heatdrop.ai

A locally-runnable voice-interactive AI VTuber that can see your screen, interrupt you mid-sentence, and sulk in desktop pet mode.

Self-hosted chat UI that unifies OpenAI, Anthropic, Google, AWS, and two dozen other providers under one roof.

AstrBot is a Python framework for wiring LLMs into QQ, WeChat, Telegram, Slack, and a dozen other chat platforms without writing platform glue.

Electron GUI for Hermes Agent, an AI assistant that can message you on 16 platforms and won't stop learning.

A locally-run frontend that wrangles dozens of LLM APIs, image generators, and TTS into one obsessively customizable interface.

A Node.js bridge that turns your personal WeChat into a multi-model AI switchboard—at the cost of dancing with Tencent's ban hammer.

Jarvis treats conversation as context, not a series of disconnected commands.

RikkaHub is a native Android client that lets you switch between OpenAI, Google, Anthropic, DeepSeek, and dozens of other LLM providers without juggling separate apps.

Khoj wires any LLM into your notes, files, and messaging apps so you can query your own data instead of begging ChatGPT to remember context.

Aurora wraps OpenAI's web ChatGPT into an OpenAI-compatible API you can host yourself, no subscription required.

A headless component library for AI chat interfaces, because streaming and auto-scroll are harder than they look.

YOLO turns your note vault into an agent workspace where AI can read, write, edit, and organize instead of just chatting.

A C# toolkit that wires lip-sync, facial expressions, and speech recognition to multiple LLMs so your 3D character can actually hold a conversation.

Voice-controlled desktop assistant wiring together pyttsx3, OpenCV, and a dozen APIs to handle your email, news, weather, and YouTube downloads.

A Word add-in that goes beyond chat, giving AI agents direct access to read, write, and reformat your .docx files via 25+ built-in tools.

A JavaScript class for real-time lip-sync with full-body 3D avatars, built on Three.js and used in everything from AI dating profiles to Twitch adventures.

A Java-based bot platform that predates the LLM era and still runs on JUnit and Objective-C.

A thin wrapper that lets beginners train chatbots and phoneme converters without drowning in TensorFlow boilerplate.

ChatLearner bolts rule-based reasoning onto TensorFlow's seq2seq model so it can tell time, solve arithmetic, and read jokes—things pure neural chatbots flunk.

A static-site generator that scrapes HN, extracts article content, and feeds it to ChatGPT for summaries with auto-generated illustrations.