ComfyUI Node
Flux Klein Identity Feature Transfer ⚡
Multi-reference identity-preserving feature transfer for FLUX.2 Klein - schedules, presets, per-reference masks, optional sigma-aware strength scaling.
Flux Klein Identity Feature Transfer ⚡
- model
- sigmas
- subject_mask_1
- subject_mask_2
- subject_mask_3
- subject_mask_4
- subject_mask_5
- subject_mask_6
- subject_mask_7
- subject_mask_8
- MODEL
◄presetHARD_LOCK►
◄enabledtrue►
◄reference_index0►
◄reference_indicesall►
◄similarity_floor0.040►
◄softmax_temperature0.0250►
◄mask_threshold1.00►
◄double_blocks0-7:mid_img=0.55►
◄single_blocks0:mid_img=0.22; 1:mid_img=0.24; 3:mid_img=0.28; 4:mid_img=0.22; 6:mid_img=0.26; 7:mid_img=0.27; 8:mid_img=0.25; 10:mid_img=0.27; 13:mid_img=0.27►
◄debugfalse►
◄mask_behaviorfocus_only►
Category🤖 CCTech/Flux Klein
Inputs (21)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| preset | COMBO | HARD_LOCK | 4 options: HARD_LOCK, MID_LOCK, SOFT_LOCK, custom |
| enabled | BOOLEAN | true | — |
| reference_index | INT | 00–15 | — |
| reference_indices | STRING | all | Zero-based references used by the transfer. 'all', comma-separated indices like '0,2,3', or ranges like '0-3'. |
| similarity_floor | FLOAT | 0.0400–0.95 | — |
| softmax_temperature | FLOAT | 0.02500.0001–0.25 | — |
| mask_threshold | FLOAT | 1.000–1 | — |
| double_blocks | STRING | 0-7:mid_img=0.55 | — |
| single_blocks | STRING | 0:mid_img=0.22; 1:mid_img=0.24; 3:mid_img=0.28; 4:mid_img=0.22; 6:mid_img=0.26; 7:mid_img=0.27; 8:mid_img=0.25; 10:mid_img=0.27; 13:mid_img=0.27 | — |
| debug | BOOLEAN | false | — |
| mask_behavior | COMBO | focus_only | focus_only: the mask limits this node's reference bank while Klein still sees the complete reference. zero_unmasked_tokens: also blocks each wired reference's unmasked tokens as attention sources in every block. References without a wired mask remain complete either way. |
| sigmasopt | SIGMAS | Optional sampler sigma schedule. When connected, block strengths decay per sampling step by delta_sigma_0 / delta_sigma_step. | |
| subject_mask_1opt | MASK | — | |
| subject_mask_2opt | MASK | — | |
| subject_mask_3opt | MASK | — | |
| subject_mask_4opt | MASK | — | |
| subject_mask_5opt | MASK | — | |
| subject_mask_6opt | MASK | — | |
| subject_mask_7opt | MASK | — | |
| subject_mask_8opt | MASK | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| MODEL | MODEL | — |