About Silicon Network

What it is

Silicon Network brings compute from many places into one: independent operators, decentralized networks, GPU clouds and the large public clouds. Each of them has its own console, prices and terms.

You start from the workload instead of the source. Say what you want to run and what matters most, price or speed, and choose from the sources that can run it.

What works today

AI inference. The Silicon Network API serves open models to clients that already speak the OpenAI, Claude or Ollama formats. Requests are served by operators' Macs running the Silicon Network engine on Apple Silicon.

The network is waiting for its first operators. No Mac has joined yet, so the API cannot serve a request until one does.

What is planned

Training, rendering, mining, trading bots and servers, and sources beyond our own operators. Each workload is listed only once a source for it is connected and checked. Until then the chooser says it is not available.

Why Macs first

Apple Silicon shares one pool of memory between the CPU and the GPU, so a Mac with a lot of memory can hold a large open model. Most of these machines sit idle for much of the day. The engine puts that time to work on inference.

The engine runs models with MLX, Apple's open source array framework. The models are open models, published by the people who trained them.