> ## Documentation Index > Fetch the complete documentation index at: https://docs.aiid.edu.kg/llms.txt > Use this file to discover all available pages before exploring further. # Kling All-in-One Reference > This series supports OpenAI video generation formats (see the link [Video generation Sora compatible formats](/api-reference/en/video-generation-sora-compatible-format/create-video)). Kling 4.0 native task API. Authenticate using this site's Bearer API Key. The request body does not require model; settings.resolution=720p uses kling-4.0, while 1080p uses kling-4.0-1080p, with 1080p as the default. After successful creation, use the returned data.id to call GET /tasks?task_ids=task ID; separate multiple IDs with commas. The returned data is an array, and the result video is available at data[].outputs[].url. Currently supported generation scenarios include prompts, first_frame, last_frame, refer_image, and feature_video. middle_frame, base_video, element, voice, audio references, callbacks, and custom external task IDs are not supported. Non-empty options.callback_url, non-empty options.external_task_id, non-empty contents.id, and other unsupported fields will return 400. options can be omitted or passed as an empty object; watermark_info.enabled=false can also be passed. Enabling watermarks is not currently supported. Supported durations range from 4 to 30 seconds, with a default of 10 seconds; 3 seconds is not supported. For text-to-video and reference scenarios, the default aspect ratio is 16:9; explicitly specifying adaptive will return 400. For first- and last-frame scenarios, explicitly specifying settings.aspect_ratio is not accepted; the aspect ratio is determined by the dimensions of the first frame image. For general-purpose reference, `contents` must include a non-empty `prompt` and at least one supported media item. For legacy first/last-frame scenarios, you can provide `first_frame`, `last_frame`, or both, but processing support depends on the capabilities of the currently available model. Requests with incompatible capabilities will be rejected; frame images will not be silently converted to regular reference images. For new first/last-frame integrations, we recommend using the image-to-video endpoint. First/last frames cannot be used together with `refer_image` reference images or `feature_video` reference videos; mixing them returns a 400 error. In reference scenarios, `refer_image` reference images and `feature_video` reference videos can be combined. `first_frame`, `last_frame`, and `refer_image` can each contain at most one item, while `feature_video` can contain up to five items. Multi-reference image scenarios that rely on ID references are not currently supported. ## OpenAPI ````yaml api-reference/en/openapi.json POST /omni-video/kling-4.0 openapi: 3.0.0 info: title: Overseas Expansion Camp API Reference Documentation version: 1.0.0 description: Public AI Gateway API Reference servers: - url: https://api.aiid.edu.kg security: - BearerAuth: [] tags: - name: OpenAI format (Chat) - name: OpenAI Format (Responses) - name: Gemini Format for Image Generation - name: Image generationOpenAI DALL-E format - name: List models - name: Video generationHappyHorse and Wan - name: Video generation Kling format - name: Video generation Omni and Veo formats - name: Video generation Seedance - name: Video generation Sora compatible formats - name: Video generation Vidu - name: Music generation task format paths: /omni-video/kling-4.0: post: tags: - Video generation Kling format summary: Kling All-in-One Reference description: >- This series supports OpenAI video generation formats (see the link [Video generation Sora compatible formats](/api-reference/en/video-generation-sora-compatible-format/create-video)). Kling 4.0 native task API. Authenticate using this site's Bearer API Key. The request body does not require model; settings.resolution=720p uses kling-4.0, while 1080p uses kling-4.0-1080p, with 1080p as the default. After successful creation, use the returned data.id to call GET /tasks?task_ids=task ID; separate multiple IDs with commas. The returned data is an array, and the result video is available at data[].outputs[].url. Currently supported generation scenarios include prompts, first_frame, last_frame, refer_image, and feature_video. middle_frame, base_video, element, voice, audio references, callbacks, and custom external task IDs are not supported. Non-empty options.callback_url, non-empty options.external_task_id, non-empty contents.id, and other unsupported fields will return 400. options can be omitted or passed as an empty object; watermark_info.enabled=false can also be passed. Enabling watermarks is not currently supported. Supported durations range from 4 to 30 seconds, with a default of 10 seconds; 3 seconds is not supported. For text-to-video and reference scenarios, the default aspect ratio is 16:9; explicitly specifying adaptive will return 400. For first- and last-frame scenarios, explicitly specifying settings.aspect_ratio is not accepted; the aspect ratio is determined by the dimensions of the first frame