> ## Documentation Index > Fetch the complete documentation index at: https://docs.omniall.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create video > Create a Kling video task. **Path:** `POST https://api.omniall.ai/v1/videos` Text-to-video vs image-to-video: if `image` / `images` / `input_reference` / `metadata.last_frame_*` / reference video is present → image-to-video; else text-to-video. Auth: `Authorization: Bearer sk-...` ## OpenAPI ````yaml openapi/kling-video.yaml POST /v1/videos openapi: 3.1.0 info: title: Omniall AI — Kling video version: 1.0.0 description: >- Kling async video via OpenAI-style Video API. **Create:** `POST https://api.omniall.ai/v1/videos` **Poll:** `GET /v1/videos/{task_id}` · **Download:** `GET /v1/videos/{task_id}/content` Model aliases (case-insensitive). See [Model Square](https://omniall.ai/pricing). **Duration priority:** top-level `seconds` > top-level `duration` > `metadata.seconds` / `metadata.duration` / `metadata.video_duration` > default **5**. **Resolution priority:** `metadata.output_config.resolution` > `metadata.resolution` > top-level `size` > default **720P**. Auth: `Authorization: Bearer sk-...` servers: - url: https://api.omniall.ai security: [] paths: /v1/videos: post: tags: - Kling summary: Create video description: >- Create a Kling video task. **Path:** `POST https://api.omniall.ai/v1/videos` Text-to-video vs image-to-video: if `image` / `images` / `input_reference` / `metadata.last_frame_*` / reference video is present → image-to-video; else text-to-video. Auth: `Authorization: Bearer sk-...` operationId: klingCreateVideo parameters: - name: Authorization in: header required: true schema: type: string example: Bearer sk-... description: Bearer API key, e.g. `Bearer sk-...`. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/KlingCreateRequest' examples: t2v_720p: summary: Text-to-video (720P) value: model: kling-2.6 prompt: Cyberpunk city night, slow camera push-in, cinematic seconds: '5' metadata: output_config: resolution: 720P aspect_ratio: '16:9' t2v_1080p_mute: summary: Text-to-video 1080P (mute) value: model: kling-3.0 prompt: Eagle circling above snowy mountain peaks seconds: '10' metadata: output_config: duration: 10 resolution: 1080P aspect_ratio: '16:9' audio_generation: Disabled t2v_with_audio: summary: Text-to-video with audio value: model: kling-2.5 prompt: Street band playing jazz, crowd watching seconds: '10' metadata: output_config: duration: 10 resolution: 1080P aspect_ratio: '16:9' audio_generation: Enabled i2v_single: summary: Single image-to-video value: model: Kling-2.5 prompt: Make the character walk forward and smile, natural light image: https://example.com/character.png seconds: '5' metadata: output_config: duration: 5 resolution: 720P aspect_ratio: '9:16' audio_generation: Disabled i2v_multi: summary: Multi-image reference value: model: Kling-3.0-Omni prompt: >- Image 1 is character Chen Zhou; image 2 is a dark void system space; camera moves from far to near images: - https://example.com/role.png - https://example.com/scene.png seconds: '15' metadata: output_config: duration: 15 resolution: 1080P aspect_ratio: '9:16' portrait_size: summary: Portrait via size value: model: kling-3.0-omni prompt: Character slowly turns in front of a blue panel seconds: '15' size: 720x1280 metadata: output_config: duration: 15 resolution: 720P aspect_ratio: '9:16' audio_generation: Enabled first_last_frame: summary: First / last frame value: model: Kling-3.0 prompt: Smooth transition from the first frame to the last frame image: https://example.com/first.png seconds: '8' metadata: last_frame_url: https://example.com/last.png output_config: duration: 8 resolution: 1080P aspect_ratio: '16:9' audio_generation: Enabled motion_std: summary: Motion control (std) value: model: Kling-2.6 prompt: Character follows the motion in the reference video seconds: '5' metadata: scene_type: motion_control motion_level: std video_url: https://example.com/motion-ref.mp4 output_config: duration: 5 resolution: 720P aspect_ratio: '16:9' motion_pro: summary: Motion control (pro) value: model: Kling-2.6 prompt: Dancer recreates advanced moves from the reference video seconds: '10' metadata: scene_type: motion_control motion_level: pro video_url: https://example.com/dance-ref.mp4 output_config: duration: 10 resolution: 1080P aspect_ratio: '16:9' avatar: summary: Digital human (avatar) value: model: kling-avatar-1080p prompt: Digital human presenting today's news, natural expression image: https://example.com/avatar-face.png seconds: '5' metadata: scene_type: avatar_i2v output_config: duration: 5 resolution: 720P aspect_ratio: '9:16' lip_sync: summary: Lip sync value: model: kling-identify-face prompt: Character lip-sync with the audio image: https://example.com/face.png seconds: '4' metadata: scene_type: lip_sync output_config: duration: 4 ref_video_file_infos: summary: Reference video (file_infos) value: model: Kling-3.0-Omni prompt: Generate a shot in the same style as the reference video seconds: '5' metadata: file_infos: - type: Url category: Video url: https://example.com/ref-video.mp4 reference_type: feature output_config: duration: 5 resolution: 720P aspect_ratio: '16:9' audio_generation: Disabled motion_file_infos: summary: Motion control + file_infos value: model: Kling-3.0-Omni prompt: Make the character move following the reference video seconds: '5' metadata: scene_type: motion_control motion_level: pro file_infos: - type: Url category: Video url: https://example.com/motion-ref.mp4 output_config: duration: 5 resolution: 720P aspect_ratio: '16:9' multi_shot_ext_info: summary: Multi-shot (ext_info) value: