---
title: open-tools-models
canonical_url: "https://www.modelscope.cn/models/sebponk/open-tools-models"
md_url: "https://www.modelscope.cn/models/sebponk/open-tools-models.md"
repository: sebponk/open-tools-models
chinese_name: "Open Tools 模型"
last_updated: 2026-11-03
license: other
pipeline_tag: text-driven-segmentation
tasks:
  - text-driven-segmentation
library_name:
  - onnx
downloads: 0
stars: 0
tags:
  - onnx
  - image-segmentation
  - background-removal
  - image-inpainting
---

# open-tools-models

> open-tools-models - sebponk 在 ModelScope 开源的模型。Open Tools 工具百宝箱使用的 AI 模型和运行环境（ONNX Runtime），程序内按需下载。

sebponk/open-tools-models 是 ModelScope 魔搭社区上的text-driven-segmentation模型，采用 other 许可。

- **Repository**: sebponk/open-tools-models
- **License**: other
- **Tasks**: text-driven-segmentation
- **Tags**: onnx, image-segmentation, background-removal, image-inpainting
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-11-03

Source: https://www.modelscope.cn/models/sebponk/open-tools-models

---

# Open Tools 模型 / Open Tools Models

[Open Tools 工具百宝箱](https://github.com/Bulua/open-tools)使用的 AI 模型和运行环境。程序内按需下载，不需要手动下载这里的文件。
同样的文件在[魔搭](https://www.modelscope.cn/models/sebponk/open-tools-models)和 [Hugging Face](https://huggingface.co/penghui002/open-tools-models) 各放一份，国内用户优先从魔搭下载。

AI models and runtime used by [Open Tools](https://github.com/Bulua/open-tools). The app downloads them on demand; you don't need to download anything here by hand.
The same files are mirrored on [ModelScope](https://www.modelscope.cn/models/sebponk/open-tools-models) and [Hugging Face](https://huggingface.co/penghui002/open-tools-models).

## 文件 / Files

所有文件都没有修改，只是转存；各自的许可证见 [LICENSE.md](LICENSE.md)。
All files are redistributed unmodified under their original licenses; see [LICENSE.md](LICENSE.md).

| 文件 / File | 用途 / Use | 来源 / Source | 许可证 / License |
| --- | --- | --- | --- |
| `models/u2netp.onnx` | 去背景（轻量）/ Background removal (light) | [rembg](https://github.com/danielgatis/rembg) · [U²-Net](https://github.com/xuebinqin/U-2-Net) | Apache-2.0 |
| `models/silueta.onnx` | 去背景（标准）/ Background removal (standard) | [rembg](https://github.com/danielgatis/rembg) · U²-Net | Apache-2.0 |
| `models/isnet-general-use.onnx` | 去背景（精细）/ Background removal (fine) | [rembg](https://github.com/danielgatis/rembg) · [IS-Net (DIS)](https://github.com/xuebinqin/DIS) | Apache-2.0 |
| `models/u2net_human_seg.onnx` | 人像抠图 / Portrait segmentation | [rembg](https://github.com/danielgatis/rembg) · U²-Net | Apache-2.0 |
| `models/lama.onnx` | 图片去水印（精细）/ Watermark removal (fine) | [LaMa](https://github.com/advimman/lama) · ONNX: [Carve/LaMa-ONNX](https://huggingface.co/Carve/LaMa-ONNX)（`lama_fp32.onnx`） | Apache-2.0 |
| `models/migan.onnx` | 图片去水印（轻量）/ Watermark removal (light) | [MI-GAN](https://github.com/Picsart-AI-Research/MI-GAN) · ONNX: [andraniksargsyan/migan](https://huggingface.co/andraniksargsyan/migan) | MIT |
| `runtime/<version>/win-x64/onnxruntime.dll` | AI 运行环境 / AI runtime | [ONNX Runtime](https://github.com/microsoft/onnxruntime)（NuGet `Microsoft.ML.OnnxRuntime.DirectML`） | MIT |
| `runtime/<version>/win-x64/DirectML.dll` | 显卡加速 / GPU acceleration | [DirectML](https://github.com/microsoft/DirectML)（NuGet `Microsoft.AI.DirectML`） | Microsoft DirectML License |