LibreAI / AI Chat / AnyModelBot
// ai-chat · modelflare-ai/AnyModelBot

AnyModelBot

AnyModelBot is an open-source ai chat project by modelflare-ai (modelflare-ai/AnyModelBot on GitHub). It has 2,742 GitHub stars and 45 forks, is released under the MIT license, and was last updated on GitHub on 9 November 2026.

Local-first personal AI agent for your own computer — any model, pay-per-use, on WeChat, Telegram, Slack and more, MIT

★ 2.7k stars⑂ 45 forksMITupdated 2026-11
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

GitHub stars
2,742
Forks
45
License
MIT
Last updated on GitHub
9 November 2026
Category
AI Chat
Listing
Full install guide
Official site
Official site
// how it works

How it works

  1. Install it

    Download the desktop app from the Releases page, or run the Docker image.

  2. Sign in once

    First launch opens a browser to sign in (or register) with Modelflare; or add your own provider API key instead.

  3. Pick a model and chat

    Choose a default model, then chat in the app or connect WeChat, Feishu, Telegram, Slack or Discord.

// features

Why people love it

Any model, no lock-in

GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more via Modelflare, or direct with your own keys.

Local-first

Conversations, memory, files and API keys stay on your device.

Lives in your chat apps

WeChat, Feishu, Telegram, Slack and Discord — the bot works where your team already talks.

Acts with permission checks

Reads and writes files, runs commands and operates the desktop, asking first outside its workspace.

Ready out of the box

Background tasks, scheduled rules and "needs you" prompts need no extra configuration.

MIT, no subscription

Open source and free to run; you pay only for model usage.

// run it

Run it yourself

  1. Step 1 — Pull the image

    Fetch the Docker image from the project’s container registry.

    docker pull ghcr.io/modelflare-ai/anymodel-bot
  2. Step 2 — Run it

    Start the container with a persistent data volume on port 3600.

    docker run -d --name amb -p 3600:3600 -v amb-data:/data ghcr.io/modelflare-ai/anymodel-bot
  3. Step 3 — Get the login token

    Read the UI token from the container, then open the local console with it.

    docker exec amb cat /data/ui-token
// use cases

What you can build

  • A personal agent that lives on your own computer
  • Chat with any model without vendor lock-in
  • Let the bot work inside your team chat apps
  • Automate files and desktop tasks with approval gates

For comparison, the same shelf also holds botato — A local-first grokbot replacement and antigravity-telegram-suite — Antigravity Telegram bot. The complete list is the AI Chat shelf. Before running a model locally, check how much VRAM you need for local LLMs — model size has to match your hardware.

// faq

Questions, answered

Is AnyModelBot free?

The app is MIT licensed with no subscription. Model usage is pay-per-use via Modelflare, or billed by your own provider keys.

Which models does it support?

70+ through Modelflare, or direct API keys for OpenAI, Anthropic, Google and other providers.

Where does my data live?

On your device — conversations, memory, files and API keys stay local.

Which chat apps can it join?

WeChat, Feishu, Telegram, Slack and Discord.

Docker or desktop?

Both — a desktop installer for Windows, macOS and Linux, or the ghcr.io/modelflare-ai/anymodel-bot image.

// keep exploring

Similar tools

AI Chat

botato

A local-first grokbot replacement.

★ 58 starsApache-2.0
View guide →
AI Chat

antigravity-telegram-suite

Antigravity Telegram bot — remote-control your AI agent via Telegram.

★ 172 starsMIT
View guide →
AI Chat

vibe_tavern

A modern, local-first AI roleplay platform you host yourself.

★ 74 starsAGPL-3.0
View guide →
AI Chat

LobeChat

Open-source AI chat and super-agents for your team — self-host it, plug in any model

★ 85.6k starsLobeHub Community
View guide →
AI Chat

Open-LLM-VTuber

Talk to any LLM hands-free with voice interruption and a Live2D avatar, running locally on your computer.

★ 14.0k starsMIT
View guide →
AI Chat

app

Open ai chat for privacy-first ai roleplay & byok (bring your own key) &.

★ 139 starsAGPL-3.0
View guide →
// alternatives

Alternatives in AI Chat

The most-starred other AI Chat projects in the LibreAI catalogue — the realistic shortlist if AnyModelBot is not the right fit.

  • NextChat★ 88.8k

    ✨ Zero-config AI chat assistant. No API key needed — sign up and instantly chat with GPT-5, Claude 4, Gemini 2.5, DeepSeek & 100+ top models. Pay-as-you-go s…

  • LobeChat★ 85.6k

    Open-source AI chat and super-agents for your team — self-host it, plug in any model

  • cherry-studio★ 52.5k

    AI productivity studio with smart chat, autonomous agents and 300+ assistants for frontier LLMs

  • LibreChat★ 45.3k

    Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI…

  • chatbox★ 41.9k

    Powerful AI Client

// more in ai-chat

More in AI Chat

Nearby AI Chat projects in the catalogue, closest in stars to AnyModelBot — keep exploring the shelf either side of it.

  • Sidekick★ 3.3k

    A native macOS app to chat with a local LLM over your own files, folders and websites.

  • super-agent-party★ 2.7k

    Self-hosted all-in-one AI companion: VTuber-style assistant meets agent platform.

  • big-AGI★ 7.1k

    big-AGI — a multi-model AI suite with Beam multi-model chats, personas, text-to-image, voice and PDF import.

  • HuixiangDou★ 2.5k

    HuixiangDou — LLM knowledge assistant built for group chats, with rejection so it answers only when it should

  • Octop★ 8.0k

    Octop — Tencent Cloud's self-hosted AI assistant: multi-user, multi-agent, workspace memory included.

  • Klee★ 1.8k

    Native macOS chat app running LLMs fully locally on Apple Silicon.

Get AnyModelBot

Free and open — grab the source, star the repo, and start creating.