Image-to-Video

Doc version：v1.0.0 | Last updated：2026-07-22
This platform fully supports the official Tongyi Wanxiang (Wan) video generation API. Requests and responses are transparently proxied; parameter semantics are identical to the official API.

Generate video using an image as the first frame, combined with a text prompt. Image-to-Video has two request protocols sharing the same endpoint, distinguished by model and request body structure:

Protocol	Applicable models	Input method	Capabilities
New (recommended)	`wan2.7-i2v` and other wan2.7-series models	`input.media` array	First-frame-to-video, first/last-frame-to-video, audio-driven generation, video continuation
Legacy	`wan2.6-i2v`, `wan2.6-i2v-flash`, `wan2.5-i2v-preview`, `wan2.2-i2v-plus`, `wan2.2-i2v-flash`, `wanx2.1-i2v-turbo`, `wanx2.1-i2v-plus`	`input.img_url` field	First-frame-to-video only; wan2.6/wan2.5 support audio (`audio_url`) and auto dubbing, wan2.2/wanx2.1 support video effect templates

Create Task

POST https://platform.dataeyes.ai/ali/api/v1/services/aigc/video-generation/video-synthesis

Request Headers

Parameter	Type	Required	Default	Description
`Content-Type`	string	Yes	application/json	Data exchange format
`Authorization`	string	Yes		Authentication, `Bearer {API_KEY}`

The X-DashScope-Async: enable header required by the official API is added automatically by the platform; you do not need to include it.

Request Body — New Protocol (wan2.7)

Parameter	Type	Required	Default	Description
`model`	string	Yes		Model name, e.g. `wan2.7-i2v`
`input.prompt`	string	No		Text prompt, Chinese or English, up to 5000 characters; excess is automatically truncated
`input.negative_prompt`	string	No		Negative prompt, up to 500 characters; excess is automatically truncated
`input.media`	array	Yes		List of media assets; each element is a `{type, url}` object
`input.media[].type`	string	Yes		Asset type: `first_frame` (first-frame image), `last_frame` (last-frame image), `driving_audio` (driving audio), `first_clip` (leading video clip, for video continuation). Each type may appear at most once in the array
`input.media[].url`	string	Yes		Asset URL. Public HTTP(S) URLs are supported; images also support Base64 (`data:{MIME_type};base64,{base64_data}`), while audio/video do not support Base64
`parameters.resolution`	string	No	1080P	Resolution tier, `720P` or `1080P`; directly affects cost. Output is scaled automatically to the tier's total pixel count, keeping the aspect ratio as close to the input asset as possible
`parameters.duration`	integer	No	5	Video duration in seconds, integer in [2, 15]; directly affects cost. For video continuation, this is the final total duration (input clip + continuation), and billing is based on the total duration
`parameters.prompt_extend`	boolean	No	true	Whether to enable intelligent prompt rewriting; noticeably improves short prompts but increases processing time
`parameters.watermark`	boolean	No	false	Whether to add an "AI-generated" watermark at the bottom-right corner of the video
`parameters.seed`	integer	No	Random	Random seed, range [0, 2147483647]; identical seeds do not guarantee identical results

Media Combination Rules

input.media only supports the following combinations; invalid combinations return an error:

Task type	Valid combination	Description
First-frame-to-video	`first_frame`	Generate a video with the image as the first frame; the model dubs automatically (background music / sound effects)
First frame + audio-driven	`first_frame + driving_audio`	Drive the visuals with the provided audio (lip sync, motion beat matching)
First/last-frame-to-video	`first_frame + last_frame`	Generate a transition video from the first frame to the last frame
First/last frames + audio-driven	`first_frame + last_frame + driving_audio`	First/last-frame transition + audio-driven
Video continuation	`first_clip`; `first_clip + last_frame`	Continue after the input video clip; a last frame can optionally be specified

Request Body — Legacy Protocol (wan2.6 and Earlier)

