UplinkSwitch

    • Roo Code

    OpenAI Compatible provider, base URL, key, one model id.

    Checked 2026-10-12

    ·All guides·Models

    Where the settings are

    The Roo Code settings panel (the gear icon) → API Provider. Roo Code only takes an OpenAI-compatible base URL here, so the Anthropic-style variables from the Claude Code guide do not apply.

    Steps

    1. API Provider: select OpenAI Compatible.
    2. Base URL: https://api.uplinkswitch.com/v1.
    3. API Key: your UplinkSwitch key from the console.
    4. Model: enter one id. Any id on the Models page works:
      model ids
      claude-sonnet-5-5
claude-opus-5-5
gpt-6.1-sol
gemini-3.1-pro-preview
deepseek-v4-pro-0813
kimi-k3
    5. Model Configuration (optional): Max Output Tokens, Context Window, Image Support, Computer Use, Input Price, Output Price. For the prices, use the per-million figures on the Models page.

    Check it

    From a shell, the same key should list the models:

    shell
    curl https://api.uplinkswitch.com/v1/models -H "Authorization: Bearer sk-uplink-live-..."

    Then start a task in Roo Code. The request shows in the console within seconds.

    Good to know

    • Roo Code uses native tool calling only. Tools go out in OpenAI's tools schema and calls stream back as tool events; there is no XML fallback. A model that cannot return tool calls cannot be used in Roo Code.
    • Tool arguments stream incrementally, so Roo Code starts a tool before the model finishes writing the call.
    • The Model field takes one id at a time. To switch models, edit the field.
    • Roo Code's OpenAI Compatible settings have no caching switch. Cached input bills at the cache read price on the Models page.

    Related