Wan 2.7 Video API

Alibaba video generation and editing API for text, image, reference, and video-edit workflows

Overview

Wan 2.7 Video is an Alibaba video model for text-to-video, image-to-video, reference-guided generation, and video editing. Use this page when you are ready to call the API after trying the model in the APIXO playground.

Capability	Value
Model ID	wan-2-7-video
Modes	text-to-video, image-to-video, reference-to-video, video-edit
Prompt	Required for text-to-video, reference-to-video, and video-edit; optional for image-to-video
Resolutions	720p, 1080p
Aspect ratios	16:9, 9:16, 1:1, 4:3, 3:4
Mode-specific ratio rules	Default 16:9 for text-to-video and reference-to-video; not supported for image-to-video; optional for video-edit
Duration	text-to-video/image-to-video: 2-15 (default 5); reference-to-video: 2-15 or 2-10 when any reference video is used (default 5); video-edit: 0 or 2-10 (default 0)
Public media fields	image_urls, video_urls, audio_urls

Public input contract

• Public media input fields are image_urls, video_urls, and audio_urls.
• Use audio_urls for audio input.
• For reference-to-video, audio_urls aligns by slot order: first image_urls, then video_urls.
• For reference-to-video, use empty string "" in audio_urls to skip a specific reference slot without shifting later audio bindings.

Endpoint and authentication

Base URL:

https://api.apixo.ai/api/v1

Method	Endpoint	Purpose
POST	/generateTask/wan-2-7-video	Submit a generation or edit task
GET	/statusTask/wan-2-7-video?taskId={taskId}	Poll task status and retrieve results

All requests require your APIXO API key:

Authorization: Bearer YOUR_API_KEY

Submit requests also require:

Content-Type: application/json

Copy-paste async quickstart

This minimal request submits a text-to-video task and returns a taskId.

curl -X POST "https://api.apixo.ai/api/v1/generateTask/wan-2-7-video" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "async",
    "input": {
      "mode": "text-to-video",
      "prompt": "a cinematic rainy night in tokyo",
      "resolution": "1080p",
      "ratio": "16:9",
      "duration": 5
    }
  }'

Successful response:

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678"
  }
}

Save the taskId; you need it to poll for the final result.

Poll for result

curl -X GET "https://api.apixo.ai/api/v1/statusTask/wan-2-7-video?taskId=task_12345678" \
  -H "Authorization: Bearer YOUR_API_KEY"

Processing response:

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678",
    "state": "processing",
    "createTime": 1767965610929
  }
}

Success response:

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678",
    "state": "success",
    "resultJson": "{\"resultUrls\":[\"https://file.apixo.ai/video.mp4\"]}",
    "createTime": 1767965610929,
    "completeTime": 1767965652317,
    "costTime": 41388
  }
}

Failed response:

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678",
    "state": "failed",
    "failCode": "PromptInvalid",
    "failMsg": "Prompt is invalid or rejected by provider",
    "createTime": 1767965610929,
    "completeTime": 1767965620132
  }
}

Parse resultJson after state becomes success:

const payload = JSON.parse(data.resultJson);
const videoUrls = payload.resultUrls;

Request body

Text-to-video

{
  "request_type": "async",
  "input": {
    "mode": "text-to-video",
    "prompt": "a cinematic rainy night in tokyo",
    "audio_urls": [
      "https://example.com/background.mp3"
    ],
    "negative_prompt": "blurry, low quality",
    "resolution": "1080p",
    "ratio": "16:9",
    "duration": 8,
    "prompt_extend": true,
    "watermark": false,
    "seed": 12345
  }
}

Image-to-video (first and optional last frame)

{
  "request_type": "async",
  "input": {
    "mode": "image-to-video",
    "image_urls": [
      "https://example.com/first-frame.png",
      "https://example.com/last-frame.png"
    ],
    "audio_urls": [
      "https://example.com/driving-audio.mp3"
    ],
    "prompt": "smooth camera motion with gentle wind",
    "resolution": "720p",
    "duration": 6
  }
}

Reference-to-video

{
  "request_type": "async",
  "input": {
    "mode": "reference-to-video",
    "prompt": "keep character style and outfit consistency, but only image 1 should speak",
    "image_urls": [
      "https://example.com/ref-1.png"
    ],
    "video_urls": [
      "https://example.com/ref-clip.mp4"
    ],
    "audio_urls": [
      "https://example.com/ref-voice-1.mp3",
      ""
    ],
    "resolution": "1080p",
    "ratio": "16:9",
    "duration": 6
  }
}

Video-edit

{
  "request_type": "async",
  "input": {
    "mode": "video-edit",
    "video_urls": [
      "https://example.com/input.mp4"
    ],
    "image_urls": [
      "https://example.com/reference.png"
    ],
    "prompt": "change the style into clay animation",
    "negative_prompt": "blurry, low quality",
    "resolution": "720p",
    "ratio": "1:1",
    "duration": 0,
    "audio_setting": "auto",
    "prompt_extend": true,
    "watermark": false
  }
}

Parameters

request_type · string · default: "async" · required

Result delivery mode. Use async for polling with statusTask, or callback for webhook delivery.

callback_url · string

Required when request_type is callback. Must be a public HTTPS URL that can receive the final task payload. See Webhooks.

input · object · required

Wan 2.7 Video input parameters.

