Get started
Compute

Workload

AI inference

Open models behind one API that OpenAI, Claude and Ollama clients already speak: chat, embeddings, rerank, images and audio.

Available

What it is

Inference is running a trained model on new input: a prompt goes in, an answer comes out. It is the part of AI that applications pay for on every request.

Requests are short and many. What matters is the time to the first token, the speed of the rest, and the price per token.

What we offer today

The Silicon Network API serves open models for chat, embeddings, rerank, speech to text, text to speech and images. It speaks the OpenAI, Claude Messages and Ollama formats, so most clients need only a new base URL and a key.

Requests are served by Silicon Network operators: independent people and companies running the engine on their own Macs. No Mac has joined yet, so the API cannot serve a request until the first operators are online.

Billed per token at the rates on the pricing page.

Sources

The kinds of source that can run this workload, now or later. No provider is named until it is connected.

  • Silicon Network operators

    Independent machines that operators add to the network, today Macs running the Silicon Network engine

    Serving today

  • Decentralized networks

    Open networks of independent machines

    Planned

  • GPU clouds and marketplaces

    Companies that rent out GPU machines

    Planned

  • Public clouds

    The large general-purpose clouds

    Planned

Outside sources would serve the same API, so a request could go to whichever source fits its price and speed. None is connected yet.

Status

Available: the API is live, and serving starts when the first Macs join.

Open in the chooserRead the quickstart

The chooser opens with AI inference selected.

Other workloads