OpenAI announces function calling and other API updates: What does it mean?

The new upgrades to API include – a new function calling capability in the Chat Completions API, updated and more steerable versions of GPT-4 and GPT 3.5 turbo, new 16K context window of GPT-3.f turbo, 75 per cent cost reduction in state-of-the-art embeddings model, 25 per cent reduction on input tokens for GPT-3.5 turbo, deprecation timeline for the gpt-3.5-turbo-0301 and gpt-4-0314 models.

In its blog, the company said that all these models come with the same data privacy and security guarantees that were introduced in March according to which the consumer owns all outputs generated from their requests, and their API data will not be used for training.

What are function calls?

This is believed to be the next-level integration for developers. With GPT-4 and GPT-3.5-turbo, developers can return JSON objects for function calls, leading to limitless possibilities when it comes to interacting with APIs and tools.

“Developers can now describe functions to gpt-4-0613 and gpt-3.5-turbo-0613 and have the model intelligently choose to output a JSON object containing arguments to call those functions. This is a new way to more reliably connect GPT’s capabilities with external tools and APIs,” read the post on OpenAI’s official blog.

JSON here stands for JavaScript Object Notation, an encoding system developed to overcome the need for any ad-hoc code for applications to communicate with servers that communicate in a predefined way. According to OpenAI, Function calling allows developers to get structured data back from the model. In simple words, function calls will allow the model to use tools developers design for it.