Build on OpenGrid
Everything you need to consume open-model inference, contribute compute, or implement the wire protocol.
The inference API follows the OpenAI v1 contract. Most applications need only a new
baseURL and OpenGrid API key.Consumer quickstart
Create a key and stream your first chat completion.
Provider setup
Enroll a device and apply safe resource limits.
Protocol guide
Version negotiation, transport, signatures, and receipts.
Privacy & security
Understand the trust boundary before sending data.
Endpoints
The OpenAI-compatible surface is rooted at https://api.opengrid.computer/v1. Dashboard traffic uses a separate browser session and should not be called from production applications.
GET /v1/models
POST /v1/chat/completions
POST /v1/completions
POST /v1/embeddings