Verified project record

InternLM/MindSearch

An LLM-based multi-agent framework that performs concurrent web searches to produce synthesized answers. It requires configuring both an external search API and an LLM server, and involves running separate backend and frontend components.

Project overview

Project type: AI Agent · Workflow · AI Search
Use cases: Search & Research
Deployment: Refer to project documentation
License: Apache-2.0

Best for

Developers and researchers who need a multi-agent framework to perform simultaneous multi-query web searches.
Users who prefer interaction via a web GUI, CLI, or API and are comfortable configuring external LLM and search services.

Key capabilities

Deploys asynchronous agents to perform simultaneous multi-query web searches.
Provides a backend API server to handle search requests.

Limitations and risks

The system requires an LLM server such as InternLM2.5-7b-chat or the GPT-4 API, which may incur costs depending on the chosen provider.
You must provide a WEB_SEARCH_API_KEY for the selected search engine (Bing, Brave, Google Serper, etc.) or Tencent Search credentials.
GPU and minimum hardware specifications are not documented.
Telemetry data collection practices are not documented in the available information.

Getting started

The setup difficulty is medium. You need to clone the repository, install dependencies, and set up environment variables in a .env file.
After configuration, start the backend FastAPI server, then start the frontend component (React, Gradio, or Streamlit).

Evidence and sources

GitHub project description: 🔍 An LLM-based Multi-agent Framework of Web Search Engine (like Perplexity.ai Pro and SearchGPT)
README: git clone https://github.com/InternLM/MindSearch cd MindSearch pip install -r requirements.txt
README: title={MindSearch: Mimicking Human Minds Elicits Deep AI Searcher}
README: Improved the UI to embody the simultaneous multi-query search.
README: MindSearch is now deployed on Puyu! 👉 [Try it](https://internlm-chat.intern-ai.org.cn/) 👈

Reviewed at: 2026-07-18T06:55:38Z