AI Stays Local

DocumentationReference

Local API

An OpenAI-compatible HTTP endpoint, bound to loopback, so existing tools can use a local model without knowing anything about this product.

The endpoint

Addresshttp://127.0.0.1:5000, bound to loopback by default. Exposing it on a network is a decision you make explicitly.
Base path/v1
Health/health
AuthenticationA bearer token generated on, and held on, the machine running the runtime. No third-party provider API key is required, because there is no third-party provider in the path.

A request

curl
curl http://127.0.0.1:5000/v1/chat/completions \
  -H "authorization: Bearer $AISL_TOKEN" \
  -H "content-type: application/json" \
  -d '{"model":"<model>","messages":[{"role":"user","content":"Hello"}]}'

The token above is read from your environment. Never paste a real token into a terminal you are sharing, a screenshot, or an issue report.

Streaming

Streamed responses are supported. Delivery is currently buffered rather than token-native, which is worth knowing if you are building an interface that shows text arriving.

Last updated 2026-09-22T10:05:46-04:00 · Request access

These pages describe the contract: ports, endpoints, volumes, the permission model and the behaviour you can expect. The command reference for the CLI is published with the build each tester receives, so that it matches the binary in their hands rather than a page written ahead of it.

Available through a controlled private beta. Larger model profiles are added as they complete hardware validation.