Skip to content
LUNAROUTEDocs

LunaRoute documentation

Connect your agents to fixed-cost inference, with private handling of prompts and responses. Start with an API key, then use the client and API format you already know.

Get startedExplore the API
  your coding agent
         |
         v
    [ LunaRoute ]
         |
         +-- priority requests
         +-- background work
         |
         v
  curated open-weight models

Make your first request

Section titled “Make your first request”

Create an API key in the dashboard, then send a request to the LunaRoute endpoint:

Terminal window
curl https://gw.lunaroute.com/v1/chat/completions \
  -H "LUNAROUTE-API-KEY: $LUNAROUTE_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "glm-5.3",
    "messages": [{"role": "user", "content": "Say hello."}]
  }'

The first request guide covers streaming, Anthropic Messages, Responses, and background requests.

Start here

Section titled “Start here”
Connect a coding agentConfigure your client with the LunaRoute CLI.Understand your concurrencyPriority scheduling, automatic overflow, and queues.Use tools over MCPConnect search, document conversion, and image tools.Set up a personal agentConnect OpenClaw, Hermes, NanoClaw, or Mira.

What you get

Section titled “What you get”

Subscriptions run on shared infrastructure, with a fixed price for the plan you choose. Your plan determines how many requests can run at once. Read usage and billing and private handling of prompts and responses.

We support a curated set of open-weight models. The selection evolves as we find other models we can support well. Use the model catalog to find current model IDs.

Compare subscription plans, or explore Enterprise for dedicated infrastructure.

LunaRoute documentation