AnyModelBot
🤖 Open-source, local-first alternative to Grok Bot, OpenAI dots and Claude Duet. Your own AI bots on any model: GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more through one Modelflare key. Reach them in the app, WeChat or Feishu; they remember you, work in the background, run commands and use your computer. Pay per use.
- ai-agent
- ai-assistant
- chatbot
- claude
- computer-use
- deepseek
- desktop-app
- dsh
- electron
- feishu
- gemini
- gpt
- grok
- llm
- local-first
- modelflare
- multi-model
- openai-compatible
- personal-ai-agent
- Stars
- 1,343
- Forks
- 14
- + today
- +10
- Created
- 6d
Ranking data as of November 2, 2026 (UTC).
Star History
Today, hour by hour
Overview
🤖 Open-source, local-first alternative to Grok Bot, OpenAI dots and Claude Duet. Your own AI bots on any model: GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more through one Modelflare key. Reach them in the app, WeChat or Feishu; they remember you, work in the background, run commands and use your computer. Pay per use. It ranks #21 on GitTiger, gaining +10 stars on November 2, 2026 (UTC).
The project is written in TypeScript and has 14 forks. It was created 6d ago.
What it is
AnyModelBot is a TypeScript, Electron-tagged desktop AI agent app under the MIT license. The repository description pitches it as an open-source, local-first alternative to several bot products, giving you your own AI bots on many models (GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more) through one Modelflare key, with pay-per-use billing.
Capabilities described
- **Platforms:** the README summary says it runs on Windows, macOS and Linux. - **Bots as personas:** each bot has its own persona, default model, workspace, memory and avatar; you can create as many as you like. - **Reach:** chat in the app, or via WeChat, Feishu, Telegram, Slack or Discord. - **Actions:** it can read and write files, run commands and operate your desktop, asking first before leaving its workspace. - **Models and cost:** one model picker lists context length, capability flags and reference prices, with balance and daily usage shown in-app. - **Data boundary:** requests go through Modelflare to your chosen provider, and Modelflare keeps request logs; the rest stays on your device. Retention details are in its trust center.
Getting started
First launch opens a browser to sign in to Modelflare. For Docker or servers, paste an API key instead. Other steps: see the README.
git clone https://github.com/modelflare-ai/AnyModelBot.git
cd AnyModelBot
# see README for setupFrequently Asked Questions
Which models can it use?
Per the repository description, GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more, all through one Modelflare key.
Is everything kept local?
Not entirely. The README summary says requests are forwarded through Modelflare, which keeps request logs; other content stays on your device.
How do I sign in?
The first launch opens a browser to log in or register with Modelflare. On Docker or a server, paste an API key instead.
Which chat channels are supported?
The app itself, plus WeChat, Feishu, Telegram, Slack and Discord according to the README summary.