An open-source framework designed to build intelligent search and research agents with capabilities for deep web search, content extraction, and comprehensive answer generation.
One-Minute Overview#
II-Researcher is a powerful deep search agent framework that uses BAML functions to perform intelligent web searches and generate comprehensive answers to questions. It's designed for developers and researchers who need advanced research capabilities, particularly for scenarios requiring deep information retrieval and analysis.
Core Value: Provides deeper, more accurate answer generation through multi-step reasoning and reflection mechanisms compared to traditional search approaches
Quick Start#
Installation Difficulty: Medium - Requires configuring multiple API keys and optional LiteLLM server setup
pip install ii-researcher
Core Capabilities#
- Intelligent Web Search - Supports multiple search providers (Tavily and SerpAPI), implements advanced query strategies and information source selection
- Multi-Step Reasoning and Reflection - Supports multiple rounds of thinking and self-correction, optimizes search strategies and information processing workflows
- Configurable Pipeline System - Supports custom reasoning methods, configurable LLM model selection
- Comprehensive Answer Generation - Automatically generates structured reports, includes reference sources and traceable information links
- Asynchronous Operation - Parallel processing of search and scraping tasks, optimized resource usage and response speed
Tech Stack#
Python, JavaScript, TypeScript, FastAPI, Next.js, LiteLLM, BAML, Tavily/SerpAPI, Firecrawl