> ## Documentation Index > Fetch the complete documentation index at: https://docs.geekai.co/llms.txt > Use this file to discover all available pages before exploring further. # Create Task Note: For video model name settings, refer to the [System Supported Video Model List](https://geekai.dev/models), currently supported video generation models are `cogvideox-flash` (free), `cogvideox`, and `cogvideox-2`. The response data format is fully compatible with OpenAI, returning the binary data of the video file. Different AI platforms support different video sizes for their video generation models. Please refer to the table below for the specific supported sizes:
Platform Model Supported Sizes
Zhipu CogVideoX Series 720x480, 1024x1024, 1280x960, 960x1280, 1920x1080, 1080x1920, 2048x1080, 3840x2160 (By default, the shorter side is 1080, and the longer side is scaled proportionally)
### cURL Request Example ```bash theme={null} curl --location 'https://geekai.dev/api/v1/videos/generations' \ --header 'Content-Type: application/json' \ --header 'Authorization: Bearer {YOUR_GEEKAI_API_KEY}' \ --data '{ "model":"cogvideox-flash", "prompt":"Kittens and puppies are playing in the grass", "async": false }' ``` ## OpenAPI ````yaml openapi.yaml POST /videos/generations openapi: 3.1.0 info: title: 极客智坊 API文档 description: 极客智坊 API 消息接口规范文档 version: 1.0.0 servers: - url: https://geekai.co/api/v1 description: 国内代理入口 - url: https://geekai.dev/api/v1 description: 海外代理入口 security: - bearerAuth: [] - apiKeyAuth: [] paths: /videos/generations: post: tags: - Video summary: 视频生成接口 description: 基于文本提示/图片生成视频的接口，支持多种视频生成模型和参数配置。 operationId: createVideo requestBody: required: true content: application/json: schema: type: object required: - model - prompt properties: model: type: string description: 视频模型 example: cogvideox-flash prompt: type: string description: 文本提示 example: 彼得兔开小汽车，游走在马路上，脸上的表情充满开心喜悦 negative_prompt: type: string description: 反向提示词，用来描述不希望在视频中看到的内容，可以对视频进行限制，仅支持该字段模型生效 image: oneOf: - type: string description: 视频首帧图片URL - type: array description: 参考图片URL（仅适用于旧模型，新模型采用images字段替代，具体以模型示例文档为准） items: type: string image_tail: type: string description: 用于视频尾帧的图片URL，仅支持该字段模型生效 images: type: array description: 参考图片URL items: type: string video: oneOf: - type: string description: 待编辑视频URL - type: array description: 参考视频URL items: type: string audio: oneOf: - type: string description: 音频URL - type: array description: 参考音频URL items: type: string with_audio: type: boolean description: 是否带音效，仅支持音效的模型生效 default: false size: type: string description: >- 视频尺寸，尺寸越大价格可能越高，不同平台支持的尺寸参数不同，请参考[创建视频API](https://docs.geekai.co/cn/api/video/generations)设置 example: 720x720 resolution: type: string description: >- 视频分辨率，不同平台支持的分辨率参数不同，请参考[创建视频API](https://docs.geekai.co/cn/api/video/generations)设置 example: 720p aspect_ratio: type: string description: 视频宽高比，默认为16:9，仅支持该字段模型生效 example: '16:9' quality: type: string description: >- 视频质量，仅支持该字段模型生效，智谱清言支持 speed、quality 两个配置，可灵 AI 和 Google Veo3 支持 std、pro 两个配置，默认 std，kling-v3 开始支持 4k example: speed duration: type: integer description: 视频时长(秒)，默认5秒(海螺默认6s） default: 5 fps: type: integer description: 视频帧率，仅支持该字段模型生效 enum: - 24 - 30 - 60 default: 30 watermark: type: boolean description: 是否添加AI生成水印，默认为false，仅支持该字段模型生效 default: false async: type: boolean description: >- 是否异步生成，默认false，即同步等待视频生成成功后返回生成结果，如果异步需要通过调用视频获取接口获取生成结果，推荐使用异步 default: true extra_body: type: object description: 额外参数配置项，以适配不同视频模型的多样化配置 properties: draft: type: boolean description: 是否生成样片，仅支持该字段模型生效 default: false return_last_frame: type: boolean description: 是否返回最后一帧图片，仅支持该字段模型生效 default: false movement_amplitude: type: string description: 运动幅度，仅支持该字段模型生效 camera_strength: type: string description: 镜头运动强度，仅支持该字段模型生效 image_list: type: array description: 用于视频生成的图片列表，支持图片URL/Base64编码，仅支持该字段模型生效 items: type: object properties: image_url: