Integration

Open Interpreter with Meridian

Point Open Interpreter at Meridian's OpenAI-compatible API and run code-generating models through your existing Meridian key. No separate billing, no new accounts.

Quick start

interpreter \
  --api_key nim_live_*** \
  --api_base https://meridian.getnimbus.net/api/v1 \
  --model gpt-5-codex

Replace nim_live_*** with your live key from the API keys page. The base URL must end in /api/v1 — Open Interpreter appends /chat/completions automatically.

Available models

  • gpt-5-codex— primary code-generation model
  • gpt-5-mini— faster, lower cost
  • claude-4-opus— alternative reasoning profile

Environment variable

Set these once and omit the flags on every invocation:

export OPENAI_API_KEY=nim_live_***
export OPENAI_API_BASE=https://meridian.getnimbus.net/api/v1

Rate limits

Requests count against your Meridian plan quota. Check usage for current consumption. The API returns standard 429 responses when you exceed your tier — Open Interpreter will retry automatically.