Nodaro DocsGeneration settings nodes
The Generation Settings nodes each hold one value, such as a model or a tone. Only Tone reaches a run, through the Look input of image and video nodes.
Generation settings nodes each hold one value: a model choice, an aspect ratio, a duration, a motion level, a number of scenes, a tone or a style description. They never run a model and cost no credits.
Only Tone reaches a run. The other six hold their value on the canvas, but the editor does not let you wire them into the generation nodes. Set the model, the frame, the length and the number of scenes in each generation node's own settings instead.
The settings nodes
| Node | Holds | Default | Changes a run |
|---|---|---|---|
| Provider | A model choice for image, video, voice or script generation | Image, Nano Banana | No |
| Aspect Ratio | A frame shape: 16:9, 9:16, 1:1, 4:3 or 4:5 | 16:9 | No |
| Duration | A target length, from 1 to 600 seconds | 60 seconds | No |
| Motion | A motion level: Subtle, Moderate or Dynamic | Moderate | No |
| Scene Count | A number of scenes, from 1 to 20 | 5 | No |
| Tone | Your own words for the tone, such as cinematic | Empty | Yes, through the Look input of image and video nodes |
| Style Guide | Your own description of the visual style | Empty | No |
Where to find them
Press Tab on the canvas and open the Models tab. The settings nodes are in the Generation Settings family, for example Models › Generation Settings › Tone.
What connects
- Tone connects to the Look input of Generate Image and Generate Video. Its words are added to the end of the prompt, like the wording of a picker. It also connects to the start-state and end-state inputs of Camera Motion and Transition.
- Provider, Aspect Ratio, Duration, Motion, Scene Count and Style Guide cannot be connected to Generate Image, Generate Video or Generate Script. The canvas refuses the wire, so their values never reach a run.
Where to set each value
| Value | Where to set it |
|---|---|
| The model | Provider in each generation node's settings panel |
| The frame shape | Aspect Ratio in each image or video node |
| The length | Duration (seconds) in each video node, or Target Length (seconds) in Generate Script |
| The amount of motion | Motion hint (injected into prompt) in Generate Video |
| The number of scenes | Number of Scenes in Generate Script |
| The visual style | Style in Generate Image, Style Guide in Generate Script, or a Style picker |
To run many shots on one model, drive them from a List into one generation node. The node runs once per item, so there is only one model to set. See Lists and batching.
Settings nodes and pickers
| Settings nodes | Creative Controls pickers | |
|---|---|---|
| Holds | A value, such as 16:9, or your own text | A choice from a catalog |
| Adds to the prompt | Only Tone, through the Look input | Tested wording for the choice |
| Prompt hint toggle | No | Yes |
| Credits | None | None |
Every settings node
Generation Settings
- ProviderThe Provider node holds one model choice for image, video, voice or script generation. It cannot be wired into generation nodes, so set the model on each node.
- Aspect RatioThe Aspect Ratio node holds one frame shape, such as 16:9 or 9:16. It cannot be wired into image or video nodes, so set the ratio on each node.
- DurationThe Duration node holds a target length in seconds. It cannot be wired into video or script nodes, so set the length on each node.
- MotionThe Motion node holds a motion level, Subtle, Moderate or Dynamic. It cannot be wired into Generate Video, which has its own Motion hint setting.
- Scene CountThe Scene Count node holds a number of scenes, 5 by default. It cannot be wired into Generate Script, which has its own Number of Scenes setting.
- ToneWrite a tone once, such as cinematic, cheerful or dark and moody, and connect it to the Look input of image and video nodes, which add it to their prompt.
- Style GuideThe Style Guide node holds written visual direction. It cannot be wired into generation nodes, so put your style in Generate Script, Generate Image or a picker.
Frequently asked questions
Related
Tone
Creative Controls
Creative controls
Choosing a model
Node reference
Last updated on
Forced Alignment
Time every word of a known transcript against its audio and get each word's start and end as data, for karaoke highlights, timed graphics and captions.
Provider
The Provider node holds one model choice for image, video, voice or script generation. It cannot be wired into generation nodes, so set the model on each node.