---
license: other
license_name: masafee-lora-license
license_link: LICENSE
base_model: gsdf/Counterfeit-V3.0
tags:
  - lora
  - stable-diffusion
  - text-to-image
  - diffusers
  - kohya-ss
library_name: diffusers
pipeline_tag: text-to-image
language:
  - en
  - ja
---

# Masafee LoRA

[![GitHub](https://img.shields.io/badge/GitHub-masafykun%2Fmasafee--lora-181717?logo=github)](https://github.com/masafykun/masafee-lora)
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20397946.svg)](https://doi.org/10.5281/zenodo.20397946)
[![ORCID](https://img.shields.io/badge/ORCID-0009--0000--7977--2756-A6CE39?logo=orcid&logoColor=white)](https://orcid.org/0009-0000-7977-2756)

Stable Diffusion LoRA for the original character **Masafee** — a red panda from a personal [LINE sticker set](https://store.line.me/stickershop/product/19794926/ja).
Trained entirely on a single consumer GPU (NVIDIA GeForce RTX 3060, 12GB) with [kohya-ss/sd-scripts](https://github.com/kohya-ss/sd-scripts).

## Model details

| | |
|---|---|
| Base model | Counterfeit-V3.0 |
| Method | LoRA (rank 32) |
| Learning rate | 1e-4 |
| Epochs | 30 (sweet spot before overfitting) |
| Training images | 13 hand-picked LINE stickers |
| Trigger word | `masafee` |
| Recommended weight | `0.8` |
| Hardware | RTX 3060 12GB (≈9 min / 16 epochs) |

## Usage

```python
from diffusers import StableDiffusionPipeline
import torch

pipe = StableDiffusionPipeline.from_pretrained(
    "gsdf/Counterfeit-V3.0",
    torch_dtype=torch.float16,
).to("cuda")
pipe.load_lora_weights(
    "masafy/masafee-lora",
    weight_name="cf60-000030.safetensors",
)

prompt = "masafee, red panda character, sitting on a park bench, sunny day, masterpiece"
image = pipe(prompt, num_inference_steps=30, guidance_scale=7.5).images[0]
image.save("out.png")
```

Apply at `network_mul=0.8` for the most faithful character likeness without rigidity.

## Paper & full reproduction

The full experiment writeup (Japanese / English, two-column PDF), all training scripts, the 6-condition sweep, the epoch ablation, and side-by-side comparison grids are in the companion GitHub repo:

🔗 **https://github.com/masafykun/masafee-lora**

### Key findings from the paper

1. **Base model choice dominates** — switching the base model (SD1.5 → Counterfeit-V3.0) had a far larger effect on output quality than tuning LR or LoRA rank.
2. **Epoch ~30 is the sweet spot** — extending to 40–60 epochs produced visible overfitting without improving character fidelity.
3. **Output quality is bounded by training image count** — with only 13 stickers, certain poses and lighting conditions remain out of distribution.

## Citation

If you use this model in academic work, please cite it via the Zenodo archive:

```bibtex
@software{suzuki_masafee_lora_2026,
  author       = {Suzuki, Masato},
  title        = {{Masafee LoRA: Single-GPU LoRA Fine-tuning of an
                   Original Character on Stable Diffusion}},
  year         = {2026},
  version      = {v1.0.0},
  doi          = {10.5281/zenodo.20397946},
  url          = {https://doi.org/10.5281/zenodo.20397946},
  orcid        = {0009-0000-7977-2756}
}
```

**Persistent archive (Zenodo DOI):** [10.5281/zenodo.20397946](https://doi.org/10.5281/zenodo.20397946)

## License

- **LoRA weights (this file):** released for **personal and research use only**. The weights encode a specific copyrighted character.
- **Code, scripts, and paper text** (in the GitHub repo): MIT.
- **Character "Masafee" itself, the original sticker images, and any images generated using this LoRA:** © 2026 masafykun / Masato Suzuki. **Redistribution, commercial use, and use to imply official endorsement are not permitted.**

By downloading this model you agree to use it only for personal experimentation or academic research.

---

Made with 🐾 by [masafykun](https://github.com/masafykun) · [masafy.org](https://masafy.org/) · [ORCID](https://orcid.org/0009-0000-7977-2756)