ComfyUI Node
Text Encode (MOSS Image)
MOSS-Image text conditioning with optional vision prompting. Reference images are passed as repeated MOSS vision blocks through the Qwen3.5-9B text encoder, with an optional per-image mask crop. No VAE or latent input is needed because MOSS-Image conditions only through the text encoder.
Text Encode (MOSS Image)
- clip
- image1
- mask1
- positive
- negative
◄prompt—►
◄negative_prompt—►
◄system_prompt—►
◄resolution1024►
◄mask_padding0.00►
◄vision_positionbefore prompt►
◄print_promptfalse►
Categorymodel/conditioning/moss image
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| clip | CLIP | — | |
| prompt | STRING | Text prompt. With reference images connected, the vision tokens are placed before or after this text according to vision_position. | |
| negative_prompt | STRING | Negative prompt. The same reference images and template are used for the negative conditioning. | |
| system_promptopt | STRING | Optional system instruction. MOSS-Image has no trained default system prompt, so leaving this unconnected sends no system message. Connect MOSS-Image System Prompt (or any STRING) to add one. | |
| image1opt | IMAGE | — | |
| mask1opt | MASK | — | |
| resolutionopt | INT | 10240–4096 | Reference images are resized to about resolution x resolution pixels, at multiples of 32, preserving aspect ratio. 0 keeps each reference at its own size, rounded to a multiple of 32. |
| mask_paddingopt | FLOAT | 0.000–1 | Context kept around the mask before cropping, as a fraction of the image size added on EACH side. 0 = tight crop to the mask; 0.1 = ~10% margin of surroundings. Only applies when a mask is connected. |
| vision_positionopt | COMBO | before prompt | Where the image (vision) tokens sit in the user turn relative to your text. 'before prompt' = image then text (default); 'after prompt' = text then image. No effect without an image. |
| print_promptopt | BOOLEAN | false | Print the full assembled prompt sent to the MOSS-Image text encoder (system instruction + vision blocks + your text) to the console. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| positive | CONDITIONING | — |
| negative | CONDITIONING | — |