Node Settings
Overview
Configure node parameters to control how each node processes data in your workflow. Understanding parameter types helps you fine-tune your AI generation results.
Accessing Node Settings
Node settings appear directly in the node body on the canvas. Click on any parameter field to edit its value.
For additional options:
- Right-click on a node
- Select Properties from the context menu
- View and edit all node properties
Parameter Types
Nodes use different parameter types depending on what kind of input they need.
Numeric Parameters
Type | Description | Example Values |
|---|---|---|
Integer (INT) | Whole numbers only | 1, 512, 1024 |
Float (FLOAT) | Decimal numbers | 0.5, 7.5, 0.75 |
Seed | Random seed for generation | Any integer; -1 for random |
Numeric fields often have constraints:
- Min/Max values: Limits the range you can enter
- Step: Controls increment when using arrows or sliders
Text Parameters
Type | Description | Example |
|---|---|---|
String | Single-line text input | Filenames, simple values |
Text (multiline) | Multi-line text area | Prompts, descriptions |
Selection Parameters
Type | Description | How to Use |
|---|---|---|
Dropdown | Choose from predefined options | Click to open list, select option |
Combo | Select from available items | Often populated dynamically (e.g., model lists) |
Boolean Parameters
Value | Meaning |
|---|---|
True / On | Feature enabled |
False / Off | Feature disabled |
Toggle by clicking the checkbox or switch.
Common Node Parameters
Sampler Nodes
Parameter | Type | Description |
|---|---|---|
steps | Integer | Number of denoising steps (more = higher quality, slower) |
cfg | Float | Classifier-free guidance scale (how closely to follow prompt) |
sampler_name | Dropdown | Which sampling algorithm to use |
scheduler | Dropdown | How to schedule the denoising process |
denoise | Float | Amount of noise to remove (1.0 = full generation) |
Image Size Nodes
Parameter | Type | Description |
|---|---|---|
width | Integer | Image width in pixels |
height | Integer | Image height in pixels |
batch_size | Integer | Number of images to generate |
Loader Nodes
Parameter | Type | Description |
|---|---|---|
ckpt_name | Combo | Checkpoint model to load |
lora_name | Combo | LoRA model to apply |
strength | Float | How strongly to apply the LoRA |
Input Types
Inputs on the left side of nodes accept connections from other nodes.
Input Appearance | Meaning |
|---|---|
Solid dot | Required input (must be connected) |
Hollow dot / ring | Optional input (can be left empty) |
Optional inputs allow the node to run with default behavior when not connected.
Widgets vs Inputs
Some parameters can work as both widgets (editable in the node) and inputs (connected from another node):
- Widget mode: Edit the value directly in the node
- Input mode: Value comes from a connected node
Right-click on convertible parameters to toggle between modes.
Tips & Best Practices
- Start with defaults: Most nodes have sensible defaults; adjust after seeing initial results
- Small changes: Adjust one parameter at a time to understand its effect
- Save presets: Note down parameter combinations that work well for your use cases
- Check tooltips: Hover over parameters for descriptions of what they do
Troubleshooting
Parameter won't accept my value?
Solution: The value may be outside the allowed range. Check for min/max constraints on numeric fields.
Dropdown is empty?
Solution: The dropdown may populate from loaded resources. Ensure required models or files are loaded earlier in the workflow.
Changes not taking effect?
Solution: Re-run the workflow after changing parameters. Some changes only apply on the next execution.
Related Documentation
- Nodes