PanelPainter-Project is an open-source initiative to automate black-and-white manga coloring using fine-tuned LoRAs. This project is dedicated to training models that maintain clean line art while achieving smooth, anime-style color fills.
Update – V3.0 Major Release
PanelPainter V3 is a major milestone. Unlike previous "helper" versions, V3 is a standalone coloring lora trained on Qwen Image Edit 2511.
Training Data: Trained on 903 hand-picked panels using Natural Language Captions.
Methodology: Combines the "real line art" training method discovered in V2 with a significantly larger dataset to generalize across different manga styles.
Trigger Word:
Color this panelpainter
Workflows & Resources
For the best results, use the dedicated workflows on RunningHub.
PanelPainter V3 (Qwen 2511)
RunningHub: V3 BF16 Workflow (Fast, Balanced)
RunningHub: V3 AIO Workflow (All-In-One)
PanelPainter V2.5 (Qwen 2511)
RunningHub: V2.5 AIO App (With VL Prompting) – Includes Vision Language prompting for better adherence.
Usage Guide
V3 has been tested on various styles including Chainsaw Man, Frieren, Komi Can't Communicate, and Oshi no Ko.
Recommended Generation Settings
LoRA: PanelPainter V3 (Weight: 1.0)
Helper LoRA: 4-Step Lighting (Weight: 1.0)
Steps: 4
Sampler: Euler
Scheduler: Simple
CFG: 1.0
Prompting Use the trigger word in your prompt:
Color this panelpainter
You can also add specific lighting or atmospheric tags:
Color this panelpainter, sunset lighting, warm tones
Project History & Development
Version 3.0 (Current)
Base: Qwen 2511
Summary: Scaled up to 903 images using the "real line art" training method. Solves the generalization issues of V2 while maintaining color quality.
Version 2.0 (Stable)
Base: Qwen 2509 (Compatible with 2511)
Summary: The breakthrough version. Switched from synthetic data to real line art (150 images). Proved that small, high-quality datasets outperform massive synthetic ones.
Version 1.0 (Legacy)
Summary: Trained on 7,000 synthetic grayscale images. Failed to handle real ink imperfections. Kept for archival purposes.
Credits & Acknowledgements
Training: Trained on Musubi Tuner (Thanks to kohya-ss).
Dataset Contributors: Special thanks to @Rox_Jr & @lucifer_brine04 for their help with dataset curation.
Description
Initial experimental release of PanelPainter.
Trained for 7,000 steps on 7,500 NSFW doujin panels (nhentai).
Optimized for use with Qwen Image Edit Plus / 2509 models.
Focused on preserving clean line art and achieving flat anime-style colors.
Works best when paired with Qwen-Rapid-AIO-NSFW-v8.safetensors.
May produce inconsistent results on SFW or action manga panels due to NSFW-based training data.
Ideal LoRA weight range: 0.45–0.6.
FAQ
Comments (24)
Great Lora
Great Lora, with what tool did you train this, could I please know?
musubi-tuner with custom sampler - https://github.com/kohya-ss/musubi-tuner/blob/main/docs/advanced_config.md#style-friendly-snr-sampler
Does this lora works with nunchaku?
Technically it should work, but you can give a try
@Kokoboy yes it does, it need some tweaks and it works even better, ill give you examples later
The AIO model includes 4 step and 8 step acceleration Lora, which may require additional configuration if it is the original version
@Rating_Agent help
@qq369885545 what?
Are you using nunchaku checkpoint provided or a nsfw aio rapid?
@CawfeeA Nunchaku
@Rating_Agent How did you get it to work? Its just overblown bright color on nunchaku
@Rating_Agent so where are examples?how to tweak
I added my eaxmples, i found this lora working best set on 1.2 with QWEN-EDIT UPSCALER & DENOISE V3 FIXED plased before it at 1.0
hey where can I find the qen edit upscaler?
@VeerGeer thanks
Dude, can u run this workflow in local?
@didiasol let me check
@Rating_Agent Any progress?
@didiasol yeah, works good. better than v1
@didiasol i use nunchaku fp4 btw
@Rating_Agent How can u guys deal with TT img enc?This node cause eveyting that output images are blank.
@didiasol just unload the model and try again