JetBrains AI Assistant with Meridian

Route JetBrains AI Assistant through Meridian for unified billing, rate limiting, and model access — no plugin required.

  1. 1

    Open AI Assistant settings

    In any JetBrains IDE, open Settings → Tools → AI Assistant → AI Providers. Click the + button and choose Custom OpenAI Endpoint.
  2. 2

    Enter the Meridian endpoint

    Set the URL field to:https://meridian.getnimbus.net/api/v1
  3. 3

    Authenticate with your API key

    Paste your Meridian API key into the API Key field:nim_live_••••••••••••••••••••••••••

    Keys are available in your Meridian dashboard. Use a nim_live_ key for production.

  4. 4

    Select a default model

    In the Default Model dropdown, enter one of:
    • gpt-5-codex — best for code generation and refactoring
    • azure-swc/gpt-4.1 — Azure-deployed GPT-4.1, low latency

    Any model in your Meridian workspace is available. JetBrains sends standard OpenAI Chat Completions requests — fully compatible.

  5. 5

    Verify connectivity

    Click Test Connection. A green checkmark confirms Meridian is reachable. Start a chat — completions stream through Meridian with your workspace policies applied.

Compatibility

JetBrains AI Assistant uses the OpenAI Chat Completions API shape (/v1/chat/completions) with streaming. Meridian proxies this transparently — no translation layer, no feature loss. Works with IntelliJ IDEA, PyCharm, WebStorm, GoLand, CLion, Rider, and all other JetBrains IDEs (2024.2+).