Kani: A Lightweight and Highly Hackable Framework for Building Language Model Applications
This provides a more hackable tool for researchers, hobbyists, and industry professionals building complex language model applications, though it is incremental as it improves upon existing frameworks rather than introducing a new paradigm.
The paper tackles the problem of existing frameworks for language model applications being too opinionated and limiting customizability, by presenting Kani, a lightweight and flexible open-source framework that supports core building blocks like model interfacing and function calling, enabling developers to accelerate development while maintaining control.
Language model applications are becoming increasingly popular and complex, often including features like tool usage and retrieval augmentation. However, existing frameworks for such applications are often opinionated, deciding for developers how their prompts ought to be formatted and imposing limitations on customizability and reproducibility. To solve this we present Kani: a lightweight, flexible, and model-agnostic open-source framework for building language model applications. Kani helps developers implement a variety of complex features by supporting the core building blocks of chat interaction: model interfacing, chat management, and robust function calling. All Kani core functions are easily overridable and well documented to empower developers to customize functionality for their own needs. Kani thus serves as a useful tool for researchers, hobbyists, and industry professionals alike to accelerate their development while retaining interoperability and fine-grained control.