Skip to content
AI-Point.uz
← Back to catalog
MCP serverCommunity

Ollama / vLLM Bridge

Call your local Ollama or vLLM model over MCP with schema-validated JSON output

setheerwagen · ↓ 0 installs

Install

Add the config to your client settings. Replace values in <angle brackets> with your own.

Config file: claude_desktop_config.json
{
  "mcpServers": {
    "mcp-ollama-vllm": {
      "command": "uvx",
      "args": [
        "mcp-ollama-vllm==1.0.6"
      ],
      "env": {
        "LOCAL_BACKEND": "ollama",
        "LOCAL_HOST": "<LOCAL_HOST>",
        "LOCAL_API_KEY": "<LOCAL_API_KEY>",
        "LOCAL_TIMEOUT": "600",
        "LOCAL_EMBED_MODEL": "nomic-embed-text",
        "LOCAL_EMBED_PATH": "/v1/embeddings",
        "LOCAL_SCHEMA_MODE": "auto"
      }
    }
  }
}

Tools

The tool list will appear once the server has a sandbox or the editors fill it in.

Report a problem ↗ (Opens GitHub in a new tab)