How Canvas selects the route
With no media connected, Seedance runs text-to-video. With one source image, it runs image-to-video.
Available in Canvas
Seedance 2.0
Seedance 2.0 is available as a video-generation node in Keter Canvas. The current reliable paths are text-to-video and animation from one source image, with the result connected to the rest of the workflow.
ai modelvideo generationcanvasseedance2.0
What's included
What this section gives you.
Text-to-video
Start from a prompt alone to create a moving concept without preparing source media first.
Single-image animation
Connect one source image when the composition should remain anchored while Seedance adds motion.
Controls inside Canvas
Choose Quality or Mini mode, set duration, resolution and aspect ratio, then review the generated video beside the inputs that produced it.
When to use it
Common jobs this section is built for.
These examples make it easier to choose where to start and which nearby section to open next.
A scene from a prompt
Use text-to-video for a first moving concept when there is no approved source frame yet.
Animate a still
Use one source image when the composition and direction of the shot should stay anchored.
Continue the workflow
Keep the generated clip connected to editing, audio, review and export steps in the same Canvas graph.
Details
What to know before you use it.
Current controls
The available node supports 4–15 second clips, common landscape, portrait and square formats, and Quality or Mini execution. Quality can use the higher resolution option; Mini is limited to its supported tiers.
Multi-reference status
The runtime accepts reference media, but Canvas does not yet expose deterministic numbered ordering for multi-reference sets. Use text or one source image when reference order matters.
Workflow
How it works inside Keter Labs.
Every section in Keter Labs follows the same shape: a clear brief, a fast first draft, refinement in the editor, and a result saved into the project.
1
Step 1
Open Canvas and add a Seedance video node to the workflow.
2
Step 2
Write the prompt and either leave media disconnected for text-to-video or connect one image for image-to-video.
3
Step 3
Choose the mode, duration, resolution and aspect ratio that fit the intended delivery.
4
Step 4
Run the node, review motion and continuity, then connect the result to editing, audio, review or export steps.
Related sections
Where to go next.
AI model
Kling 3.0
Kling 3.0 is covered as a production option inside Keter Labs, not as a standalone novelty. The page explains where the model fits, what kind of input it needs, and how to judge whether the result is ready for a real project.
Open Kling 3.0AI model
Google Veo 3.1
Google Veo 3.1 is covered as a production option inside Keter Labs, not as a standalone novelty. The page explains where the model fits, what kind of input it needs, and how to judge whether the result is ready for a real project.
Open Google Veo 3.1Documentation
Video AI models
Video AI models is a practical Keter Labs help page on how to choose a video model based on duration, motion, style, speed, and generation budget. It helps you quickly understand the preparation, limitations, sequence of steps, and criteria for a finished result without a long search across multiple pages.
Open Video AI models
Video
Video generator
An AI video generator for short ad scenes, social content, product motion, mood videos and first drafts of clips. Start from text, an image or a reference, define the motion and bring the clip to a format you can show the team.
Open Video generatorReady when you are
Bring this workflow into your next project.
Spin up assets, lock in versions and align your team around a single creative platform.
Open Canvas for Seedance