Initial release: V1, V2, and V3 models

README.md

@@ -1,3 +1,126 @@
 ---
 license: apache-2.0
+license_name: apache-2.0
+tags:
+- lora
+- manga
+- coloring
+- anime
+- qwen
+- dataset
+- diffusers
+- image-to-image
+viewer: false
 ---
+
+# PanelPainter-Project
+
+**PanelPainter-Project** is the central repository for the PanelPainter manga coloring LoRAs.
+
+This project is dedicated to training LoRAs to automate the coloring of black-and-white manga panels. I am releasing all the files here, including datasets, logs, and experimental versions, so others can see exactly how it was trained.
+
+## Project Structure
+
+This repository contains everything used to create the models:
+
+### 1. LoRA Models (`/loras`)
+This directory contains the model weights for all iterations of the project:
+
+* **V3 (Latest Release):** `PanelPainter_v3_Qwen2511.safetensors`
+    * **Base:** Qwen Image Edit 2511
+    * **Note:** The latest model trained on the expanded 903-image dataset.
+* **V2 (Stable):** `PanelPainter_v2_Qwen2509.safetensors`
+    * **Base:** Qwen Image Edit 2509 (Compatible with 2511).
+    * **Note:** Standard release (High quality, low variety).
+* **V1 (Legacy):** `PanelPainter_v1_Legacy.safetensors`
+    * **Base:** Qwen Image Edit 2509
+    * **Note:** Archived experimental version (synthetic data).
+
+### 2. Training Logs (`/logs`)
+**Content:** Tensorboard logs and charts from my training runs. You can check these to see how the loss converged and how the model learned over time for each version.
+
+### 3. Training Dataset
+The datasets used for this project are hosted separately:
+
+* **PanelPainter-Dataset**
+    * Contains the curated image pairs used for training the active versions.
+
+---
+
+## Version History & Development Log
+
+### Version 3.0 (Current Release)
+* **Status:** Released.
+* **Base Architecture:** Qwen 2511.
+* **Strategy:** Scaling Up High-Quality Data.
+* **Dataset:** Expanded to 903 images.
+* **Summary:** This version combines the correct "real line art" training method discovered in V2 with a significantly larger dataset. This improves the model's ability to generalize across different manga styles while maintaining the color quality of V2.
+
+### Version 2.0
+* **Status:** Released / Stable.
+* **Base Model:** Trained on Qwen Image Edit 2509, also it works on Qwen 2511 as well.
+* **The Breakthrough:** After V1 failed, this version switched to training on real line art instead of synthetic grayscale.
+* **Dataset:** A tiny, hyper-curated set of 150 images (70% Doujin / 30% SFW).
+* **Outcome:** Despite the small size, it proved that high-quality real line art outperforms massive synthetic datasets. It produces good colors but lacks variety due to the small sample size.
+
+### Version 1.0
+* **Status:** Archived / Deprecated.
+* **Base Model:** Qwen Image Edit 2509.
+* **The Mistake:** Trained on 7,000 images generated by simply desaturating colored pages (synthetic grayscale).
+* **Outcome:** The model learned to color "perfect gray" inputs but failed on real, imperfect ink lines.
+* **Lesson:** Quantity does not matter if the data distribution doesn't match real usage.
+
+---
+
+## Training Configuration (V3)
+
+**Hardware:** Trained on an A40 GPU on Runpod for approximately two days.
+
+Below is the exact accelerate command used to train the V3 model on Musubi Tuner:
+
+```bash
+accelerate launch --num_cpu_threads_per_process 1 --mixed_precision bf16 \
+  /workspace/musubi-tuner/src/musubi_tuner/qwen_image_train_network.py \
+  --dataset_config dataset_edit.toml \
+  --dit /workspace/Training_Models_Qwen/Qwen_Image_Edit_2511_BF16.safetensors \
+  --vae /workspace/Training_Models_Qwen/qwen_train_vae.safetensors \
+  --text_encoder /workspace/Training_Models_Qwen/qwen_2.5_vl_7b_bf16.safetensors \
+  --model_version edit-2511 \
+  --network_module networks.lora_qwen_image \
+  --output_dir /workspace/output_panelpainter \
+  --output_name panelpainter_v3_part1 \
+  --mixed_precision bf16 \
+  --max_data_loader_n_workers 0 \
+  --learning_rate 3e-4 \
+  --network_dim 128 \
+  --network_alpha 128 \
+  --optimizer_type adafactor \
+  --optimizer_args "scale_parameter=False" "relative_step=False" "warmup_init=False" "weight_decay=0.01" \
+  --lr_scheduler cosine \
+  --lr_warmup_steps 150 \
+  --timestep_sampling qinglong_qwen \
+  --discrete_flow_shift 2.2 \
+  --max_train_epochs 8 \
+  --save_every_n_epochs 1 \
+  --save_state \
+  --gradient_checkpointing \
+  --gradient_checkpointing_cpu_offload \
+  --gradient_accumulation_steps 4 \
+  --blocks_to_swap 20 \
+  --sdpa
+```
+
+## Data Privacy & License
+
+* **Project License:** Apache 2.0
+* **Dataset Disclaimer:** The /dataset folder contains copyrighted manga panels.
+* **For Learning Only:** I am sharing this strictly to show how the model was trained.
+* **Copyright:** The original art belongs to the creators/publishers.
+* **No Selling:** Please do not sell or repackage these images.
+
+## Acknowledgements
+
+Trained on Musubi Tuner. Thanks to kohya-ss.
+
+## External Links
+* **Public Model Page:** [Civitai: PanelPainter](https://civitai.com/models/2103847/panelpainter-manga-coloring)

loras/v1/PanelPainter_v1_Legacy.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42974fc604542faf47cf55aea4da2ae97c8d51a2aec7068553df362486ed4123
+size 295241616

loras/v2/PanelPainter_v2_Qwen2509.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:139c996de30e0a617573b5cc25a0d5a0c974ad7258635b3ae9b59593b9303e66
+size 2359632080