Artificial Intelligence

Run large language models locally

Open Source OpenAI alternative, self-host AI models

LangChainGo is a framework for developing applications powered by language models

Go implementation of the Model Context Protocol for building MCP servers and clients in Go

Framework created by Google for building AI-powered and agentic applications in idiomatic Go

Framework for building LLM-based multi-agent systems

AI gateway exposing a unified OpenAI-compatible API across OpenAI, Anthropic, Gemini, Groq, xAI, Ollama and other providers, with routing, usage tracking, rate limits, and guardrails

Embeddable vector database for Go with Chroma-like interface and zero third-party dependencies. In-memory with optional persistence

CLI proxy that reduces LLM token usage by 60-90% with declarative YAML filters. Drop-in for Claude Code, Cursor, Copilot, and Gemini. rtk alternative in Go

A Go library for building stateful, multi-actor applications with LLMs, built on the concept of LangGraph，with a lot of builtin Agent architectures

OpenAI-compatible LLM gateway that routes requests across 30 providers with fallback, rate limiting, budgets, guardrails, and observability

Go SDK for building AI applications. One SDK, 20+ providers. Inspired by Vercel AI SDK

Framework for building security engineering AI agents in Go. Read-only by construction, built-in sandbox, 45 agent blueprints for AWS, GCP, Azure, K8s, GitHub & GitLab deep research

Agentic harness focused on terminal workflows, following UNIX principles. No external dependencies, ideal for scripting/prototyping agentic apps, engine is also exposed as a package

Manage, load-balance, and failover packs of Ollamas

Framework for building real-time voice AI agents over WebRTC, wiring speech-to-text, LLMs, and text-to-speech into a streaming pipeline

A context-efficient terminal-based AI coding agent. Provider agnostic, supports MCPs, Agent Skills, Subagents, and more. Comes with a simple and straightforward TUI

MCP server providing AI assistants with web search, content extraction, and multi-source research capabilities. Single binary, 5 search providers with circuit-breaker failover, 4-tier scraping pipeline

CLI that scans AI agent session transcripts to identify and safely quarantine unused skills, MCP servers, and agents across Claude Code, Codex CLI, Hermes, OpenCode, Cursor, and OpenClaw

Framework for building stateful AI agents in Go

Multi-agent orchestration & workflow engine. Declarative YAML workflows, LLM coordinator with hub-and-spoke mailboxes, race-safe delivery. One YAML file, one Go binary. Runs on any goai-supported provider

Go library for driving the Claude Code CLI non-interactive prompt surface from Go programs

AI Agent runtime engine with long-lived sessions for Claude Code, OpenCode, pi-mono and other CLI AI tools. Provides full-duplex streaming, multi-platform integrations, and secure sandbox

A Go toolkit for building AI agents and applications across multiple providers with unified LLM, embeddings, tool calling, and MCP integration

OpenTelemetry-native LLM observability and budget guardrails for cost-constrained production environments

AI gateway for routing, securing, and monitoring LLM traffic across 10+ providers. OpenAI-compatible API, WASM policy plugins, canary rollouts, real-time dashboard

YAML-driven multi-agent AI runtime for Go with Erlang-style supervision, MCP tool server support, and a CLI

AI Agent execution runtime with event sourcing, checkpoint recovery, and At-Most-Once execution guarantee. Written in Go

Model Context Protocol (MCP) server wrapping the Antigravity CLI to run prompts and peer reviews

LocalStack-style local emulation of OpenAI and Gemini APIs; single Docker container, llama.cpp + Gemma 3 backend

Terminal-based AI workflow engine with YAML-driven pipelines, 20+ LLM providers (DeepSeek, Qwen, GLM, Mistral, etc.), and a TUI for workflow management

Idiomatic Go port of CrewAI (multi-agent orchestration). Zero dependencies, stdlib only

AI-native content backend with typed lifecycle management, native MCP tools for every content type, and zero runtime dependencies

Meaning-based search over PDF, Markdown, DOCX, source code, and other file types, using generative AI embedding models to vectorize files into a vector database

Official Go client SDK for the Dakera self-hosted agent memory server, providing typed interfaces for memory store/recall, session management, namespace operations, and decay configuration

Framework for building dependable AI agents with typed dependencies, zero external dependencies, and native MCP support

Terminal contribution heatmap, streak tracker, and token leaderboard for AI coding assistants