Free AI Tools for Open Source

An open-source, OpenAI-compatible LLM gateway you run yourself, providing a single endpoint for 40+ providers with virtual keys, budgets, and usage tracking.

Let coding agents see your browser by capturing screenshots of open tabs via a CLI and browser extension.

Macro combines email, chat, docs, tasks, calls, and CRM into one open-source app with shared AI memory.

Run any LLM locally on your Mac faster than LM Studio or Ollama, with chat, coding agents, images, video, and voice, all fully offline and open source.

Valmis is an open-source AI agent platform for automating business workflows securely, with isolated containers, proxy API access, and multi-step automation.

A secure file system and code execution sandbox for AI agents with built-in job queues, Docker isolation, and execution history.

A feature-rich Pomodoro timer that lives inside the Claude Code terminal, keeping focus without context switching.

A unified Python library that combines MCP servers, agent skills, custom functions, and HTTP endpoints into a single tool set for building AI agents with any LLM.

Local-first memory for AI coding agents. Automatically stores and recalls project context via MCP, offline, with ~35ms recall.

A self-hosted, browser-based AI CSV analyzer that uses multiple AI providers to analyze and visualize CSV and Excel data.

An open-source, OpenAI-compatible LLM gateway you run yourself, providing a single endpoint for 40+ providers with virtual keys, budgets, and usage tracking.

Let coding agents see your browser by capturing screenshots of open tabs via a CLI and browser extension.

Macro combines email, chat, docs, tasks, calls, and CRM into one open-source app with shared AI memory.

Run any LLM locally on your Mac faster than LM Studio or Ollama, with chat, coding agents, images, video, and voice, all fully offline and open source.

Valmis is an open-source AI agent platform for automating business workflows securely, with isolated containers, proxy API access, and multi-step automation.

A secure file system and code execution sandbox for AI agents with built-in job queues, Docker isolation, and execution history.

A feature-rich Pomodoro timer that lives inside the Claude Code terminal, keeping focus without context switching.

A unified Python library that combines MCP servers, agent skills, custom functions, and HTTP endpoints into a single tool set for building AI agents with any LLM.

Local-first memory for AI coding agents. Automatically stores and recalls project context via MCP, offline, with ~35ms recall.

A self-hosted, browser-based AI CSV analyzer that uses multiple AI providers to analyze and visualize CSV and Excel data.

AXIOM is a bootable Rust kernel that optimizes transformer inference by replacing generic OS abstractions with inference-specific primitives.

A CLI tool that checks package versions for known vulnerabilities across npm, PyPI, and many other ecosystems, designed for AI agents to avoid using outdated packages.

Moxie is an open-source money agent that finds zombie subscriptions, duplicate charges, and missed refunds, then drafts fixes with your consent and logs every step in a tamper-evident audit trail.

An open-source ML pipeline that diagnoses car faults from audio by cleaning sound, embedding with CLAP, and training calibrated linear classifiers.

Kage is a git-native verification layer for coding agent memory that checks every claim against your code to prevent hallucinations and stale facts.

An open-source Agent UI library in VanillaJS for building agentic front-end experiences with WebMCP, SSE streaming, and customizable theming.

Open-source, privacy-first AI screen memory tool that captures and analyzes your screen activity locally to let you search and chat with your screen history.

Fleet is a local-first web console for creating, configuring, monitoring, and operating Dockerized Hermes AI agents across trusted machines.

Vibe-zsh is an AI-powered Oh-My-Zsh plugin that transforms natural language into shell commands using any OpenAI-compatible API.

Keeps your Mac awake only while AI coding agents are actively working, and lets it sleep when they finish.

An open protocol for capability-aware AI routing, plus a reference router that routes between local and cloud AI models based on task, cost, and latency.

NanoEuler is an open-source GPT-2-style language model built entirely from scratch in C/CUDA with hand-written backprop, BPE tokenizer, FlashAttention, and training pipelines.

A skill for coding agents that delegates coding, planning, review, and orchestration tasks to the zerostack CLI agent.

Open-source reliability testing harness for AI agent skills, computing pass@k scores across Claude Code, Codex, and Pi backends.

Open-source alternative to Claude in Slack: self-hosted AI agent that reads threads, answers questions, calls tools, and renders rich results.

AI-powered open-source pipeline that converts diarized audio (podcasts, interviews) into YouTube-ready MP4 with semantic B-roll, burned subtitles, and optional voice synthesis.

Secure data room for sharing confidential documents with enterprise-grade security and analytics.

A local-first MCP gateway that reduces token usage by ~90% by exposing 3 meta-tools instead of hundreds, working across all major AI clients.

An open-source CLI toolkit for simulating AI training math, estimating model size, memory, training time, and token budget.

Purr is a free, open-source, on-device voice-to-text dictation app for macOS Apple Silicon with no cloud round-trips.

Lightweight, themeable AI chat widget for websites built with vanilla JS, supporting streaming, voice, tool calls, and a plugin system.

Beautiful, AI-native markdown editor and LLM Wiki for creating and managing knowledge bases with AI assistance.

Browser-native AI content workstation that uses live webpage context and persona alchemy to generate targeted, multi-stage outputs, all running client-side with BYOK.

A transpiler from stateful imperative workflows to declarative DSPy programs for reliable, optimizable LLM steps with zero boilerplate.

Open source authorization engine for AI agents that provides confidence-aware gating, human-in-the-loop review, policy-as-code, and full audit trail.

peerd is the first AI agent harness native to the browser, allowing AI agents to drive tabs, run sandboxed compute, and share peer-to-peer with no backend.

HALO is a methodology and toolkit for recursively self-improving AI agent harnesses using RLMs and production traces.

Alma is a local-first MCP server that gives AI agents a private, consent-based memory of your personal facts and preferences.

Oak is an agent-native version control system with a lazy working tree, JSON-first CLI, and branch management designed for AI coding agents.

Skybridge is a full-stack TypeScript framework for building type-safe MCP apps and servers with React-powered UI.

OpenHuman is an open-source, personal AI super intelligence that offers a simple, private, and powerful agentic assistant with local memory and managed services.

A self-hostable workspace where humans and AI agents collaborate in shared rooms over a Nostr relay.

A local, zero-dependency dashboard for Claude Code that provides live token usage, context monitoring, lost-session recovery, full-text search, and phone-based approval of tool calls.

Velane is an open source AI agent code runtime that connects to your IDE via MCP, enabling autonomous creation, testing, and deployment of workflows with 800+ integrations.

Juno is a local voice layer for Mac that converts speech to clean text on-device, with voice actions, rewrites, and no subscription.

PII GUI is a local-first desktop application for detecting and redacting personal information in documents using regex and on-device ONNX models.

Claw Patrol is an open-source security firewall for AI agents that guards credentials, gates actions, and logs all traffic.

Memento is an AI tool that surfaces your important history from email, organizing it around people, projects, and topics that matter to you.

Free open-source desktop companion that runs a private AI backend on Mac/PC and connects MyLLM iOS app over trusted HTTPS via Tailscale.

MCP server for SharkClean / SharkNinja robot vacuums that enables agentic control via any MCP client: start cleanings, clean rooms, dock, and check status in plain language.