💻 Dev

Koog is a JVM framework for building reliable, enterprise-grade AI agents.

gptme is a personal AI agent that runs in the terminal, capable of writing code, using terminal commands, and browsing the web.

Lemonade is a local AI server that enables running optimized LLMs on your own GPU and NPU.

OpenMemory is a local, persistent storage solution for long-term memory in LLMs and AI agents.

AdalFlow is a library for building and automatically optimizing language model applications.

Langroid is a lightweight and extensible Python framework for building LLM-based multi-agent applications.

RubyLLM is a unified Ruby framework supporting multiple AI providers for building chatbots, AI agents, and more.

LazyLLM is a low-code development tool for building multi-agent large language model applications.

fast-agent is a flexible LLM interaction tool for coding, development, and evaluation.

vocode-core is an open-source library for building voice-based LLM applications.

Free AI code completion plugin for local or API hosting, compatible with Visual Studio Code.

AgentScope Java is a agent-oriented programming framework for building applications based on large language models.

Acontext is an open-source skill memory layer for AI agents, automatically capturing and storing the learning outcomes of agents.

Transform approved specifications into long-term autonomous AI coding agents.

GoClaw is a multi-tenant AI agent platform supporting multiple LLM providers and messaging channels.

Neo.mjs is a self-evolving software organism that autonomously operates within applications through a multi-model cluster.

A lightweight and powerful multi-agent workflow and voice agent framework.

MemMachine is an open-source tool that provides a long-term memory layer for AI agents.

BotSharp is an open-source AI multi-agent framework based on .NET.

A unified virtual file system that enables AI agents to access data across services.

Open-source enterprise-grade agent designed for Chinese teams, supporting conversation, task orchestration, and automated integration.

A framework-agnostic, Git-based standard for defining AI agents.

One-click commenting, code transformation, code generation from UI designs, and AI-powered batch file processing in VSCode.

Rapid-MLX is the fastest local AI engine for Apple Silicon, supporting multiple models and tool calls.