OpenAI-Compatible Endpoint Adapter for LM Studio

Use any OpenAI-compatible API with LM Studio by routing chat requests through a configurable base URL (default: OpenRouter). The plugin streams responses, forwards tool calls, and lets you choose any model ID supported by your provider.

Features

• Works with OpenAI-compatible providers (OpenRouter by default).
• Streaming chat responses.
• Tool call passthrough when the provider supports tools.
• Simple per-chat model selection.
• Optional debug logging for request/response diagnostics.

Requirements

• Node.js + npm
• LM Studio installed, with the lms CLI available (LM Studio must be run at least once before lms works).

Optional (Windows): bootstrap lms into PATH if needed:

Quick Start (Local Development)

npm run dev runs lms dev, which starts the plugin dev server, verifies manifest.json, installs deps if needed, and rebuilds on changes. Note: lms dev is part of LM Studio Plugins (beta).

Configuration

Configuration is done in LM Studio plugin settings.

Global (shared across chats):

• API Key: provider API key (e.g., sk-or-... for OpenRouter)
• Base URL: provider base URL (default: https://openrouter.ai/api/v1)

Per-chat:

• Model: any provider model ID, e.g. allenai/molmo-2-8b:free
• Debug Logging: enable verbose request logging and status output

Usage

• Set API Key and Base URL in global settings.
• Set the Model in the chat/plugin settings.
• Start a chat in LM Studio.

If the model is missing, the plugin will return an error asking you to set it.

Troubleshooting

• 400 Provider returned error: the model ID may be invalid or the provider rejected the request. Enable Debug Logging to inspect the payload.
• Invalid or missing API key: verify the key in global settings.
• 429 Rate limit: try again later or use a paid key.

Development Notes

• Source code lives in src/.
• Build output goes to dist/ (generated by TypeScript).
• Key files: src/generator.ts, src/config.ts, manifest.json.

Publishing

npm run push wraps lms push to upload the current folder as a plugin revision. lms push prompts for confirmation unless you pass -y.

If you have not authenticated with LM Studio Hub yet, run:

Security

Never hardcode API keys. Use the protected global config field instead.

License

ISC