Video Model Router

Which AI video model can actually generate your shot? A free browser-side capability matrix over Veo 3.2, Veo 3.1, Kling 2.6 Pro, Sora 2, LTX-2 Pro, Runway Gen-3 Turbo, Luma Dream Machine, Pika 2.2, CogVideoX and Seedance 2.0 - checking duration, aspect ratio, resolution, audio, dialogue, reference slots, face preservation, motion transfer and keyframes, and listing every ruled-out model with the published limit that ruled it out. Then five lanes: choose the model, write the Veo 3.2 prompt with its typed reference slots, port a prompt to another model with an honest account of what is lost, plan an image-to-video, or extend a clip. Lanes derived from the open-source Agent Skills pexoai/pexo-skills (veo-3.2-prompter), guia-matthieu/clawfu-skills (ai-video-prompting) and josiahsiegel/claude-plugin-marketplace (fal-model-guide, fal-text-to-video, fal-image-to-video, fal-video-to-video). Not affiliated with or endorsed by Google, OpenAI, Kuaishou, Runway, Lightricks, Luma, Pika, Zhipu or ByteDance.

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.