ComfyUI Node
Eric Qwen-Edit Style Transfer
A ComfyUI node in Eric Qwen-Edit with 17 inputs and 1 output.
Eric Qwen-Edit Style Transfer
- pipeline
- style_image
- content_image
- image
◄style_modefull_style►
◄custom_prompt►
◄additional_guidance►
◄style_strength1.0►
◄vae_target_size1024►
◄vl_styletrue►
◄vl_contenttrue►
◄ref_stylefalse►
◄ref_contenttrue►
◄negative_promptblurry, distorted, low quality►
◄steps8►
◄true_cfg_scale4.0►
◄seed0►
◄max_mp8.0►
CategoryEric Qwen-Edit
Inputs (17)
| Name | Type | Default | Description |
|---|---|---|---|
| pipeline | QWEN_EDIT_PIPELINE | — | |
| style_image | IMAGE | Reference image providing the style (passed as Picture 1) | |
| content_image | IMAGE | Image to apply the style to (passed as Picture 2) | |
| style_mode | COMBO | full_style | Type of style transfer to apply |
| custom_promptopt | STRING | When non-empty, overrides the style_mode template entirely. Reference images as Picture 1 (style) and Picture 2 (content). | |
| additional_guidanceopt | STRING | Extra instructions appended to the style prompt (e.g., 'but keep faces realistic') | |
| style_strengthopt | FLOAT | 1.00.1–2 | How strongly to apply the style (affects CFG emphasis). 1.0=balanced, >1=stronger style, <1=more original content |
| vae_target_sizeopt | INT | 10240–2048 | Fixed resolution for VAE encoding. 0 (default) = encode refs at output resolution (matches Edit node behavior, best for high-res). Set to e.g. 1024 to force refs to ~1MP (only useful at low output res). |
| vl_styleopt | BOOLEAN | true | Include style image in VL/semantic path (text encoder sees its content) |
| vl_contentopt | BOOLEAN | true | Include content image in VL/semantic path |
| ref_styleopt | BOOLEAN | false | Include style image in VAE/ref path (pixel-level latent). Default False — style is semantic-only, preventing structure bleed. |
| ref_contentopt | BOOLEAN | true | Include content image in VAE/ref path (preserves its pixel structure) |
| negative_promptopt | STRING | blurry, distorted, low quality | What to avoid |
| stepsopt | INT | 81–100 | Inference steps (8 for lightning LoRA, 50 for base model) |
| true_cfg_scaleopt | FLOAT | 4.01–20 | True CFG scale |
| seedopt | INT | 00–18446744073709550000 | Random seed |
| max_mpopt | FLOAT | 8.00.5–16 | Max output megapixels. VAE refs scale to match output resolution. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |