locallama-mcp

Heratiki/locallama-mcp
★ 41 stars TypeScript 💻 Code/Dev Tools Updated 1d ago
An MCP Server that works with Roo Code/Cline.Bot/Claude Desktop to optimize costs by intelligently routing coding tasks between local LLMs free APIs and paid APIs.
View on GitHub →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "locallama-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "Heratiki/locallama-mcp"
      ]
    }
  }
}

README Excerpt

LocalLama MCP is a local-first, provider-neutral Model Context Protocol server for modern coding-agent workflows. It is being revived to support current MCP-capable tools such as Codex, Claude Code, Claw Code, Cursor, GitHub Copilot Agent mode, and generic MCP clients. The project routes coding work across local models, free or low-cost remote models, and paid frontier models using cost, latency, context capacity, benchmark history, and task fit. It is no longer designed around Cline or Roo Code

Tools (20)

CODE_SEARCH_ENABLEDCODE_SEARCH_EXCLUDE_PATTERNSCODE_SEARCH_INDEX_ON_STARTCODE_SEARCH_REINDEX_INTERVALCOST_THRESHOLDDEFAULT_LOCAL_MODELDEPENDENCY_ANALYSIS_ENABLEDLM_STUDIO_ENDPOINTLOCK_FILE_CHECK_ACTIVE_PROCESSMAX_SUBTASKSOLLAMA_ENDPOINTOPENROUTER_API_KEYPYTHON_DETECT_VENVPYTHON_PATHPYTHON_VENV_PATHQUALITY_THRESHOLDREMOVE_STALE_LOCK_FILESSUBTASK_GRANULARITYTASK_DECOMPOSITION_ENABLEDTOKEN_THRESHOLD

Topics

clinebotmcp-serverroocodevscode