ComfyUI Node
Eric Qwen Conditioning Interpolate
A ComfyUI node in Eric Qwen-Edit/Conditioning with 7 inputs and 1 output.
Eric Qwen Conditioning Interpolate
- cond_A
- cond_B
- conditioning
◄alpha1.00►
◄methodlerp►
◄scopeall_tokens►
◄prefix_N0►
◄scope_N7►
CategoryEric Qwen-Edit/Conditioning
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| cond_A | QWEN_CONDITIONING | Baseline conditioning (alpha=0 result) | |
| cond_B | QWEN_CONDITIONING | Target conditioning (alpha=1 result) | |
| alpha | FLOAT | 1.00-5–5 | 0=cond_A (no change), 1=cond_B (full target), >1=exaggerated, <0=opposite. Text prompt dominates coarse intent; alpha shapes the character. |
| methodopt | COMBO | lerp | lerp: linear interpolation. Fast, good for alpha 0.3-1.2. slerp: spherical interpolation. Preserves embedding magnitude. More uniform perceptual steps. Better for wide alpha sweeps or large extrapolation (>1.5). |
| scopeopt | COMBO | all_tokens | all_tokens: interpolate everything. skip_prefix_N: text tokens only → expression editing. keep_prefix_N: image tokens only → style transfer. last_N_tokens: not recommended for Qwen. |
| prefix_Nopt | INT | 00–512 | 0 = auto from Encode metadata (recommended). |
| scope_Nopt | INT | 71–128 | Only used for last_N_tokens scope. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| conditioning | QWEN_CONDITIONING | — |