Back to Docs

Windsurf with Meridian

Connect Meridian as a custom provider in Windsurf for AI-assisted development powered by our models.

Prerequisites

  • Windsurf IDE installed (latest version recommended)
  • A Meridian API key — generate one here
  • Active Meridian subscription with API access

Step 1 — Configure Provider

Open Windsurf settings and add Meridian as a custom OpenAI-compatible provider. Navigate to Settings → AI Providers → Add Custom Provider.

{
  "providers": {
    "meridian": {
      "baseUrl": "https://api.getnimbus.net/v1/chat/completions",
      "apiKey": "YOUR_MERIDIAN_API_KEY",
      "models": ["meridian-1"]
    }
  }
}

Step 2 — Set Environment Variables

Alternatively, configure Meridian via environment variables in your Windsurf workspace or system environment.

MERIDIAN_API_KEY=sk-mrd-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
MERIDIAN_BASE_URL=https://api.getnimbus.net/v1/chat/completions

Step 3 — Select Model

After saving the provider configuration, open the Windsurf chat panel and select meridian-1 from the model dropdown. Meridian is now ready for inline completions, chat, and agent mode.

Rate Limits

Free tier

100 req/day

Pro tier

5,000 req/day

Max tokens/req

8,192

Concurrent

4 streams

Troubleshooting

401 Unauthorized

Verify your API key is active and has not expired.

Model not appearing

Ensure the provider base URL ends with /v1 and the model name matches exactly.

429 Rate Limited

Upgrade to Pro or wait for the rate window to reset (UTC midnight).