:

• Compatibility style: use image for first-frame generation, or images with exactly 2 items for first-frame plus last-frame generation.
• Latest style: use top-level media, which maps to the upstream input.media protocol.

Implemented official media.type values are first_frame, last_frame, driving_audio, and first_clip.

Supported media combinations in the current project implementation:

• first_frame
• first_frame + driving_audio
• first_frame + last_frame
• first_frame + last_frame + driving_audio
• first_clip
• first_clip + last_frame

prompt is optional for this channel capability. size supports only 720P and 1080P. seconds supports integer strings from 2 to 15.

POST /v1/videos

cURL

wan27_multimodal

curl --request POST \
  --url https://api.powertokens.ai/v1/videos \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "wan2.7-i2v",
  "prompt": "The portrait gradually starts speaking with subtle head motion.",
  "media": [
    {
      "type": "first_frame",
      "url": "https://example.com/assets/portrait.png"
    },
    {
      "type": "driving_audio",
      "url": "https://example.com/assets/voice.mp3"
    }
  ],
  "seconds": "8",
  "size": "720P"
}
'

200

{
  "id": "<string>",
  "task_id": "<string>",
  "object": "video",
  "model": "<string>",
  "status": "queued",
  "progress": 123,
  "created_at": 123
}

Authorizations

Authorization · string · header · required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Option 1

Option 2

Option 3

model · enum<string> · required

Ali wan2.7-i2v model.

Available options: wan2.7-i2v

image · string · required

Compatibility field for single first-frame input. The project maps it to upstream media: [{type: first_frame, url: ...}].

prompt · string

Optional text prompt. When omitted, generation relies entirely on the provided media inputs.

images · string[]

Compatibility field for first-frame plus last-frame input. Must contain exactly 2 items in order: [first_frame, last_frame].

Required array length: 2 elements

media · object[]

Latest multimodal input style for wan2.7-i2v. Use this field for official upstream combinations such as first frame + driving audio or first clip continuation.

Show child attributes

seconds · string · default: 5

Video duration in string form. Supported values are integer strings from 2 to 15.

size · enum<string> · default: 1080P

Output resolution tier for the current project implementation.

Available options: 720P, 1080P

Response

200

application/json

Submission successful. Returns the unified OpenAI-style video task object.

id · string

Public task ID.

task_id · string

Public task ID alias.

object · string

Example:

"video"

model · string

Model name.

status · enum<string>

Unified video task status.

Available options: queued, in_progress, completed, failed, unknown

progress · integer

Task progress percentage.

created_at · integer<int64>

Unix timestamp in seconds.