ComfyUI Node
Nunchaku FLUX.2 Klein Identity Guidance
A ComfyUI node in Nunchaku/FLUX.2 Klein/Enhancer/Post-CFG with 6 inputs and 1 output.
Nunchaku FLUX.2 Klein Identity Guidance
- model
- identity_latent
- MODEL
◄strength0.30►
◄start_percent0.00►
◄end_percent0.80►
◄modeadaptive►
CategoryNunchaku/FLUX.2 Klein/Enhancer/Post-CFG
Inputs (6)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| identity_latent | LATENT | VAE-encoded reference image at full resolution. | |
| strength | FLOAT | 0.300–1 | How hard to pull toward the reference each step. 0.3 = move 30% of the distance. |
| start_percent | FLOAT | 0.000–1 | When to start correcting. 0.0 = beginning of denoising. |
| end_percent | FLOAT | 0.800–1 | When to stop correcting. 0.8 = last 20% runs freely for texture refinement. |
| mode | COMBO | adaptive | adaptive: pulls only where prediction resembles reference. direct: pulls everywhere equally. channel_match: matches color/feature statistics without copying spatial content. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| MODEL | MODEL | — |