Now live: DeepSeek V4.1 Flash and GLM-5.3, hosted in Australia. Get started →

Integrations OpenClaw

Use DeepSeek in OpenClaw, hosted in Australia

Add the Xinference Model API to OpenClaw as a custom provider and run DeepSeek V4.1 Flash or GLM-5.3 in Australian data centres.

Hosted in Australia

Set it up in four steps

  1. 1

    Get your key

    Sign up for the Model API and create an API key in the console.

  2. 2

    Run onboarding

    Run this and pick Custom provider. Paste your API key when asked.

    Terminal
    $ openclaw onboard
  3. 3

    Enter your details

    Use these values. OpenClaw names models as provider and model ID, so keep the provider ID short.

    OpenClaw custom provider
    Base URL     https://cloud.xinference.co/v1
    Model ID     deepseek-v4.1-flash
    Provider ID  xinference
  4. 4

    Set the default model

    Make DeepSeek the default. To switch in a chat, type /model xinference/glm-5.3.

    Terminal
    $ openclaw models set xinference/deepseek-v4.1-flash

Prefer a config file?

Add the provider to ~/.openclaw/openclaw.json.

~/.openclaw/openclaw.json
{
  agents: { defaults: { model: { primary: "xinference/deepseek-v4.1-flash" } } },
  models: {
    mode: "merge",
    providers: {
      xinference: {
        baseUrl: "https://cloud.xinference.co/v1",
        apiKey: "${XINFERENCE_API_KEY}",
        api: "openai-completions",
        models: [
          { id: "deepseek-v4.1-flash", name: "DeepSeek V4.1 Flash" },
          { id: "glm-5.3", name: "GLM-5.3" },
        ],
      },
    },
  },
}

Run OpenClaw on DeepSeek, hosted in Australia