The Lebrel waitlist is live. We’re building private, uncensored AI for researchers, developers, and creators who want more freedom and control. Private beta is coming soon. Join the waitlist: early.lebrel.ai/

“Uncensored” tells you little about what’s behind the answer. A tampered model could make an agent run harmful commands. Knowing which weights are running matters. Lebrel is building uncensored access with model verification, so users have more than a provider’s word to go on.

The Lebrel leaderboard is live. See where you stand: early.lebrel.ai/leaderboard

Use Lebrel your way. Chat for working directly with its models. An OpenAI-compatible API to bring them into your app or agent. One platform. Your choice of how to use it.

Keep the question open. Lebrel’s own model editions have no added refusal layer. More room to examine an argument or develop a difficult idea. That’s the experience we’re building. Get on the waitlist: early.lebrel.ai

AI is becoming part of everything we do, but users still have too little control over it. We’re building Lebrel to change that. An AI experience where privacy, freedom of choice, and transparency come first. Built for people who want to do more with AI, on their own terms.

Meet Lebrel. An AI inference platform giving you more freedom and control over how you use AI. Private, uncensored model editions built for researchers, creators, and developers. Use Lebrel through Chat or build with its API. AI on your terms.