This project provides an open-source framework for integrating agentic tools into Large Language Model (LLM) reasoning, aiming to enhance deep research capabilities. It is targeted at researchers and developers exploring advanced LLM applications.

How It Works

The framework integrates agentic tools directly into the LLM's reasoning process. This approach allows LLMs to leverage external resources and functionalities, such as search engines (You.com, Bing) and potentially other APIs (Jina), to perform deeper research and more complex tasks.

Quick Start & Requirements

• Install via conda env create -f environment.yml.
• Requires API keys for OpenAI, You.com, and optionally Jina and Bing.
• Environment variables OPENAI_API_KEY and YDC_API_KEY must be set.
• Example run command: python scripts/run_agentic_reason.py --use_jina True --jina_api_key "your jina api key" --bing_subscription_key "your bing api key" --remote_model "your remote model name" --deep_research True

Highlighted Details

• Enables LLM reasoning with agentic tools for deep research.
• Supports integration with You.com, Jina, and Bing search APIs.
• Allows specifying remote LLM models (e.g., gpt-4o).
• Includes an optional mind mapping feature.

Maintenance & Community

The project is marked as "Still in development" with a TODO list indicating ongoing work on auto-research and cleaning agentic reasoning. No specific community links or contributor details are provided in the README.

Licensing & Compatibility

The README does not explicitly state a license. The project is described as "free and open," but specific licensing terms for commercial use or closed-source linking are not detailed.

Limitations & Caveats

The project is explicitly stated to be "Still in development" and "undergoing rapid updates," suggesting potential instability and breaking changes. Key functionalities like "auto research" are still pending.