ComfyUI Node
AuK Generate / Edit X
A ComfyUI node in WorkflowX/Audio/AuK with 8 inputs and 1 output.
AuK Generate / Edit X
- model
- conditioning
- vae
- AUDIO
◄seconds3.0►
◄seed0►
◄steps32►
◄guidance2.0►
◄sway-1.00►
CategoryWorkflowX/Audio/AuK
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| model | WORKFLOWX_AUK_MODEL | AuK model from AuK Model Loader. | |
| conditioning | CONDITIONING | Conditioning from AuK Instruction Encode. | |
| vae | VAE | AuK VAE from AuK VAE Loader. Its own decode preserves AuK's volume edits. | |
| seconds | FLOAT | 3.00–3600 | Output length in seconds. 0 matches the source audio's duration. Text-only generation needs a positive value; roughly 2-3 spoken words per second. |
| seed | INT | 00–18446744073709550000 | Drives the diffusion noise and reference-latent sampling. Each batch item uses seed + index. |
| steps | INT | 321–1000 | Euler steps for Base; 32 is the upstream default and 16-50 is sensible. Flash ignores this and always runs its fixed 4-step schedule. |
| guidance | FLOAT | 2.00–100 | AuK guidance strength for Base; 2 is the upstream default. This is not KSampler CFG. Flash ignores this. |
| sway | FLOAT | -1.00-1–0 | Sway sampling for Base; -1 is the upstream default, 0 disables it. Flash ignores this. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| AUDIO | AUDIO | — |