Chat Assistants

Wechaty abstracts away the mess of messaging-platform APIs so you can write a chatbot once and run it anywhere.

Leon is an open-source AI assistant that prefers using tools and local context over cloud-based guesswork.

It is an open-source bridge between large language models and the instant messaging apps you already tolerate.

AnythingLLM bundles document chat, AI agents, and multi-user support into one local-first application so you don't have to duct-tape five tools together.

Chatbox aggregates ChatGPT, Claude, Gemini, and Ollama into one cross-platform desktop app that keeps your conversation data local.

To wrap every LLM backend, RAG pipeline, and enterprise auth scheme into a single self-hosted interface.

A maintainer cataloged every Chinese NLP repo they touched into a single, obsessively categorized list so others wouldn’t have to hunt.

To let you own a game-playing, voice-chatting AI companion that doesn't vanish when the stream ends.

Cherry Studio exists to end the browser-tab shuffle by putting every major cloud and local LLM into one ready-to-run desktop client.

Onyx bundles chat, RAG, agents, and sandboxed code into a single self-hosted interface so teams can use any LLM without shipping data to a SaaS black box.

It fans out one prompt to multiple models, lets them rank each other’s anonymous answers, and has a chairman synthesize the final response.

SillyTavern gives AI hobbyists a single, local interface to command text generators, image engines, and voice models across dozens of backends without losing control of their prompts or data.

Jan is a desktop chat client that makes running local LLMs as mundane as using ChatGPT, while quietly exposing an OpenAI-compatible API for your other tools.

LibreChat bundles every major LLM provider into a single self-hosted chat platform so teams don't have to choose—or leak data.

Hermes WebUI gives the Hermes autonomous agent a browser interface without adding a JavaScript build pipeline.

Odysseus is a self-hosted AI workspace that replicates the ChatGPT experience on your own hardware while also handling email, calendars, documents, and deep research without shipping your data to third-party clouds.

Khoj exists to unify your documents, web search, and chosen LLM into a single self-hosted brain accessible from Emacs to WhatsApp.

A Python bot framework that wires DeepSeek, Claude, or Ollama into WeChat, QQ, and Telegram with workflows, image gen, and voice replies.

PandasAI wraps pandas in an LLM so you can ask CSVs plain-English questions instead of writing groupby statements.

XiaoZhi is open-source ESP32 firmware that streams voice to LLMs like Qwen and DeepSeek, then uses the MCP protocol to let those models control your LEDs, servos, and smart home gear.