MOSS-Image Lineart Control LoRA

MOSS-Image Lineart Control LoRA

Colours a line drawing or turns a photo into a new scene that keeps its line structure, as a lineart control LoRA for MOSS-Image-SFT in ComfyUI, for people who restyle drawings and design sketches.

Made by thedeoxen on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: This is a lineart control LoRA for the MOSS-Image-SFT image model. You give it a fine white line drawing on black, and it follows those lines while your prompt sets colour, material and style. It is an edit-style LoRA, not the official MOSS colour or depth adapter. The author picked the checkpoint from training step 5000.

Who it suits: People who colour line drawings or restyle photos in ComfyUI, and want the lines to stay put, will find it fits. Designers whose sketches should lead the picture can use it too.

Industry | Who uses it | How they use it | Example
Design | Visual designers | Recolour a composition while keeping its outline | A new colour scheme on the same layout
Product imagery | Marketers | Reuse a photo's line structure for a new scene | A familiar layout in a new setting
Illustration | Artists | Colour a clean line drawing and let the prompt fill in the rest | A sketch turned into a coloured scene
Research | Image-model tinkerers | Test lineart control on MOSS-Image-SFT | Comparing line weights and strengths

If you do not use ComfyUI or MOSS-Image-SFT, you can skip it, since it is trained for that base model.

Where to use it:

• ComfyUI, with the workflow file included in the repository
• ComfyUI's stock node for encoding MOSS edit prompts
• Windows, macOS or Linux with Python, to download the files

Use case | Who | What they get back
Colour a line drawing | Illustrators | A coloured image that follows your lines
Turn a photo into a new scene | Artists | A scene that keeps the photo's line structure
Design sketches | Designers | Results shaped by the drawing rather than by lighting

How to set it up: You need ComfyUI, the MOSS-Image-SFT files and this LoRA. The author names three base files and a workflow.

• ComfyUI with the Realistic Lineart preprocessor, with coarse mode on
• MOSS-Image-SFT: the UNet moss_image_sft_int8_convrot.safetensors
• A Qwen3.5 text encoder file for MOSS-Image, with the moss_image type chosen when you load it
• The VAE flux2-vae_small.safetensors

1. Install the Hugging Face hub package.
2. Download this repository's files, which hold the LoRA and the workflow.
3. Load the MOSS-Image-SFT UNet, text encoder and VAE.
4. Apply this LoRA to the UNet with a strength of 1.0.

pip install -U huggingface_hub
hf download thedeoxen/MOSS-Image-lineart-controlnet

How to use it: Make a line map from your photo, or draw one, then hand it to the Text Encode MOSS-Image Edit node as the only reference image. On the positive and the negative encode alike, switch the edit format to ai-toolkit. Sample with 30 steps, CFG 4, and euler with simple. There is no trigger word.

Tool | What it does
Realistic Lineart node | Builds the white-on-black line map, with coarse mode on in the workflow
Text Encode MOSS-Image Edit | Takes the control image and your prompt
LoRA loader | Applies this LoRA at strength 1.0
Included workflow | Loads a photo, builds the map, encodes, and samples

Example: You load a photo, build its line map and write a prompt for a circus ringmaster in a scarlet coat in front of a painted tent. The result keeps the line structure from the photo, while the prompt supplies the clothing, colour and setting.

For newcomers and experts: A line map is a picture made only of thin lines. A good first step is to run the included workflow on one photo with the default settings, then change only the prompt.

Its adapter weights use rank 64 for the linear layers and rank 16 for the convolutions. It learned as an edit adapter. The control picture serves as the starting reference, and the text describes the image you want out. It was trained with the kv cache on, which is why the edit format is ai-toolkit. The coarse setting resembles the training maps more than the fine one does. If the lines feel too rigid, lower the strength toward 0.7.

You load a photo into the workflow
The Realistic Lineart node builds a white-on-black map
The map and your prompt go to the encode node
MOSS-Image-SFT with the LoRA samples the new image

Good to know:

• License: apache-2.0, as listed on Hugging Face
• Cost to run: none named
• Data it reaches: the images and prompts you load into ComfyUI, and the source names no web or account access
• Sparse, light white lines on black match the training better than a dense sketch
• Not for: other base models, since it was trained on and is recommended with MOSS-Image-SFT
• It is not the official MOSS colour or depth adapter

On Hugging Face, 13 November 2026

• Task: image to image
• Library: diffusers
• Downloads in the last 30 days: 1,548
• Likes: 3
• License: apache-2.0
• Languages: en
• Based on: OpenMOSS-Team/MOSS-Image-SFT
• Tags: diffusers, moss-image, controlnet, lora, comfyui, lineart, image-to-image

Links

• Model: https://huggingface.co/thedeoxen/MOSS-Image-lineart-controlnet
• Files: https://huggingface.co/thedeoxen/MOSS-Image-lineart-controlnet/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download thedeoxen/MOSS-Image-lineart-controlnet

Then follow the usage in its model card. Nothing in this kit runs by itself.