Skip to main content

Teale

Turn your Mac into an AI inference node. Run models locally, earn by sharing compute, build on an open API.

Get started​

Use Teale​

Run AI models locally on your Mac with zero configuration. Chat with state-of-the-art models in seconds.

Get started with chat

Earn with Teale​

Share your idle compute with the network and earn Teale Credits for being online and for serving requests. Credits redeem AI inference and transfer to other users.

Start earning

Build on Teale​

OpenAI-compatible API at localhost:11435. Drop-in replacement for any app that speaks OpenAI.

Explore the API

What is Teale?​

Teale is a decentralized AI inference network. Macs run inference, iPhones consume it, all peer-to-peer with no central servers.

Apple Silicon native. Teale runs optimized inference on M1 and later chips, taking full advantage of the unified memory architecture and Neural Engine. Cross-platform support is available via teale-node, a Rust binary that runs on Linux, Windows, and Android with support for NVIDIA, AMD, and Intel GPUs.

No central servers. Every node connects directly to peers. There is no cloud backend, no data collection, no single point of failure. Your conversations never leave your device unless you explicitly route them through the network.

End-to-end encrypted. All network traffic between nodes uses E2E encryption with Ed25519 identities. No one --- not even relay operators --- can read your prompts or completions.

Closed-loop credit economy. Availability and serving earn Teale Credits. Credits redeem AI inference and transfer to other users — they don't cash out. Paying customers top up the pool with USDC; the network never pays USDC back out to users or devices.

Open API. Teale exposes an OpenAI-compatible API on localhost:11435. Any tool that works with OpenAI --- LangChain, LlamaIndex, Continue.dev, custom apps --- works with Teale out of the box.

How it works​

  1. Install Teale on your Mac (download) or any platform (cross-platform).
  2. Models download automatically based on your available RAM.
  3. Chat locally through the menu bar app, CLI, or API.
  4. Optionally join the network to serve inference to others and earn Teale Credits.

Teale organizes nodes into four network tiers --- local, LAN, personal trusted network, and the wider Teale network --- so traffic stays as close to you as possible.

Next steps​