Show properties

mode · string · required

Generation mode. Supported values: text-to-video, image-to-video, reference-to-video, video-edit.

prompt · string

Prompt text. Required for text-to-video, reference-to-video, and video-edit. Optional for image-to-video. Required prompts cannot be empty after trimming. Supports Chinese and English; maximum length is 5000 characters and excess content is truncated.

negative_prompt · string

Optional negative prompt. Supports Chinese and English; maximum length is 500 characters and excess content is truncated.

resolution · string · default: "1080p"

Output resolution tier. Supported values: 720p, 1080p.

ratio · string

Supported values: 16:9, 9:16, 1:1, 4:3, 3:4. Default is 16:9 for text-to-video and reference-to-video. Not supported for image-to-video. Optional for video-edit.

duration · integer · default: "5 (or 0 for video-edit)"

Accepts integer or numeric string input. text-to-video and image-to-video: 2-15 (default 5). reference-to-video: 2-15 by default, but 2-10 when any reference video is provided (default 5). video-edit: 0 or 2-10 (default 0).

prompt_extend · boolean · default: "true"

Whether to enable prompt rewriting before generation. true (default) enables rewriting and can improve short prompts but may increase latency. false disables rewriting.

watermark · boolean · default: "false"

Whether to add a watermark. false (default) means no watermark. true adds a bottom-right watermark with fixed text AI 生成.

seed · integer

Optional random seed. Supported range: 0 to 2147483647. A fixed seed improves reproducibility, but identical seeds do not guarantee identical outputs.

image_urls · string[]

Public image URLs. image-to-video: use 1-2 images when no video_urls are provided (first frame, plus optional last frame), or at most 1 image as last frame when video_urls is present. reference-to-video: optional 1-5 images, but total image_urls + video_urls must be 1-5. video-edit: optional 1-4 reference images. Image constraints: formats JPEG/JPG/PNG (no alpha), BMP, WEBP; width and height 240-8000 px; aspect ratio 1:8-8:1; file size <= 20MB.

video_urls · string[]

Public video URLs. image-to-video: optional 1 continuation clip at video_urls[0]. reference-to-video: optional 1-5 reference videos, with total image_urls + video_urls between 1 and 5. video-edit: required and must contain exactly 1 input video URL. Video constraints by mode: image-to-video continuation clip (video_urls[0]) must be mp4/mov, 2-10s, 240-4096 px, ratio 1:8-8:1, size <= 100MB; reference-to-video reference video must be mp4/mov, 1-30s, 240-4096 px, ratio 1:8-8:1, size <= 100MB; video-edit input video must be mp4/mov, 2-10s, 240-4096 px, ratio 1:8-8:1, size <= 100MB.

audio_urls · string[]

Public audio URLs. text-to-video: optional 1 background audio URL. image-to-video: optional 1 driving audio URL, and only when video_urls is not provided. reference-to-video: optional 1-5, aligned by slot order to reference items (image_urls first, then video_urls). Length can be less than the total reference count, but cannot exceed image_urls + video_urls. To skip a reference slot while keeping later alignment, pass empty string "" at that position. Non-empty values are mapped as per-reference voice. If no audio_urls are provided for reference-to-video, reference videos can keep their original voice by default. Audio constraints: formats wav/mp3, duration 2-30s, size <= 15MB. If audio is longer than duration, truncates to the first duration seconds; if shorter, the remaining video segment is silent.

audio_setting · string

video-edit only. Supported values: auto, origin. auto (default) lets decide whether to regenerate audio based on prompt sound instructions, otherwise it may keep original audio. origin forces original input audio and skips regeneration.

Response format

Submit task response

POST /generateTask/wan-2-7-video returns a task ID when the task is accepted:

code · integer

API status code. 200 means the task was accepted.

message · string

Human-readable status message.

data.taskId · string

Unique task identifier used with the status endpoint.

Status response fields

taskId · string

Unique task identifier.

state · string

Current task state: pending, processing, success, or failed.

resultJson · string

JSON string containing generated video URLs in resultUrls. Present when state is success.

failCode · string

Machine-readable failure code. Present when state is failed.

failMsg · string

Human-readable failure message. Present when state is failed.

createTime · integer

Task creation timestamp in Unix milliseconds.

completeTime · integer

Task completion timestamp in Unix milliseconds. Present after completion.

costTime · integer

Processing duration in milliseconds. Present after completion when upstream timing is available.

