Console

Home / Agents / OpenClaw

OpenClaw

Your self-hosted, multi-channel AI assistant

OpenClaw Foundation OpenAI-compatible macOSLinuxWindows
Official install page All tools

OpenClaw is an open-source AI assistant whose Gateway runs on your own hardware, reachable through Discord, Slack, Telegram, WhatsApp, and 20+ other channels, with swappable model backends. It suits teams that want to self-host and expose an agent on chat platforms. It supports any OpenAI-compatible endpoint via models.providers.

Base URL (root)https://quantoport.comPut this in each tool's base_url or *_BASE_URL setting.
OpenAI-compatible pathhttps://quantoport.com/v1Append /v1 for anything speaking the OpenAI wire (/v1/chat/completions).
Recommended modelsclaude-sonnet-5deepseek-v4-proglm-5.3Names published in our own catalogue — confirm the live list for your key with GET /v1/models.

Skip it — one-line setup

The script installs OpenClaw from its official source, writes the gateway settings into the tool's own config file (backing up any existing one) and runs a live check. It asks for your gateway key, or take it from --key / --key-file. --uninstall restores exactly what it wrote.

shell
curl -fsSL https://quantoport.com/agents/openclaw/setup | bash

1 · Install

Install OpenClaw from its own official source.

install
curl -fsSL https://openclaw.ai/install.sh | bash
# or npm (npm 11.16+):
npm install -g openclaw@latest --allow-scripts=openclaw
# or run without installing via npx:
npx openclaw@latest

2 · Configure — point it at QuantoPort

Gateway base URL https://quantoport.com; get your key from the console.

config
International site https://quantoport.com (OpenAI format, supported directly):
Edit ~/.openclaw/openclaw.json:
{
  "agents": { "defaults": { "model": { "primary": "quantoport/gpt-5.1-codex-max" } } },
  "models": {
    "providers": {
      "quantoport": {
        "baseUrl": "https://quantoport.com/v1",
        "apiKey": "${QUANTOPORT_API_KEY}",
        "api": "openai-completions",
        "models": [ { "id": "gpt-5.1-codex-max", "name": "GPT-5.1 Codex Max" } ]
      }
    }
  }
}
Then restart the Gateway: openclaw gateway restart.
CN site https://api.shinlinkai.com (OpenAI format, supported directly): change baseUrl to https://api.shinlinkai.com/v1.

3 · Verify

One request proves you are on our channel.

verify
openclaw models list
# seeing your quantoport models means the config is live; openclaw gateway status confirms the Gateway is running, then send a prompt to verify.

4 · Common pitfalls

  • The provider's api must be "openai-completions"; for custom non-api.openai.com endpoints OpenClaw automatically disables unsupported developer-role compat.
  • Register custom models in models.providers.<id>.models[] with an id, or the agent won't see them.
  • apiKey accepts ${ENV_VAR} to inject from the environment instead of plaintext; model refs are provider/model, e.g. quantoport/<model-id>.
  • Use the model names your gateway actually serves.

Sources for the tool-side steps: github.com/openclaw/openclaw raw.githubusercontent.com/openclaw/openclaw/main/docs/concep raw.githubusercontent.com/openclaw/openclaw/main/docs/instal