Lm
Lm is an MCP server that connects Claude to any OpenAI-compatible LLM endpoint and offload routine work to a local model. It has 120 GitHub stars, is released under the Apache-2.0 license and runs locally with npx -y @houtini/lm. It works with Claude Desktop, Claude Code, Cursor and VS Code and needs an API key.
- Needs API key
- Developer tools
- Apache-2.0
- Actively maintained
Install Lm
Generated from the server's MCP registry entry. Replace your-value with your own values.
Claude Desktop
{
"mcpServers": {
"lm": {
"command": "npx",
"args": [
"-y",
"@houtini/lm"
],
"env": {
"HOUTINI_LM_API_KEY": "your-value"
}
}
}
}Settings > Developer > Edit Config. macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\. Restart Claude Desktop afterwards.
Claude Code
claude mcp add --env HOUTINI_LM_API_KEY=your-value --transport stdio lm -- npx -y @houtini/lmCursor
{
"mcpServers": {
"lm": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@houtini/lm"
],
"env": {
"HOUTINI_LM_API_KEY": "your-value"
}
}
}
}Project file; use ~/.cursor/mcp.json to enable it in every project.
VS Code
{
"inputs": [
{
"type": "promptString",
"id": "houtini_lm_api_key",
"description": "HOUTINI_LM_API_KEY",
"password": true
}
],
"servers": {
"lm": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@houtini/lm"
],
"env": {
"HOUTINI_LM_API_KEY": "${input:houtini_lm_api_key}"
}
}
}
}Config formats checked against the official docs on Oct 7, 2026: modelcontextprotocol.io (opens in a new tab), code.claude.com (opens in a new tab), cursor.com (opens in a new tab), code.visualstudio.com (opens in a new tab).
Environment variables
Variables the server reads at startup.
| Name | Required | Description |
|---|---|---|
HOUTINI_LM_ENDPOINT_URL | No | Base URL of the OpenAI-compatible API endpoint (e.g. http://localhost:1234, https://openrouter.ai/api). Legacy alias: LMSTUDIOURL. |
HOUTINI_LM_API_KEYsecret | No | Bearer token for API authentication (no auth if blank). Legacy aliases: LMSTUDIOPASSWORD, LMPASSWORD, OPENROUTERAPI_KEY. |
HOUTINI_LM_MODEL | No | Model identifier to use for requests (auto-detected if not set). Legacy alias: LMSTUDIOMODEL. |
HOUTINI_LM_PROVIDER | No | Optional provider override: 'openrouter' forces OpenRouter-specific handling (attribution headers, reasoning.exclude, no inference serialisation)… |
About Lm
Houtini LM is an MCP server that lets Claude (or any MCP client) hand bounded work to another model - a local LLM on your GPU, OpenAI's latest GPT models, a LiteLLM router, OpenRouter or a cheap cloud API - while you carry on working in the AI platform you already like. It cuts your token bill, and it gives you a second model to review your code whenever you want one.
- ai-agents
- claude-mcp
- code-generation
- developer-tool
- developer-tools
- lm-studio
- local-llm
- mcp
- mcp-client
- mcp-server
Similar MCP servers
More developer tools MCP servers
Gemini CLI
google-gemini/gemini-cli
An open-source AI agent that brings the power of Gemini directly into your terminal.
Developer toolsTypeScriptFront-End Checklist
thedaviddias/Front-End-Checklist
Review frontend code and live pages against 386 quality-gated web development rules.
Developer toolsMDXClaude Flow
ruvnet/ruflo
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Developer toolsTypeScriptCodebase Memory
DeusData/codebase-memory-mcp
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Developer toolsCBytedance Filesystem
bytedance/UI-TARS-desktop
MCP server for filesystem access.
Developer toolsTypeScriptGitHub
github/github-mcp-server
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
OfficialDeveloper toolsGo
What is Lm?
Lm is an MCP server that connects Claude to any OpenAI-compatible LLM endpoint and offload routine work to a local model. It has 120 GitHub stars, is released under the Apache-2.0 license and runs locally with npx -y @houtini/lm. It works with Claude Desktop, Claude Code, Cursor and VS Code and needs an API key. The source code is at github.com/houtini-ai/houtini-lm.
How do I install the Lm MCP server?
Add the command npx -y @houtini/lm to your MCP client: put it in claude_desktop_config.json for Claude Desktop, run claude mcp add for Claude Code, or add it to .cursor/mcp.json (Cursor) or .vscode/mcp.json (VS Code). The snippets on this page are ready to paste.
Is Lm free?
The server is open source under the Apache-2.0 license, so running it is free. It needs credentials (HOUTINI_LM_API_KEY) for the service it connects to, which may require a paid account.
Is Lm actively maintained?
The most recent commit was on Oct 6, 2026. The latest release is v2.8.0, published Mar 18, 2026. appsgit only lists MCP servers with a commit in the last six months and re-checks every server daily.