Webhook callback mode

Use callback mode when your backend should receive the final result automatically instead of polling.

curl -X POST "https://api.apixo.ai/api/v1/generateTask/wan-2-7-video" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "callback",
    "callback_url": "https://your-server.com/webhooks/apixo",
    "input": {
      "mode": "video-edit",
      "video_urls": [
        "https://example.com/input.mp4"
      ],
      "image_urls": [
        "https://example.com/reference.png"
      ],
      "prompt": "change the scene to watercolor style",
      "resolution": "1080p",
      "duration": 0,
      "audio_setting": "origin"
    }
  }'

See Webhooks for delivery requirements and retry behavior.

Billing

Wan 2.7 Video is billed per second. Unit price depends on both mode and resolution.

Mode	Resolution	APIXO official price
text-to-video	720p	$0.10 / second
text-to-video	1080p	$0.15 / second
image-to-video	720p	$0.10 / second
image-to-video	1080p	$0.15 / second
reference-to-video	720p	$0.15 / second
reference-to-video	1080p	$0.24 / second
video-edit	720p	$0.10 / second
video-edit	1080p	$0.15 / second

Billing formulas:

Mode	Formula
text-to-video	duration * unitPrice
image-to-video	duration * unitPrice
reference-to-video	duration * unitPrice
video-edit	billableSeconds * unitPrice

For video-edit, billing is based on both input and output duration:

• Input video duration must be between 2 and 10 seconds (inclusive).
• Billable input duration is capped at 10 seconds.
• If duration = 0, then billableSeconds = cappedInputDuration * 2.
• If duration = 2-10, then billableSeconds = cappedInputDuration + duration.
• Maximum billable seconds for video-edit is 20.

For current route and market comparison pricing, see Pricing.

Latency and polling

Wan 2.7 Video tasks are asynchronous. The backend does not provide a fixed public latency SLA; actual latency varies by mode, resolution, duration, prompt complexity, media fetch speed, and provider queue load.

Workflow	Recommended first poll	Poll interval
text-to-video	60s after task creation	10s
image-to-video	60s after task creation	10s
reference-to-video	60s-90s after task creation	10s-15s
video-edit	60s-90s after task creation	10s-15s

For production workloads, use callback mode to avoid frequent polling on long-running video jobs.

Rate limits and concurrency can vary by account, API key, and route. If you receive 429, slow down requests and retry with backoff. For account-level details, see System APIs.

Errors and troubleshooting

HTTP errors

Code	Meaning	What to do
400	Invalid request body, mode, parameter, or media URL shape	Fix the request before retrying
401	Missing or invalid API key	Check the Authorization header
402	Insufficient balance or quota	Add balance or switch account/key
403	Key or route cannot access the model	Check permissions and route strategy
429	Rate limit or concurrency limit reached	Retry with exponential backoff
500	Server error	Retry with backoff
502	Upstream provider error	Retry with backoff
504	Upstream timeout	Retry or use callback mode

Task failure codes

failCode is route-dependent and may come from mapped upstream failures. Common values include:

Fail code	Meaning	What to do
PromptInvalid	Prompt is invalid or rejected by provider	Rewrite prompt and retry
MissingParameter / BadRequest	Required fields are missing or malformed	Check mode-specific required fields
SensitiveContent / SensitiveContentDetected	Prompt or media failed safety checks	Change prompt or media
RateLimitExceeded	Provider or route rate limit was reached	Retry with backoff
Task TimeOut / Timeout	Upstream generation timed out	Retry or use callback mode
UnmappedUpstreamError	Upstream error could not be mapped to a known type	Retry with backoff or contact support with taskId

Parameter troubleshooting

• mode must be one of text-to-video, image-to-video, reference-to-video, video-edit.
• prompt is required for text-to-video, reference-to-video, and video-edit.
• ratio is not allowed in image-to-video.
• image-to-video requires image_urls or video_urls.
• If image-to-video includes video_urls, audio_urls is not allowed.
• reference-to-video requires at least one reference item across image_urls and video_urls, with total count <= 5.
• If reference-to-video includes audio_urls, it binds by slot order: image_urls first, then video_urls.
• If reference-to-video needs to skip a reference slot while keeping later audio aligned, pass empty string "" in that slot.
• If reference-to-video includes audio_urls, its length cannot exceed image_urls + video_urls.
• video-edit requires exactly one video_urls item.
• video-edit requires a non-empty prompt.
• video-edit input clip duration must be 2-10 seconds.

See Error Codes for the full error reference.

Related links

• Generation API Overview
• Video Models
• Generate Task
• Status Task
• Webhooks
• Error Codes
• Parameter Specification
• Routing Strategies
• Try Wan 2.7 Video in the APIXO Playground
• Pricing