Perform calling and different API updates
Builders can now describe capabilities to gpt-4-0613 and gpt-3.5-turbo-0613, and have the mannequin intelligently select to output a JSON object containing arguments to name these capabilities. This can be a new technique to extra reliably join GPT’s capabilities with exterior instruments and APIs.
These fashions have been fine-tuned to each detect when a perform must be known as (relying on the consumer’s enter) and to reply with JSON that adheres to the perform signature. Perform calling permits builders to extra reliably get structured information again from the mannequin. For instance, builders can:
- Create chatbots that reply questions by calling exterior instruments (e.g., like ChatGPT Plugins)
Convert queries corresponding to “Electronic mail Anya to see if she desires to get espresso subsequent Friday” to a perform name like send_email(to: string, physique: string), or “What’s the climate like in Boston?” to get_current_weather(location: string, unit: 'celsius' | 'fahrenheit').
- Convert pure language into API calls or database queries
Convert “Who’re my prime ten clients this month?” to an inner API name corresponding to get_customers_by_revenue(start_date: string, end_date: string, restrict: int), or “What number of orders did Acme, Inc. place final month?” to a SQL question utilizing sql_query(question: string).
- Extract structured information from textual content
Outline a perform known as extract_people_data(folks: [{name: string, birthday: string, location: string}]), to extract all folks talked about in a Wikipedia article.
These use instances are enabled by new API parameters in our /v1/chat/completions endpoint, capabilities and function_call, that permit builders to explain capabilities to the mannequin through JSON Schema, and optionally ask it to name a particular perform. Get began with our developer documentation and add evals in case you discover instances the place perform calling could possibly be improved
