HQLHLVM DocsHQL DocsGitHub
Download
HLVM for macOSmacOS 26+ · .dmg

Runtime Server

hlvm serve exposes the provider-neutral HLVM runtime host for GUI clients and tools.

hlvm serve

Default host:

http://127.0.0.1:11435

Core Endpoints

MethodPathPurpose
GET/healthRuntime health check
POST/api/chatChat, eval, and agent submissions
GET/api/chat/messagesCurrent conversation messages
GET/api/chat/streamServer-sent conversation events

Evaluate HQL

curl -X POST http://127.0.0.1:11435/api/chat \
  -H "Content-Type: application/json" \
  -d '{"mode":"eval","messages":[{"role":"user","content":"(+ 1 2)"}]}'

Chat

curl -X POST http://127.0.0.1:11435/api/chat \
  -H "Content-Type: application/json" \
  -d '{"mode":"chat","messages":[{"role":"user","content":"hello"}]}'

Engine Boundary

The runtime host is not itself a model engine. It can route to native harnesses, cloud providers, or local models. It never installs or starts Ollama. When an Ollama route is selected, the user-owned service at its standard endpoint must already be available.

Next