AI Video Generation

Video generation requires AI Multimedia

The AI Video feature is part of the AI Multimedia tier (MegaBundle and Complete Bundle). FE Commercial doesn't include video generation by default.

Video models supported

ContentRyte uses Nexus AI to route across 4 video providers:

• Sora (OpenAI) — highest quality, slowest
• Veo (Google) — fast, good for marketing clips
• Kling — strong for animated / stylized output
• Luma — best for photorealistic motion

The router picks the optimal provider based on prompt + speed needs, with automatic failover.

Two generation modes

Text-to-Video — describe what you want; AI generates from scratch.

Image-to-Video — upload a starting image; AI animates it into a 5–10 second clip.

Generating a video

1. Sidebar → Videos → Generate Video
2. Type a prompt OR upload a starting image
3. Pick model + length (5s / 10s)
4. Click Generate
5. Video generations are async — they take 2–5 minutes. You'll see a "Processing" indicator while it runs.

AI Avatar Pro

For talking-head videos:

1. Upload one clear photo of a person (yours, or any model)
2. Type or upload the script (text-to-speech or your own audio)
3. ContentRyte generates a video of that person delivering the script with synced lip movement

Combined with voice cloning, you can produce videos in your own voice and face from text.