Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News OpenAI Announces Function Calling, Allowing Developers to Describe Functions

OpenAI Announces Function Calling, Allowing Developers to Describe Functions

This item in japanese

OpenAI has introduced updates to the API, including a capability called function calling, which allows developers to describe functions to GPT-4 and GPT-3.5 and have the models create code to execute those functions.

According to OpenAI, function calling facilitates the development of chatbots capable of leveraging external tools, transforming natural language into database queries, and extracting structured data from text. These models have undergone fine-tuning to not only identify instances where a function should be invoked but also provide JSON responses that align with the function signature.

AI models can intelligently interface with external tools and APIs thanks to the crucial role played by function calling. Developers can access a large selection of functionality and services by specifying functions to these models. By using external tools to respond to queries, search databases, or extract structured data from unstructured text, this connection enables AI models to accomplish tasks that are beyond their natural capacities. As a result of function calling, AI models become more versatile and effective instruments for tackling complex challenges in the real world.

With the introduction of gpt-4-0613 and gpt-3.5-turbo-0613, developers now have the ability to describe functions to these models. As a result, the models can intelligently generate JSON objects that contain the necessary arguments to call those functions. This exciting development offers a more dependable means of connecting GPT's capabilities with external tools and APIs, opening up new possibilities for seamless integration.

These models have developed the capability to recognize situations where a function should be activated based on the user's input through careful fine-tuning. Additionally, they have been taught to provide JSON answers that match the particular function signature. Developers can now more reliably and consistently get structured data from the model by using function calling.

In addition to function calling, OpenAI is introducing an enhanced variant of GPT-3.5-turbo that offers a significantly expanded context window. The context window, measured in tokens or units of raw text, represents the amount of text considered by the model prior to generating further text. This expansion allows the model to access and incorporate a larger body of information, enabling it to make more informed and contextually relevant responses.

Function calls in AI development allow models to utilize tools designed by developers, enabling them to extend their capabilities and integrate customized functionalities. This collaborative approach bridges the gap between AI models and developer-designed tools, fostering versatility, adaptability, and innovation in AI systems.

About the Author

Rate this Article