What generation modes are available?
- Text to video
- Image to video
- Reference to video
- First / last frame
- Motion control
- Lip sync
Generate a video clip from a text prompt. No source media required — describe the scene, motion, and style.Best for: backgrounds, B-roll, concept visualizations, abstract motion.
What utility modes are available?
Remove objects / background
Clean up a clip by removing unwanted objects or replacing the background scene.
Upscale
Increase video resolution and enhance detail for final delivery or display at larger sizes.
Join videos
Concatenate multiple clips into a single output — useful for assembling a sequence from separate generation runs.
Hyperframes (animated video)
Structured multi-stage pipeline: plan, storyboard, script, voiceover, and render a cohesive short film.
Which models are available for video nodes?
Capability matrices — including max duration, audio support, and resolution — vary by model. See Video models for the full comparison.
Video generation is among the most credit-intensive operations. Check the credit estimate in the node panel before running, especially for longer clips or higher resolutions.
How do you build a video pipeline?
1
Generate or upload source media
Start with a text-to-image node (or upload an image) to create your anchor visual.
2
Add a video node
Choose a generation mode — image-to-video is a common first step when you have a strong hero image.
3
Add audio (optional)
Connect an audio node output to a lip sync node or as a reference track for the Hyperframes renderer.
4
Apply utilities
Chain upscale, remove-background, or join nodes as needed.
5
Export or schedule
Send the final clip to export or connect to the Scheduler for publishing.
What’s next?
Hyperframes
Full animated video pipeline with storyboarding
Audio nodes
Add voiceover and dialogue to your clips
Video models reference
Full model capability matrix