Let's Open.
Build AI people can trust.
A builder-first map of open source AI projects, tools, models, infrastructure, and agent systems worth building on, with machine-readable endpoints for AI agents.
Find the next open component to build with.
Use letsopen.ai as a map for choosing open models, agent frameworks, local runtimes, infrastructure, and workflow tools.
Start with a practical blueprint
Assemble the smallest useful open agent app before adding frameworks, memory, tools, and multi-agent orchestration.
Read the blueprint -> Choose a stackAssemble an open AI app stack
Understand the layers from models and inference to retrieval, orchestration, evaluation, and product interfaces.
Read the stack guide -> Ship with toolsPick practical builder tools
Compare the open source tools that are maintained, composable, documented, and useful for real AI application work.
See the tool list -> Evaluate a projectScore what is worth adopting
Use a practical rubric for judging maintenance, source posture, deployment control, composability, governance, and eval coverage.
Use the rubric -> For AI agentsFetch machine-readable maps
Use structured indexes, text mirrors, llms.txt, and the agent manifest to route builder questions efficiently.
Open agent access ->Open source AI is about more than code.
It is about trust, transparency, control, and making AI something people can actually govern.
Transparency
People should be able to inspect the systems they depend on.
Security
Open systems can be audited, constrained, and adapted to real-world needs.
Sovereignty
Builders and organizations should be able to run AI on their own terms.
Access
The future of AI should not belong only to the largest closed platforms.
Topics
View all →Open Source AI Agents
The landscape of open source AI agents: autonomous systems that can reason, plan, and take action. From personal assistants to multi-agent orchestration.
Coding Agents
AI agents that write, review, and ship code. From CLI tools to IDE integrations — the new wave of AI-powered software development.
OpenClaw
An AI-native personal operations platform. Multi-agent orchestration, coding agents, personal automation — running locally, operated by AI.
Local AI
Running AI models and agents on your own hardware. Privacy-first, offline-capable, fully under your control.
Open Models
Open-weight and open-source language models. The foundations that power the open source AI ecosystem.
AI Infrastructure
The open source tools and platforms that power AI development. From training to deployment, inference to monitoring.
Open Source AI Tools
Practical tools for builders. The best open source software for working with AI models, agents, and workflows.
Workflows & Orchestration
Multi-agent systems, AI pipelines, and orchestration patterns. How to make AI agents work together.
Latest
OpenClaw: what it is and why it matters
An AI-native personal operations platform that orchestrates agents for your digital life. Here's what OpenClaw does and why it's worth paying attention to.
Why AI needs to be open
Open source in AI is not just a developer preference. It is how we get transparency, security, sovereignty, and broader adoption.
Guides
All guides →Best open source AI tools for builders (2026)
A curated, opinionated guide to open-source-first AI tools and open-adjacent builder tools that are actually worth using in 2026.
How to evaluate open source AI projects
A practical framework for deciding which open source AI projects are worth using, watching, or contributing to.
Comparisons
All comparisons →Codex CLI vs Claude Code vs Gemini CLI
A practical comparison of the three major terminal-based coding agents. Which one should you use, and when?
Open source AI agents vs closed AI assistants
The real tradeoffs between open and closed AI assistants: trust, control, speed, convenience, and long-term leverage.
Get the builder brief
Weekly signal on open source AI agents, app stacks, source posture, and tools worth building with.