# Kling v3 Omni Video > Kling Video 3.0 Omni: Unified multimodal video generation with reference images, video editing, native audio, and multi-shot control - **Provider**: replicate - **Model ID**: kwaivgi/kling-v3-omni-video - **Category**: video_generation - **Credits**: 2610 per request - **Pricing Type**: input_based ## API Endpoint Base URL: https://api.core.today/v1 ### Create Prediction POST /predictions ### Get Status GET /predictions/{job_id} ### Cancel DELETE /predictions/{job_id} ## Authentication Header: `X-API-Key: YOUR_API_KEY` ## Input Parameters - `end_image` (string, optional): Last frame image. Requires start_image. Supports .jpg/.jpeg/.png, max 10MB, min 300px. - `aspect_ratio` (string, optional): Aspect ratio. Required when not using start frame or video editing. (Default: `16:9`; Options: `16:9`, `9:16`, `1:1`) - `prompt` (string, **required**): Text prompt for video generation. Supports <<>>, <<>> template references. Max 2500 characters. - `start_image` (string, optional): First frame image. Supports .jpg/.jpeg/.png, max 10MB, min 300px, aspect ratio 1:2.5 to 2.5:1. - `video_reference_type` (string, optional): How to use reference video: 'feature' for style/camera reference, 'base' for video editing. (Default: `feature`; Options: `feature`, `base`) - `keep_original_sound` (boolean, optional): Keep original sound from reference video. (Default: `True`) - `generate_audio` (boolean, optional): Generate native audio. Mutually exclusive with reference video. (Default: `False`) - `mode` (string, optional): 'standard' generates 720p, 'pro' generates 1080p, '4k' generates 4K. 4K does not support reference_video. (Default: `pro`; Options: `standard`, `pro`, `4k`) - `reference_video` (string, optional): Reference video (.mp4/.mov). Duration 3-10s, resolution 720-2160px per side, max 200MB. - `duration` (integer, optional): Video duration in seconds (3-15). Ignored for video editing (base). (Default: `5`; Range: min: 3, max: 15) - `reference_images` (array, optional): Reference images for elements, scenes, or styles. Supports .jpg/.jpeg/.png. Max 7 without video, 4 with video. - `multi_prompt` (string, optional): JSON array of shot definitions for multi-shot mode. Each shot: {"prompt": "...", "duration": N}. Max 6 shots, min duration 1s per shot, total must equal duration. Example: [{"prompt":"A cat jumps","duration":3},{"prompt":"It lands","duration":2}] ## Example Request ```json { "model": "kwaivgi/kling-v3-omni-video", "input": { "mode": "pro", "duration": 15, "aspect_ratio": "16:9", "generate_audio": true, "prompt": "A Wes Anderson style movie trailer with symmetrical framing and pastel colors. Opening: a perfectly centered shot of a grand pink hotel facade with teal shutters. A bellboy in a pillbox hat pushes a luggage cart precisely down the center of a hallway. Quick lateral tracking shots: a mustachioed concierge stamps a passport, twin sisters in matching yellow dresses ride a tandem bicycle through a village square, a detective examines a painting through a magnifying glass while standing in an absurdly tiny elevator. Every shot is perfectly balanced and color-coordinated in pastels. Final symmetrical shot of the entire cast standing in a row on the hotel steps. Text in a vintage serif font: \"KLING V3 OMNI VIDEO\" then \"NOW ON REPLICATE\". Whimsical harpsichord music, the sound of a bell ding." } } ``` ## Response Format ```json { "job_id": "abc123", "status": "pending", "provider": "replicate", "model": "black-forest-labs/flux-schnell", "created_at": "2026-01-01T00:00:00Z", "result": null, "error": null } ``` Status values: `pending`, `processing`, `completed`, `failed`, `cancelled` ## Usage Flow 1. POST /predictions with model and input → receive job_id 2. Poll GET /predictions/{job_id} until status is `completed` or `failed` 3. Result contains output URL(s) or data ## Output Type url ## Tags kwaivgi, kling, video-generation, multimodal, video-editing, native-audio, multi-shot ## Documentation https://replicate.com/kwaivgi/kling-v3-omni-video