Tencent/CognitiveKernel-Pro
A deep research agent framework with open-source training recipes for building and fine-tuning autonomous research agents.
Not currently ranked — collecting fresh signals.
star history
CognitiveKernel-Pro is a state-of-the-art open-source deep research agent that uses free tools (with optional Google Search API). The project provides fully reproducible supervised fine-tuning recipes for training agent foundation models without requiring reinforcement learning, along with RL training recipes (subgoal-GRPO) for further improvements. It publishes datasets on Hugging Face for both query data and SFT training, as well as trained model weights (Qwen3-8B-CK-Pro). The associated research has been accepted at ACL 2026.
Frequently asked
- What is Tencent/CognitiveKernel-Pro?
- A deep research agent framework with open-source training recipes for building and fine-tuning autonomous research agents.
- Is CognitiveKernel-Pro open source?
- Yes — Tencent/CognitiveKernel-Pro is an open-source project tracked on heatdrop.
- What language is CognitiveKernel-Pro written in?
- Tencent/CognitiveKernel-Pro is primarily written in Python.
- How popular is CognitiveKernel-Pro?
- Tencent/CognitiveKernel-Pro has 524 stars on GitHub.
- Where can I find CognitiveKernel-Pro?
- Tencent/CognitiveKernel-Pro is on GitHub at https://github.com/Tencent/CognitiveKernel-Pro.