local runtime · paired

OpenAI-compatible local runtime

Local text inference through the user’s companion when the runtime follows the supported OpenAI-compatible response contract.

Curious Obmil studies an observatory instrument beside a smaller happy GoldCap with one main cap, two small caps, and living gold veins
Curious checks the signal. GoldCap keeps the local path close to the user.
DecaCap interface
User-configured loopback chat-completions endpoint
Published evidence
The companion supports a user-configured local endpoint and token. Runtime-specific behavior must be tested with the hardware checker and companion health test.
Exact-model probe
Not published Account-, model-, region-, quota-, runtime-, and hardware-specific verification still requires a real user-owned key or device.
Last reviewed
2026-07-26
REQUIREMENTS
  • A local server bound to loopback
  • A compatible chat-completions response
  • User-configured endpoint and optional token
  • A locally available model
LIMITATIONS
  • Compatibility is contract-based, not a promise that every OpenAI-like server behaves identically
  • Remote LAN or internet endpoints are outside the intended local-only boundary
IMPLEMENTATION TEST TARGETS
  • Loopback endpoint configuration
  • Optional local token
  • Companion-authenticated browser path
  • Text response extraction
WHAT THIS DOES NOT PROVE
  • Access to a particular model on a user account
  • Quality, speed, pricing, quota, or safety behavior
  • Hardware fit or training-toolkit compatibility