model: Kling-3.0 seconds: '5' metadata: file_infos: - type: Url category: Image url: https://example.com/ref-image.png output_config: duration: 5 resolution: 720P aspect_ratio: '16:9' input_compliance_check: Disabled ext_info: >- {"AdditionalParameters":"{\"multi_prompt\":[{\"index\":1,\"prompt\":\"Woman dancing\",\"duration\":\"2\"},{\"index\":2,\"prompt\":\"Woman jumps up\",\"duration\":\"3\"}],\"shot_type\":\"customize\",\"multi_shot\":true}"} content_array: summary: content array style value: model: kling-3.0-omni seconds: '10' content: - type: text text: Character slowly turns in front of a system panel images: - https://example.com/ref.png metadata: output_config: duration: 10 resolution: 720P aspect_ratio: '9:16' audio_generation: Enabled responses: '200': description: Task accepted content: application/json: schema: $ref: '#/components/schemas/KlingTask' example: id: task_xxxxxxxx task_id: task_xxxxxxxx object: video model: kling-2.5-720p status: queued progress: 0 created_at: 1780891541 components: schemas: KlingCreateRequest: type: object required: - model description: Kling create-video body. properties: model: type: string description: >- Model alias (case-insensitive). Generation: `kling-1.0` … `kling-3.0-omni`, `kling-o1`. Specialty: `kling-identify-face`, `kling-avatar-720p`, `kling-avatar-1080p`. example: kling-2.5 prompt: type: string description: Text prompt (required for most scenarios). seconds: type: string description: >- Duration in seconds as string, e.g. `"5"`, `"10"`. Highest priority for duration. example: '5' duration: oneOf: - type: number - type: string description: Duration in seconds; ignored if top-level `seconds` is set. size: type: string description: >- Size shortcut, e.g. `720x1280`, `720P`, `1080P`. Can imply resolution + aspect ratio. example: 720x1280 image: type: string description: Single reference image URL. images: type: array items: type: string description: >- Multi reference image URLs (max 3). Prefer `images` for Omni referenced billing. maxItems: 3 input_reference: type: string description: Reference image URL (similar to `image`). mode: type: string description: e.g. `pro` for motion-control Pro tier. example: pro content: type: array description: >- OpenAI-style multimodal content. Text is taken from `type=text` parts. items: type: object properties: type: type: string description: text / image_url / video_url … text: type: string image_url: type: object properties: url: type: string metadata: $ref: '#/components/schemas/KlingMetadata' KlingTask: type: object properties: id: type: string task_id: type: string object: type: string example: video model: type: string description: Concrete billing tier name. status: type: string description: >- queued / pending / processing / in_progress / completed / failed / cancelled progress: type: integer created_at: type: integer completed_at: type: integer seconds: type: string size: type: string video_url: type: string error: type: object properties: code: type: string message: type: string KlingMetadata: type: object description: Kling extended options. properties: scene_type: type: string enum: - motion_control - avatar_i2v - lip_sync description: >- Scene type. Omit for normal T2V/I2V. `motion_control` requires a public reference video. motion_level: type: string enum: - std - pro description: 'Motion-control tier (default `std`). Also `mode: "pro"`.' video_url: type: string description: >- Reference video URL (required for motion_control; also usable for Omni). last_frame_url: type: string description: Last-frame image URL (first/last frame). last_frame_file_id: type: string description: Last-frame file id. file_infos: type: array maxItems: 3 description: Structured file inputs (max 3). items: $ref: '#/components/schemas/KlingFileInfo' ext_info: type: string description: >- Advanced extension JSON **string** (not object), e.g. multi-shot for kling-3.0 / omni. output_config: $ref: '#/components/schemas/KlingOutputConfig' resolution: type: string description: Resolution shortcut (= `output_config.resolution`). audio_generation: type: string enum: - Enabled - Disabled description: Audio shortcut (= `output_config.audio_generation`). audio: type: boolean description: '`true` enables audio.' seconds: type: string duration: oneOf: - type: number - type: string video_duration: oneOf: - type: number - type: string mode: type: string KlingFileInfo: type: object properties: type: type: string enum: - File - Url description: Url requires `url`; File requires `file_id`. category: type: string enum: - Image - Video url: type: string file_id: type: string usage: type: string description: e.g. FirstFrame / Reference. reference_type: type: string enum: - feature - base description: Video reference type. KlingOutputConfig: type: object properties: duration: type: number description: Output duration in seconds; keep aligned with top-level `seconds`. example: 5 resolution: type: string enum: - 720P - 1080P description: Default 720P. aspect_ratio: type: string enum: - '16:9' - '9:16' - '1:1' description: T2V default 16:9. audio_generation: type: string enum: - Enabled - Disabled description: Default Disabled (mute). storage_mode: type: string enum: - Permanent - Temporary media_name: type: string maxLength: 64 class_id: type: integer expire_time: type: string description: ISO 8601. person_generation: type: string enum: - AllowAdult - Disallowed input_compliance_check: type: string enum: - Enabled - Disabled output_compliance_check: type: string enum: - Enabled - Disabled enhance_switch: type: string enum: - Enabled - Disabled ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.