Add Web Search MCPs to LM Studio (Tavily + Brave) - OpenAI-Compatible Responses API
Make your local LLM actually useful with real time information from the Internet! Add web search MCPs and call them from both the chat UI and LM Studio’s new responses API (OpenAI-compatible). This walkthrough covers Tavily + Brave setup, model selection (Qwen-3 8B example), and running a Python client against your local server.
What I cover:
Version & preflight checks (I demo on 0.3.30).
MCP JSON: where to paste it and how to secure your API key.
Live demo: model calls tool → shows search query → returns live answer.
API example: Python + curl for streaming & reasoning control.
Pro tip:…
Watch on YouTube ↗
(saves to browser)
DeepCamp AI