type: string description: 图片URL/Base64编码数据 type: type: string description: 帧类型，first_frame：首帧，end_frame：尾帧 video_list: type: array description: 用于视频生成的视频列表，支持视频URL，仅支持该字段模型生效 items: type: object properties: video_url: type: string description: 参考视频URL refer_type: type: string description: 参考类型：feature（特征参考视频）或 base（待编辑视频） keep_original_sound: type: string description: 是否保留原始音频，yes 保留，no 不保留 element_list: type: array description: 参考主体列表，基于主体库中主体的 ID 配置，仅支持该字段模型生效 items: type: object properties: element_id: type: long description: 主体ID voice_list: type: array description: 生成视频时所引用的音色的列表，仅支持该字段模型生效 items: type: object properties: voice_id: type: string description: 音色ID multi_shot: type: boolean description: 是否开启多镜头模式，仅支持该字段模型生效 shot_type: type: string description: 分镜方式，仅支持该字段模型生效 multi_prompt: type: array description: 多镜头模式下的分镜提示词列表，仅支持该字段模型生效 items: type: object properties: index: type: integer description: 分镜索引，从0开始 prompt: type: string description: 分镜提示词 duration: type: string description: 分镜时长(秒) camera_control: type: object description: 镜头控制，仅支持该字段模型生效 properties: type: type: string description: 预定义的运镜类型 config: type: object description: 运镜类型的配置参数 properties: horizontal: type: float description: 水平运镜 vertical: type: float description: 垂直运镜 pan: type: float description: 平移运镜 tilt: type: float description: 俯仰运镜 roll: type: float description: 滚动运镜 zoom: type: float description: 缩放运镜 static_mask: type: string description: 静态笔刷，仅支持该字段模型生效 dynamic_masks: type: array description: 动态笔刷，仅支持该字段模型生效 items: type: object properties: mask: type: string description: 运动笔刷涂抹的 mask 图片 trajectories: type: array description: 运动笔刷的轨迹点列表 items: type: object properties: x: type: number description: 轨迹点的 x 坐标 'y': type: number description: 轨迹点的 y 坐标 real_person_mode: type: boolean description: 是否开启真人模式 retries: type: integer description: 自动重试次数 default: 0 responses: '200': description: 成功响应 content: application/json: schema: type: object required: - model - task_id - task_status properties: model: type: string description: 使用的视频模型 example: cogvideox-flash task_id: type: string description: 任务ID format: uuid task_status: type: string description: 任务状态 enum: - pending - running - succeed - failed example: running video_result: type: object description: 视频生成结果(仅在同步模式且生成成功时返回) properties: id: type: string description: 视频ID url: type: string format: uri description: 视频URL cover_image_url: type: string format: uri description: 封面图片URL duration: type: number description: 实际视频时长(秒) revised_prompt: type: string description: 增强后的提示文本 error: type: object description: 错误信息(仅在生成失败时返回) properties: code: type: string description: 错误代码 message: type: string description: 错误描述 '400': $ref: '#/components/responses/ValidationError' '401': $ref: '#/components/responses/Unauthorized' '413': description: 提示文本过长或图片过大 content: application/json: schema: $ref: '#/components/schemas/Error' '500': $ref: '#/components/responses/StandardError' components: responses: ValidationError: description: 参数验证错误 content: application/json: schema: type: object properties: code: type: string example: validation_error message: type: string example: 参数验证失败 details: type: object additionalProperties: type: string example: field: 错误描述 Unauthorized: description: 未授权 content: application/json: schema: type: object properties: code: type: string example: unauthorized message: type: string example: Invalid API key or token StandardError: description: 标准错误响应 content: application/json: schema: $ref: '#/components/schemas/Error' schemas: Error: type: object required: - code - message properties: code: type: string description: 错误代码 example: invalid_request message: type: string description: 错误信息描述 example: 请求参数不合法 securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: JWT description: API认证token apiKeyAuth: type: apiKey in: header name: GeekAI-API-Key description: API密钥认证 ````