Wan 3.0 Video API
Alibaba Wanxiang 3.0 all-in-one reference video model. Text, first frame, first + last frame, multi-modal reference (images, video, audio) and document or web page references, 2-30 seconds at 480p/720p/1080p.
- Text to video
- Image to video
Starting at
$0.047
per second of output
Wan 3.0 Video playground
Input
Text description. Required unless at least one media input is supplied. In reference mode you can address assets positionally with "图N / 视频N / 音频N" labels; indices run per media type.
Image URLs without explicit roles. In the frame family the first item is the start frame and an optional second is the end frame; in the reference family every item is a reference image (up to 10). For Wan 3.0, Prime and Pro, each image must be at least 240x240 pixels.
Reference videos, up to 5 clips of 1-15s each (total <= 15s). Input video duration plus the requested output duration must stay within 30 seconds.
Reference audio, up to 5 clips of 1-15s each (total <= 15s).
Reference document driving the generation (docx/xlsx/pptx/pdf/txt/md and similar, <=100MB, <=50 pages). Cannot be combined with linkUrl. A prompt is optional when this is supplied.
Public, login-free web page to turn into a video. Cannot be combined with fileUrl.
Runs are charged to your balance.
Output

Output will appear here.
Check the price and run the model. Your result will appear here; browser use does not need an API key.
Wan 3.0 Video API pricing
| Option | Price per second |
|---|---|
| 480p | $0.047 |
| 720p | $0.094 |
| 1080p | $0.188 |
Quick start
Full API reference →1curl -X POST https://api.api-stock.com/api/v1/generation/create \2 -H "Authorization: Bearer $API_STOCK_KEY" \3 -H "Content-Type: application/json" \4 -d '{5 "model": "wan3.0-video",6 "input": {7 "size": "adaptive",8 "resolution": "480p",9 "duration": 5,10 "audio": true11 }12 }'
Parameters
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
prompt | string | No | — | Text description. Required unless at least one media input is supplied. In reference mode you can address assets positionally with "图N / 视频N / 音频N" labels; indices run per media type. |
imageUrls | string[] | No | — | Image URLs without explicit roles. In the frame family the first item is the start frame and an optional second is the end frame; in the reference family every item is a reference image (up to 10). For Wan 3.0, Prime and Pro, each image must be at least 240x240 pixels. |
imageWithRoles | object[] | No | — | Images with explicit roles. Use this when you need first/last frame or reference images stated outright rather than inferred from order. |
videoUrls | string[] | No | — | Reference videos, up to 5 clips of 1-15s each (total <= 15s). Input video duration plus the requested output duration must stay within 30 seconds. |
audioUrls | string[] | No | — | Reference audio, up to 5 clips of 1-15s each (total <= 15s). |
fileUrl | string | No | — | Reference document driving the generation (docx/xlsx/pptx/pdf/txt/md and similar, <=100MB, <=50 pages). Cannot be combined with linkUrl. A prompt is optional when this is supplied. |
linkUrl | string | No | — | Public, login-free web page to turn into a video. Cannot be combined with fileUrl. |
generationType | string | No | — | How bare imageUrls are classified: frame (first/last frame) or reference. Omit to let it be inferred — reference when the request already carries reference-family inputs, frame otherwise.One of: frame, reference |
size | string | No | adaptive | Aspect ratio. `adaptive` lets the model pick one from the inputs.One of: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16 |
resolution | string | No | 480p | Output resolution. Each tier is billed at its own per-second rate, so 1080p costs four times 480p for the same duration.One of: 480p, 720p, 1080p |
duration | number | No | 5 | Duration in seconds (2-30). Billing is per second at the resolution's rate. With reference videos, total input video duration plus this value must be at most 30 seconds. |
audio | boolean | No | true | Include an audio track in the output. |
seed | number | No | — | Random seed for controlling randomness |
Frequently asked questions
How much does Wan 3.0 Video cost?
Wan 3.0 Video starts at $0.047 per second of output on API Stock. The final charge depends on the selected settings.
When was Wan 3.0 Video released?
Wan 3.0 Video was released on August 24, 2026 and is available through API Stock.
How do I get an API key for Wan 3.0 Video?
Create an account and API key on API Stock, then fund your balance. The same key works with Wan 3.0 Video and other catalog models.
How do I call the Wan 3.0 Video API?
POST to https://api.api-stock.com/api/v1/generation/create with your API key and model "wan3.0-video". Use the returned task id for polling or receive a webhook.
What input parameters does Wan 3.0 Video accept?
Wan 3.0 Video accepts 13 input parameters. See their types, defaults and allowed values above.
What happens if a Wan 3.0 Video request fails?
API Stock automatically retries the request on a reserve provider. If every provider fails, the charge is refunded to your balance.
Can I use Wan 3.0 Video with the same API key as other models?
Yes. Use your API Stock API key and balance for Wan 3.0 Video and other catalog models.
Other Video models
- PixVerse V6PixVerse$0.024/ generation
- Seedance 2.0 MiniByteDance$0.025/ second
- Gemini Omni 1.2 FlashGoogle$0.029/ second
- Seedance 2.0 FastByteDance$0.057/ second
- Seedance 2.5 FlashByteDance$0.063/ second
- Kling 3.0 Motion ControlKling$0.065/ second
- Seedance 1.5 ProByteDance$0.065/ generation
- Wan 3.0 Video PrimeAlibaba$0.074/ second
- Wan 2.7 VideoAlibaba$0.086/ second
Read next
Guides
- Text to videoWhich text-to-video model fits the shot you have in mind, prompts that actually work, and the full API run from create to finished file.8 min read
- Image to videoTurning a still into a shot — which model to animate it with, first and last frame control, motion transfer, and the prompts that keep the subject intact.7 min read
- Give your coding agent image, video and music toolsConnect Claude Code, Codex, OpenCode or OpenClaw to API Stock and use generated media in a page, app or content draft.3 min read