image. For general-purpose reference, `contents` must include a non-empty `prompt` and at least one supported media item. For legacy first/last-frame scenarios, you can provide `first_frame`, `last_frame`, or both, but processing support depends on the capabilities of the currently available model. Requests with incompatible capabilities will be rejected; frame images will not be silently converted to regular reference images. For new first/last-frame integrations, we recommend using the image-to-video endpoint. First/last frames cannot be used together with `refer_image` reference images or `feature_video` reference videos; mixing them returns a 400 error. In reference scenarios, `refer_image` reference images and `feature_video` reference videos can be combined. `first_frame`, `last_frame`, and `refer_image` can each contain at most one item, while `feature_video` can contain up to five items. Multi-reference image scenarios that rely on ID references are not currently supported. operationId: create_kling4_native_omni_video requestBody: required: true content: application/json: schema: type: object additionalProperties: false required: - contents properties: settings: type: object additionalProperties: false properties: resolution: type: string enum: - 720p - 1080p default: 1080p duration: type: integer minimum: 4 maximum: 30 default: 10 aspect_ratio: type: string enum: - '21:9' - '16:9' - '9:16' - '1:1' options: type: object additionalProperties: false description: >- You can omit it, pass an empty object, or explicitly disable the watermark. properties: watermark_info: type: object additionalProperties: false required: - enabled properties: enabled: type: boolean enum: - false contents: type: array minItems: 2 description: >- A non-empty prompt and supported media must both be provided. Image-to-video requests must include first_frame. Do not use first_frame/last_frame together with refer_image/feature_video; otherwise, a 400 error is returned. See the API documentation for media quantity limits. Legacy first-and-last-frame requests are validated based on the capabilities of currently available models; for new first-and-last-frame integrations, the image-to-video endpoint is recommended. items: oneOf: - type: object additionalProperties: false required: - type - text properties: type: type: string enum: - prompt text: type: string minLength: 1 - type: object additionalProperties: false required: - type - url properties: type: type: string enum: - first_frame - last_frame - refer_image - feature_video url: type: string format: uri pattern: ^https?:// not: items: properties: type: enum: - prompt oneOf: - description: >- Legacy first/last-frame compatibility scenarios: `first_frame`, `last_frame`, or both, subject to the capabilities of currently available models. Do not combine with reference assets; omit `aspect_ratio`, and the aspect ratio follows the first frame. properties: contents: items: properties: type: enum: - prompt - first_frame - last_frame settings: not: required: - aspect_ratio - description: >- Example scenario: `refer_image` and `feature_video` can be used together. The default aspect ratio is 16:9. properties: contents: items: properties: type: enum: - prompt - refer_image - feature_video settings: properties: aspect_ratio: type: string enum: - '21:9' - '16:9' - '9:16' - '1:1' default: '16:9' examples: kling4_native_720p: summary: Kling 720p value: settings: resolution: 720p duration: 10 aspect_ratio: '16:9' options: {} contents: - type: prompt text: A cinematic product shot with natural motion - type: refer_image url: https://example.com/reference.jpg - type: feature_video url: https://example.com/motion.mp4 kling4_native_1080p: summary: Kling 1080p value: settings: resolution: 1080p duration: 10 aspect_ratio: '16:9' options: {} contents: - type: prompt text: A cinematic product shot with natural motion - type: refer_image url: https://example.com/reference.jpg - type: feature_video url: https://example.com/motion.mp4 responses: '200': description: Success content: application/json: schema: type: object required: - code - message - request_id - data properties: code: type: integer example: 0 message: type: string request_id: type: string data: type: object properties: id: type: string description: Task ID on this site. Pass it as-is when querying. status: type: string enum: - submitted - processing - succeeded - failed create_time: type: integer format: int64 description: Creation time, in Unix milliseconds. update_time: type: integer format: int64 description: Update time, in Unix milliseconds. external_id: type: string maxLength: 0 examples: success: summary: Success value: code: 0 message: SUCCEED request_id: request_example data: id: task_example status: submitted create_time: 1792710000000 update_time: 1792710000000 external_id: '' '400': description: The request parameters are invalid or contain unsupported fields. components: securitySchemes: BearerAuth: type: http scheme: bearer description: |- Use Bearer Token authentication. Format: `Authorization: Bearer sk-xxxxxx` ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.