https://github.com/onion99/minegpt

MineGPT 是一个基于Kotlin Multiplatform 开发的本地小型语言模型(SLM)对话应用; MineGPT is a lightweight local SLM (Small Language Model) chat application built with Kotlin Multiplatform. It aims to provide a cross-platform, performant, and user-friendly AI assistant experience.
https://github.com/onion99/minegpt

ai android chatbot chatgpt compose compose-multiplatform desktop-ai gguf kotlin-multiplatform large-language-models llamacpp local-ai mobile-ai

Last synced: 8 days ago
JSON representation

MineGPT 是一个基于Kotlin Multiplatform 开发的本地小型语言模型(SLM)对话应用; MineGPT is a lightweight local SLM (Small Language Model) chat application built with Kotlin Multiplatform. It aims to provide a cross-platform, performant, and user-friendly AI assistant experience.

• Host: GitHub
• URL: https://github.com/onion99/minegpt
• Owner: Onion99
• License: gpl-3.0
• Created: 2025-03-27T04:09:00.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2025-05-07T11:01:08.000Z (10 days ago)
• Last Synced: 2025-05-07T12:19:58.553Z (9 days ago)
• Topics: ai, android, chatbot, chatgpt, compose, compose-multiplatform, desktop-ai, gguf, kotlin-multiplatform, large-language-models, llamacpp, local-ai, mobile-ai
• Language: C++
• Homepage:
• Size: 30.8 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![Kotlin](https://img.shields.io/badge/Kotlin-2.1.0-blue.svg?style=flat&logo=kotlin)](https://kotlinlang.org)

[![ComposeMultiplatform](https://img.shields.io/badge/Compose_Multiplatform-1.7.3-blue.svg?style=flat)](https://www.jetbrains.com/compose-multiplatform/)

![ProjectBanner](docs/project_cover.webp)

README 文件語言: [English](/docs/README_EN.md) | [中文](/README.md)

## 📜 简介

MineGPT 是一个基于Kotlin Multiplatform 开发的本地小型语言模型(SLM)对话应用

![ProjectApp](docs/project_chat.webp)

## 💠 测试模型

> 可以从这里下载 https://huggingface.co/models?library=gguf

- https://huggingface.co/legraphista/Qwen2-7B-Instruct-IMat-GGUF/blob/main/Qwen2-7B-Instruct.IQ1_M.gguf

- https://huggingface.co/unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/resolve/main/DeepSeek-R1-Distill-Qwen-1.5B-Q2_K.gguf

## 🚀未来计划

- 🖥️ 先进一步完善Desktop的使用

- ⚙️ 添加JNA来构建CMake

- 🌐 更多平台支持

- 🌍 多语言

- 💾 数据存储

- ⬇️ 应用下载SLM模型(GGUF模型)

## 🧩 依赖组件

感谢以下开发者,谢谢你们的付出

- Kotlin Multiplatform (KMP)

- Compose Multiplatform (CMP)

- JetBrains Adaptive Layout

- JetBrains Navigation Compose

- JetBrains Lifecycle ViewModel

- Koin

- Ktor

- Coil

- BuildKonfig

- Okio I/O

- llama.cpp

## 🙏 感谢

* [ggerganov/llama.cpp](https://github.com/ggerganov/llama.cpp) 是一个纯 C/C++ 框架，用于执行机器学习模型。它提供了一个原始的 C 样式 API 来与 LLM 交互转换为 GGML/llama.cpp 原生的 GGUF 格式

* [shubham0204/SmolChat-Android](https://github.com/shubham0204/SmolChat-Android) 是一个Android平台上运行SLM模型的开源项目,本项目也学习于此