GitHub - topoteretes/cognee: Deterministic LLMs Outputs for AI Applications and AI Agents
Apr 22, 2024 - github.com
Cognee is an open-source framework designed for AI engineers to create self-improving deterministic outputs for large language models (LLMs) using graphs, LLMs, and vector retrieval. It allows users to add new information and use LLMs and cognee to create knowledge. Users can query cognee for the knowledge and add alternative data types. It can be installed using pip or poetry and can be used with Ollama or Anyscale as the LLM provider.
The framework also offers a demo notebook for users to understand its functionality better. It can be used on Whatsapp with one of their partners by typing specific commands. The framework's working is illustrated with an architecture diagram. For more detailed information, users can refer to the documentation.
Key takeaways:
Cognee is an open-source framework for creating self-improving deterministic outputs for LLMs, primarily used by AI Engineers.
It allows users to add new pieces of information and use LLMs and cognee to create knowledge. Users can query cognee for the knowledge.
It supports different data types and allows targeting specific directories to add by providing dataset_name.
Users can try Cognee on Whatsapp with one of their partners by typing specific commands to save and query knowledge.
