OpenAI Compatible Endpoint for LM Studio

Use any OpenAI-compatible API in LM Studio via OpenRouter.ai.

Features

• Model Dropdown - Select from cached free models (24+ available)
• Auto Mode - Automatically uses the first available free model
• Custom Model ID - Enter any model manually (e.g., openai/gpt-4o)
• Request Tracking - Shows request count (25 free requests max)
• Background Caching - Model list refreshes automatically every hour

Installation

Or install locally without pushing to Hub:

Configuration

Global Settings (apply to all chats)

• API Key (optional for free models)

  • Get your key from OpenRouter.ai
  • Enter as sk-or-v1-xxxxx

• Base URL (default: https://openrouter.ai/api/v1)

Per-Chat Settings

• Model - Dropdown to select a model

Usage

• Open LM Studio and activate this plugin
• On first run, the model cache is empty — the plugin will load models on first generation
• Or manually refresh by restarting the plugin after initial use
• Select a model or use Auto mode
• Start chatting

Rate Limits

• Free tier: ~25 requests per day (varies by model)
• Free models: Look for the :free suffix (e.g., x-ai/grok-4.1-fast:free)
• Some models may return 429 after a few requests

Available Free Models

The plugin caches free models on startup. Common ones include:

• x-ai/grok-4.1-fast:free
• stepfun/step-3.5-flash:free
• arcee-ai/trinity-mini:free

Development Commands

Files

• src/ - TypeScript source
• dist/ - Compiled JavaScript
• models-cache.json - Cached model list (generated at runtime)