Parameter	Type	Required	Default	Description
`model`	string	Yes		Model name, e.g. `wan2.6-i2v`, `wan2.6-i2v-flash`, `wan2.2-i2v-plus`, `wanx2.1-i2v-turbo`
`input.prompt`	string	No		Text prompt, Chinese or English; wan2.6/wan2.5 series up to 1500 characters, wan2.2/wanx2.1 series up to 800 characters, excess is automatically truncated. Ignored when a `template` effect is used
`input.negative_prompt`	string	No		Negative prompt, up to 500 characters
`input.img_url`	string	Yes		First-frame image URL. Public HTTP(S) URL or Base64 (`data:{MIME_type};base64,...`)
`input.audio_url`	string	No		wan2.6 / wan2.5 series only. Audio URL (public URL); if provided, the video is generated with this audio, otherwise the model dubs automatically
`input.template`	string	No		wan2.2 / wanx2.1 series only. Video effect template name (e.g. `flying`); when used, `prompt` is ignored, and effect availability depends on the model
`parameters.resolution`	string	No	Model-dependent	Resolution tier; directly affects cost. Aspect ratio is kept as close to `img_url` as possible. See the tier table below for per-model values
`parameters.duration`	integer	No	Model-dependent	Video duration in seconds; directly affects cost. See the tier table below for per-model values
`parameters.prompt_extend`	boolean	No	true	Whether to enable intelligent prompt rewriting
`parameters.shot_type`	string	No	single	wan2.6 series only. `single` (single shot) / `multi` (multi-shot storytelling); only takes effect when `prompt_extend=true`, precedence: shot_type > prompt
`parameters.audio`	boolean	No	true	`wan2.6-i2v-flash` only. Whether to generate a video with audio; pricing differs between audio and silent output. Precedence: audio > audio_url — when `audio=false`, the output is silent and billed as silent even if `audio_url` is provided
`parameters.watermark`	boolean	No	false	Whether to add an "AI-generated" watermark
`parameters.seed`	integer	No	Random	Random seed, range [0, 2147483647]

The wan2.2 and wanx2.1 series generate silent videos by default.

Input Asset Requirements

Images (New protocol first_frame / last_frame; legacy img_url)

Item	Requirement
Format	JPEG, JPG, PNG (transparency not supported), BMP, WEBP
Resolution	Both width and height within [240, 8000] pixels
Aspect ratio	1:8 to 8:1
File size	wan2.7 / wan2.6 / wan2.5: up to 20MB; wan2.2 / wanx2.1: up to 10MB
Input method	Public HTTP(S) URL, or Base64 (supported MIME types: image/jpeg, image/png, image/bmp, image/webp)

Audio (New protocol driving_audio; legacy audio_url)

Item	wan2.7 (driving_audio)	wan2.6 / wan2.5 (audio_url)
Format	wav, mp3	wav, mp3
Duration	2–30 seconds	3–30 seconds
File size	≤15MB	≤15MB
Truncation rule	Audio longer than `duration` is trimmed to the first `duration` seconds; if shorter, the remainder is silent (e.g. 3s audio, 5s video → first 3s with audio, last 2s silent)	Same as left
Semantics	Provided = audio-driven (lip sync, motion beat matching); omitted = auto-generated background music / sound effects	Provided = generate the video with this audio; omitted = auto dubbing
Input method	Public HTTP(S) URL (Base64 not supported)	Same as left

Video (New protocol first_clip, video continuation)

Item	Requirement
Format	mp4, mov
Duration	2–10 seconds
Resolution	Both width and height within [240, 4096] pixels, aspect ratio 1:8 to 8:1
File size	≤100MB
Continuation rule	The continuation limit is controlled by `duration`: e.g. `duration=15` with a 3s input clip → 12s continuation, 15s total output, billed as 15s

The output video's aspect ratio is determined by the input asset (first-frame image / leading video clip), scaled automatically to the resolution tier's total pixel count; video width and height must be multiples of 16, so there may be a slight deviation from the original image's aspect ratio.

Resolution and Duration Tiers per Model

Model	resolution options	resolution default	duration values (seconds)	duration default
`wan2.7-i2v`	720P, 1080P	1080P	Integer [2, 15]	5
`wan2.6-i2v`	720P, 1080P	1080P	Integer [2, 15]	5
`wan2.6-i2v-flash`	720P, 1080P	1080P	Integer [2, 15]	5
`wan2.5-i2v-preview`	480P, 720P, 1080P	1080P	5, 10	5
`wan2.2-i2v-flash`	480P, 720P, 1080P	720P	Fixed 5, not modifiable	5
`wan2.2-i2v-plus`	480P, 1080P	1080P	Fixed 5, not modifiable	5
`wanx2.1-i2v-turbo`	480P, 720P	720P	3, 4, 5	5
`wanx2.1-i2v-plus`	720P	720P	Fixed 5, not modifiable	5

resolution and duration directly affect cost (billed by resolution tier and duration in seconds); wan2.6-i2v-flash has different prices for audio and silent output.

