xming521 /WeClone
Build a portrait of your life. Assistants that know you better. Clones that sound more like you.
★ 18,406⑂ 1,553👁 76⑃ 124● 123▤ 15
+0
Daily
18,406
stars
1,553
forks
76
watchers
124
prs
123
issues
15
releases
README
Build a portrait of your life. Assistants that know you. Clones that talk like you.
简体中文|
English|
Project Homepage |
Documentation
WeClone distills your identity, goals, preferences and current situation from your digital life into a life portrait, organized into domains such as work, family and health, that you can review, trace and share on your own terms. The portrait serves as long-term memory for your personal assistants and will power clones with your personal style.
✨ Core Features
- 🗺️ Interactive life portrait: your portrait is organized by domain and topic. Every fact traces back to its source and can be reviewed and edited. Extraction runs through Codex or any OpenAI-compatible API.
- 🎭 Your own AI clone: fine-tune an open model on your chat history and deploy it to Telegram, Discord, Slack, Feishu and more through AstrBot or LangBot. Training data is filtered for personal information.
- 🧠 Connect personal assistants: Claude Code, Codex, dots, Muse, Grok Bot and other personal assistants use your portrait through MCP or their memory. You choose which domains and privacy levels they get.
- 🗣️ Speaking style analysis: 179 style features in ten categories, covering vocabulary, syntax, tone, emotion, rhetoric and conversational rhythm.
- 🤖 Agent clone (in development): no training required. A large model drives the clone in real time from your portrait, memories and speaking style.
- 🔒 Privacy first: encrypted local storage by default, a privacy level on every fact, and full control over what is exported and shared.
[!NOTE]
WeClone now goes beyond fine-tuned clones. The original fine-tuning workflow remains fully supported, see Fine-tuned clone.
The portrait and its extraction prompts are currently in Chinese, and English chats have not been tested yet. The web UI is available in English and Chinese.
📰 Recent Updates
- [26/11/03] Life portrait: build-profile extracts, organizes and summarizes your portrait in one command, ready for fact-by-fact review in the web UI
- [26/11/03] Assistant access: Claude Code, Codex and other personal assistants read your portrait over MCP
- [26/11/03] Encrypted storage by default, and token access for chat bots
- [25/07/10] Telegram added as a data source
- [25/06/05] Fine-tuning on image data
📥 Supported Data Sources
The life portrait reads text messages only. The table shows the message types the fine-tuned clone supports. See Preparing data for exporting and setup.
| Platform | Text | Images | Voice | Video | Animated Emojis/Stickers | Links (Sharing) | Quote | Forward | Location | Files |
|----------|------|--------|-------|-------|-----------------|-----------------|-------|---------|----------|-------|
| Telegram | ✅ | ✅ | ❌ | ❌ | ⚠️ Converted to Emoji | ❌ | ❌ | ✅ | ✅ | ❌ |
| WhatsApp | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 |
| Discord | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 |
| Slack | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 | 🚧 |Top issues
#188Telegram 支持从设置导出的聊天记录8 comments · 2025-07-11#177Add Sticker/meme Feature6 comments · 2025-07-05#105部署到Google Colab的移植版脚本6 comments · 2025-05-20#37微调效果反馈6 comments · 2025-04-17#92数据预处理出现问题5 comments · 2025-05-18#47MacOS M3不支持4 comments · 2025-05-04#46Mac设备支持4 comments · 2025-04-30#200请问是否支持A770 16G,还有有没有推荐可以导出微信聊天记录的工具3 comments · 2025-09-01
Discussions
#217不在本地开发0 comments · 2026-05-05#194请问demo测试的时候报错是什么问题?1 comments · 2025-08-07#181训练过程中可以暂停或者暂存,下次继续训练吗1 comments · 2025-07-07#180wsl2中部署项目后 会出现如下错误1 comments · 2025-07-06#152用QQ聊天记录训练?0 comments · 2025-06-10#76windows系统能够部署吗?1 comments · 2025-05-15#62有何商业化的点子?2 comments · 2025-05-14#54低性能电脑上想部署有什么解决办法?1 comments · 2025-05-13
Latest release
v1.0.4 v1.0.4 2026-11-09🎉 What's Changed WeClone 1.0.4 tells you what a build will cost and how long it will take, before you start! Cost and time estimate - The confirm step of the Build portrait page estimates a build's cost, time and number of requests. A trial sends about 10 real requests to the model service to measure them, and the build reuses their answers. - Known providers such as DeepSeek are priced at their official rates, other services at the common price across providers. For a service with off-peak prices, the estimate follows the hours a build started now would run in. - With Codex, the estimate sh