Kling V3 Omni Video on Replicate

Unified multimodal video generation with reference images, video editing, native audio, and multi-shot control via Replicate.

We have integrated Kling V3 Omni Video (kwaivgi/kling-v3-omni-video) as two new video models — Standard (720p) and Pro (1080p) — powered by Replicate.

Why This Matters

  • Unified multimodal: Text-to-video, image-to-video, reference-based generation, video editing, and multi-shot control in a single model.
  • Native audio: Optional dialogue and ambient sound generation (when not using reference video).
  • Multi-shot mode: Create videos with up to 6 sequential scenes for narrative content.
  • Reference images: Up to 7 reference images for character and style consistency.

What's New

  • Catalog models: kling-v3-omni-standard (720p) and kling-v3-omni-pro (1080p) in the video model selector (Replicate backend).
  • GenAI pipeline: Server-side generation, per-second pricing, and webhook support.
  • Mastra tools: generateVideoToolKlingV3OmniStandard and generateVideoToolKlingV3OmniPro for agent-driven workflows.

Learn More

  • Model page and API: https://replicate.com/kwaivgi/kling-v3-omni-video
  • Kling V3 announcement: https://klingai.com