---
title: FLUX.1-Fill-dev-LoRA-extract
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract.md"
repository: AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract
last_updated: 2024-11-22
license: other
parameters: 2.4B
tensor_type:
  - F16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 1090
stars: 2
---

# FLUX.1-Fill-dev-LoRA-extract

> FLUX.1-Fill-dev-LoRA-extract - AI-ModelScope 在 ModelScope 开源的模型。LoRA extracts of https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev/ made using https://github.com/bmaltais/kohyass/tree/sd3-sd3.5-flux.

AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract 是 ModelScope 魔搭社区上的 2.4B 参数机器学习模型，采用 other 许可。

- **Repository**: AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract
- **License**: other
- **Parameters**: 2.4B
- **Downloads**: 1090
- **Stars**: 2
- **Last updated**: 2024-11-22

Source: https://www.modelscope.cn/models/AI-ModelScope/FLUX.1-Fill-dev-LoRA-extract

---

LoRA extracts of https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev/ made using https://github.com/bmaltais/kohya_ss/tree/sd3-sd3.5-flux.

Various network dimensions provided - your mileage may vary. 
After a few attempts, I have not been able to generate anything good. I don't know if this should even work, or what rank is ideal, or what weight.
But they are there if you want to try it too. I'd recommend trying rank 128 as that is the size of the FLUX.1-Depth-dev and FLUX.1-Canny-dev LoRAs.