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

Integrations DeepSeek Harness

Use DeepSeek in DeepSeek Harness, hosted in Australia

Add the Xinference Model API to DeepSeek Harness as a custom model API 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

    Open DeepSeek Harness

    Start the web UI, then open Settings and Models.

    Terminal
    $ npx @deepseek-ai/dsh web
  3. 3

    Add a custom model API

    Choose Add model provider and switch the card to Custom model API. Fill in these values and save.

    Custom model API
    Provider ID   xinference
    Base URL      https://cloud.xinference.co/v1
    API protocol  OpenAI Chat Completions
    API key       your Xinference API key
    Model         deepseek-v4.1-flash
  4. 4

    Pick the model

    Choose Fetch available models, or add deepseek-v4.1-flash and glm-5.3 by hand. Then select the model in the model picker.

Run DeepSeek Harness on DeepSeek, hosted in Australia