Vidu-reference-to-video-q3-mix

The balanced optimization version of Shengshu Technology's Vidu Q3 series reference-to-video model, delivering strong performance in visual quality and motion dynamics. Supports intelligent scene cutting and audio-visual synchronized generation. Generates 1-16 second videos at 720p and 1080p resolutions. Note: Current version does not support subject library calls. Ideal for creative scenarios prioritizing balanced image quality and motion performance without subject library requirements.

Request

Authorization

Header Params

Body Params application/jsonRequired

Example

{
    "prompt": "P1: Close-up on a city street. A young Asian man in a brown jacket and white t-shirt (wearing a backpack) walks along a busy pedestrian street, frowning and looking at his phone screen, searching for directions. Blurred pedestrians and shop signs (e.g., \"Cafe\") in the background.  P2: Close-up. A person's hand holds a smartphone. The phone screen clearly displays a city map navigation interface, with routes and Chinese place names visible.  P3: Medium close-up on a city street. The same Asian man stands between modern skyscrapers, looking up at blue street signs (e.g., \"Financial Center\") and the sky.  P4: Medium close-up at a subway station entrance. The Asian man stands in front of a subway entrance (marked \"Line 4 Entrance\"), facing a large subway line map, pointing with his finger at a point on the map. Other pedestrians and fare gates are around.  P5: Medium shot in a city alleyway. The man walks through an old, graffiti-covered urban back alley. Surrounded by traditional Chinese small shops (like \"Hardware Store\"), the ground is wet.  P6: Medium shot at a street food stall. The Asian man stands in front of a roadside stall (marked \"Fried Dough Sticks\"), talking to an elderly vendor. The old vendor points in a certain direction.  P7: Medium close-up on a nighttime city street. The man walks along a brightly lit night street, lifting his left hand to check his watch. Various neon signs (e.g., \"Night Market,\" \"Convenience Store\") are in the background.  P8: Medium shot outside a vintage cafe (night). The man stands outside a vintage-style cafe named \"Time Cafe,\" talking to another man. The cafe interior is warmly lit.  P9: Close-up inside a cafe (night). The same man sits inside the cafe, a latte on the table. He is facing a laptop, smiling as he types. Blurred city night view visible outside the window.",
    "imageUrls": [
        "https://www.runninghub.cn/view?filename=214f824f1ab31efbfff586db26c3a800e701e70dd8f350f7d40b18b81583ac88.jpg&type=input&subfolder=&Rh-Comfy-Auth=eyJ1c2VySWQiOiIzZjY1MTNlNWEwNjY1N2I4OGYyNjU5NTEzYmU3ZDM0YyIsInNpZ25FeHBpcmUiOjE3NzU2NTE4NjYzNzQsInRzIjoxNzc1MDQ3MDY2Mzc0LCJzaWduIjoiZDA4NzlhYjJjOTlkMWFjMDUzMDNiYjA3ODNlNjg3MjgifQ==&Rh-Identify=3f6513e5a06657b88f2659513be7d34c&rand=0.883905726617537"
    ],
    "duration": 5,
    "resolution": "720p",
    "aspectRatio": "16:9",
    "audio": "true"
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://www.runninghub.ai/openapi/v2/vidu/reference-to-video-q3-mix' \
--header 'Authorization: Bearer [Your API KEY]' \
--header 'Authorization: Bearer [Your API KEY]' \
--header 'Content-Type: application/json' \
--data-raw '{
    "prompt": "P1: Close-up on a city street. A young Asian man in a brown jacket and white t-shirt (wearing a backpack) walks along a busy pedestrian street, frowning and looking at his phone screen, searching for directions. Blurred pedestrians and shop signs (e.g., \"Cafe\") in the background.  P2: Close-up. A person'\''s hand holds a smartphone. The phone screen clearly displays a city map navigation interface, with routes and Chinese place names visible.  P3: Medium close-up on a city street. The same Asian man stands between modern skyscrapers, looking up at blue street signs (e.g., \"Financial Center\") and the sky.  P4: Medium close-up at a subway station entrance. The Asian man stands in front of a subway entrance (marked \"Line 4 Entrance\"), facing a large subway line map, pointing with his finger at a point on the map. Other pedestrians and fare gates are around.  P5: Medium shot in a city alleyway. The man walks through an old, graffiti-covered urban back alley. Surrounded by traditional Chinese small shops (like \"Hardware Store\"), the ground is wet.  P6: Medium shot at a street food stall. The Asian man stands in front of a roadside stall (marked \"Fried Dough Sticks\"), talking to an elderly vendor. The old vendor points in a certain direction.  P7: Medium close-up on a nighttime city street. The man walks along a brightly lit night street, lifting his left hand to check his watch. Various neon signs (e.g., \"Night Market,\" \"Convenience Store\") are in the background.  P8: Medium shot outside a vintage cafe (night). The man stands outside a vintage-style cafe named \"Time Cafe,\" talking to another man. The cafe interior is warmly lit.  P9: Close-up inside a cafe (night). The same man sits inside the cafe, a latte on the table. He is facing a laptop, smiling as he types. Blurred city night view visible outside the window.",
    "imageUrls": [
        "https://www.runninghub.cn/view?filename=214f824f1ab31efbfff586db26c3a800e701e70dd8f350f7d40b18b81583ac88.jpg&type=input&subfolder=&Rh-Comfy-Auth=eyJ1c2VySWQiOiIzZjY1MTNlNWEwNjY1N2I4OGYyNjU5NTEzYmU3ZDM0YyIsInNpZ25FeHBpcmUiOjE3NzU2NTE4NjYzNzQsInRzIjoxNzc1MDQ3MDY2Mzc0LCJzaWduIjoiZDA4NzlhYjJjOTlkMWFjMDUzMDNiYjA3ODNlNjg3MjgifQ==&Rh-Identify=3f6513e5a06657b88f2659513be7d34c&rand=0.883905726617537"
    ],
    "duration": 5,
    "resolution": "720p",
    "aspectRatio": "16:9",
    "audio": "true"
}'

Responses

🟢200成功

application/json

Task result query endpoint: /openapi/v2/query

Body

Examples

Submit response exampleTask success response example

{
    "taskId": "2013508786110730241",
    "status": "RUNNING",
    "errorCode": "",
    "errorMessage": "",
    "results": null,
    "clientId": "f828b9af25161bc066ef152db7b29ccc",
    "promptTips": "{\"result\": true, \"error\": null, \"outputs_to_execute\": [\"4\"], \"node_errors\": {}}"
}