Request Examples

Example 1: First-Frame-to-Video (wan2.7)

curl --request POST \
  --url 'https://platform.dataeyes.ai/ali/api/v1/services/aigc/video-generation/video-synthesis' \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.7-i2v",
    "input": {
      "prompt": "The camera slowly pushes forward as light and shadow flow across the scene",
      "media": [
        {
          "type": "first_frame",
          "url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260424/mvzfud/hh-v2v-girl.jpg"
        }
      ]
    },
    "parameters": {"resolution": "1080P", "duration": 5}
  }'

Example 2: First/Last-Frame-to-Video (wan2.7)

curl --request POST \
  --url 'https://platform.dataeyes.ai/ali/api/v1/services/aigc/video-generation/video-synthesis' \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.7-i2v",
    "input": {
      "prompt": "Petals drift down in the wind, from full bloom to withering",
      "media": [
        {
          "type": "first_frame",
          "url": "https://wanx.alicdn.com/material/20250318/first_frame.png"
        },
        {
          "type": "last_frame",
          "url": "https://wanx.alicdn.com/material/20250318/last_frame.png"
        }
      ]
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

Example 3: First Frame + Audio-Driven (wan2.7)

curl --request POST \
  --url 'https://platform.dataeyes.ai/ali/api/v1/services/aigc/video-generation/video-synthesis' \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.7-i2v",
    "input": {
      "prompt": "The character sways to the rhythm of the music with lively expressions",
      "media": [
        {
          "type": "first_frame",
          "url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260424/mvzfud/hh-v2v-girl.jpg"
        },
        {
          "type": "driving_audio",
          "url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250825/iaqpio/input_audio.MP3"
        }
      ]
    },
    "parameters": {"resolution": "1080P", "duration": 5}
  }'

Example 4: Legacy Protocol (wan2.6-i2v)

curl --request POST \
  --url 'https://platform.dataeyes.ai/ali/api/v1/services/aigc/video-generation/video-synthesis' \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.6-i2v",
    "input": {
      "img_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260424/mvzfud/hh-v2v-girl.jpg",
      "prompt": "The character slowly turns around, gazing into the distance"
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

Response Example

{
  "output": {
    "task_status": "PENDING",
    "task_id": "0385dc79-5ff8-4d82-bcb6-xxxxxx"
  },
  "request_id": "4909100c-7b5a-9f92-bfe5-xxxxxx"
}

Response Fields

Field	Type	Description
`output.task_id`	string	Task ID used for polling; valid for 24 hours. Do not create duplicate tasks — just keep polling
`output.task_status`	string	Task status; `PENDING` on successful creation. See Overview for the full enum
`request_id`	string	Unique request identifier, useful for troubleshooting
`code` / `message`	string	Error code and details, returned only when creation fails (e.g. `InvalidApiKey`, `InvalidParameter`)

Query Task

After creating a task, poll the task status with the task_id; once the status is SUCCEEDED, get the video URL from output.video_url (valid for 24 hours):

GET https://platform.dataeyes.ai/ali/api/v1/tasks/{task_id}

See 09-Query-Task for details.

02-Image-to-Video

Image-to-Video#

Create Task#

Request Headers#

Request Body — New Protocol (wan2.7)#

Media Combination Rules#

Request Body — Legacy Protocol (wan2.6 and Earlier)#

Input Asset Requirements#

Images (New protocol first_frame / last_frame; legacy img_url)#

Audio (New protocol driving_audio; legacy audio_url)#

Video (New protocol first_clip, video continuation)#

Resolution and Duration Tiers per Model#

Request Examples#

Example 1: First-Frame-to-Video (wan2.7)#

Example 2: First/Last-Frame-to-Video (wan2.7)#

Example 3: First Frame + Audio-Driven (wan2.7)#

Example 4: Legacy Protocol (wan2.6-i2v)#

Response Example#

Response Fields#

Query Task#