Continuously modify and generate video

Given a prompt, the model will return one or more predicted completions and can also provide the probabilities of alternative tokens at each position.
Create completions for the provided prompt and parameters.
Official documentation: https://platform.openai.com/docs/api-reference/chat/create

Request

Header Params

Body Params application/json

Example

{
    "model": "sora-2",
    "messages": [
        {
            "role": "user",
            "content": "生成一段高清的，牛马跳舞的视频"
        },
        {
            "role": "assistant",
            "content": "```json\n{\n  \"prompt\": \"生成一段高清的，牛马跳舞的视频\",\n  \"orientation\": \"portrait\"\n}\n```\n\n> ID: `task_01k7dzptgwf2z87wj2c3t36qk0`\n>[数据预览](https://asyncdata.net/web/task_01k7dzptgwf2z87wj2c3t36qk0) | [原始数据](https://asyncdata.net/source/task_01k7dzptgwf2z87wj2c3t36qk0)\n> 排队中......\n> 生成中.\n\n>🏃‍ 进度 76..\n\n> 生成完成 ✅\n> sid: s_68ec8accd80c8191ae1fc957af78caaa\n\n![https://filesystem.site/cdn/20251013/929d094fcd1844c3ee38be9476fd03.webp](https://filesystem.site/cdn/20251013/929d094fcd1844c3ee38be9476fd03.webp)\n[在线播放▶️](https://filesystem.site/cdn/20251013/5781a601c5ce5e216e914fef119c99.mp4)"
        },
        {
            "role": "user",
            "content": "让牛的头上顶着光圈，马的脚下踩着祥云"
        }
    ],
    "stream": true
}

Continuously modify and generate video

Request

Responses