---
title: Flux2-Klein-9B-True-V3
canonical_url: "https://www.modelscope.cn/models/wikeeyang/Flux2-Klein-9B-True-V3"
md_url: "https://www.modelscope.cn/models/wikeeyang/Flux2-Klein-9B-True-V3.md"
repository: wikeeyang/Flux2-Klein-9B-True-V3
chinese_name: Flux2-Klein-9B-True-V3
last_updated: 2026-07-01
license: apache-2.0
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - black-forest-labs/FLUX.2-klein-9B
base_model_relation: finetune
parameters: 41.9B
tensor_type:
  - I8
  - BF16
  - U8
  - F32
  - F8_E4M3
library_name:
  - safetensors
  - gguf
  - pytorch
frameworks:
  - PyTorch
language:
  - en
  - zh
downloads: 8768
stars: 54
tags:
  - gguf
---

# Flux2-Klein-9B-True-V3

> Flux2-Klein-9B-True-V3 - wikeeyang 在 ModelScope 开源的模型。Flux2-Klein-9B-True-V3

wikeeyang/Flux2-Klein-9B-True-V3 是 ModelScope 魔搭社区上的 41.9B 参数text-to-image-synthesis模型，采用 apache-2.0 许可，基于 black-forest-labs/FLUX.2-klein-9B 构建。

- **Repository**: wikeeyang/Flux2-Klein-9B-True-V3
- **License**: apache-2.0
- **Tasks**: text-to-image-synthesis
- **Parameters**: 41.9B
- **Base model**: black-forest-labs/FLUX.2-klein-9B
- **Tags**: gguf
- **Downloads**: 8768
- **Stars**: 54
- **Last updated**: 2026-07-01

Source: https://www.modelscope.cn/models/wikeeyang/Flux2-Klein-9B-True-V3

---

## Flux2-Klein-9B-True-V3

**如果您喜欢我的模型，请支持 ❤️ 我以获得更多的算力来调优模型。**

**If you like my models, please support ❤️ me for more computing resources to do finetuning work.**

[![wikeeyang](ko-fi-wikeeyang.jpg)](https://ko-fi.com/wikeeyang)![Wechat](Wechat-Sponsor-Code.jpg)

====================================================================================================

### **V3 版本有较大提升，基本解决了 V1 / V2 中存在的一些问题，美学和构图有较大改善。详见以下例图：**

**The version 3 has significant improvements, basically solving some of the issues present in V1/V2, with noticeable enhancements in aesthetics and composition. See the examples below:**

1. **不同量化精度出图质量对比，量化损失很小。High-quality quantization parameters, the loss is very low.**

   ![Quantified-Compare](1-Quantified-Compare.jpg)

2. **与官方及旧版模型对比效果。Comparison with the official and old versions.**

   ![Compare-with-offical-and-old-version](2-Compare-with-offical-and-old-version.jpg)

3. **V1 - V3 样例与对比。Examples and Comparisons of V1 - V3.**

   ![Sample-Compare-v1-3](3-Sample-Compare-v1-3.jpg)

4. **仅提示词指令图像编辑。Just use instruct prompt to edit image.**

   提示词：给他穿上粗犷的牛仔服外套，敞开着，露出他原来穿的T恤和图案，戴上一顶帽沿上翘的粗皮质牛仔帽，并让他嘴上叼个烟斗，烟斗上有一缕烟雾。保持图片的其它元素不变。

   Prompt: Put him in a rugged denim jacket, open to show the T-shirt and design he was originally wearing, add a thick leather cowboy hat with an upturned brim, and have him holding a smoking pipe with a wisp of smoke. Keep the other elements of the picture unchanged.

   ![Instruct-edit](4-Instruct-edit.jpg)

5. **仅提示词 + LoRA 换头、换装等。Prompt + LoRA to do Tryon / Tryoff and Faceswap.**

   提示词：用图2的女士的脸，完全替换掉图1女士的脸。加 bfs_head_v1 LoRA 权重 0.75 (0.6-0.9)。

   Prompt: Use the face of the woman in picture 2 to completely replace the face of the woman in picture 1.    (model + lora - bfs_head_v1 weight = 0.75)

   ![Faceswap](5-Faceswap.jpg)

6. **通过 Mask + LoRA 实现图像的区域精准编辑。Precise regional image editing using Mask + LoRA.**

   (提示词和头像图片同上。The prompt and Face picture 2 are the same as above.)

   ![Regional edit](6-Mask-BFS-LoRA.jpg)



### 使用方法, How to use：

与 V1 / V2 版本模型相同，The same with V1 / V2 vesrion.

或者参考 B 站如下大佬们的视频, Please refer to UP's video as below :

感谢：[@**T8star-Aix**](https://space.bilibili.com/385085361)， [@**AI-KSK**](https://space.bilibili.com/110353151)， [@**黑鹤001**](https://space.bilibili.com/515231056)， [@**蓝色多脑盒**](https://space.bilibili.com/1450124458)， [@**初级AAA**](https://space.bilibili.com/478987143)， [@**山竹老师AIGC**](https://space.bilibili.com/158424637)， [@**炮老师的小课堂**](https://space.bilibili.com/397686377)， [@**HyLan_L**](https://space.bilibili.com/591877893)， [@**雪煮松风**](https://space.bilibili.com/692784900)， [@**冰缘排骨**](https://space.bilibili.com/3546582953757144)， [@**暴躁小白兔学AI**](https://space.bilibili.com/3690991840135491) 以及 [@**活力网TV**](https://space.bilibili.com/92055552)， 全平台统一号的 [@**CommintAI**](https://space.bilibili.com/158429672) 等一众社区大佬对本模型的各种技术评测、详细使用介绍及进一步微调提升等方面的建议和支持。

Also on:  [Huggingface.co](https://huggingface.co/wikeeyang/Flux2-Klein-9B-True-V3)



### 特别致谢，Special thanks：

感谢：[@**亮亮rayne**](https://www.linkedin.com/in/2941b192)，[@**GuangyuanSD**](https://huggingface.co/GuangyuanSD) 在模型调制优化过程一如既往的给与技术、构图、美学、社群等方面的指导和支持。

Thanks to [@**亮亮rayne**](https://www.linkedin.com/in/2941b192) and [@**GuangyuanSD**](https://huggingface.co/GuangyuanSD) for their usual guidance and support in the model tuning and optimization process, covering technology, composition, aesthetics, community, and more.



### 模型许可：

本模型请在遵循 black-forest-labs 公司非商业许可 ([FLUX Non-Commercial License](https://huggingface.co/black-forest-labs/FLUX.2-klein-9B/blob/main/LICENSE.md)) 下使用。

This model falls under the [FLUX Non-Commercial License](https://huggingface.co/black-forest-labs/FLUX.2-klein-9B/blob/main/LICENSE.md).