A Swift package for interacting with OpenAI

• Utilizing ChatGPT using the ChatGPT API
• Request N completions using the Completions API
• Requesting N images using the Image Generation API
  • Creating an image mask

Import OpenAI and then add your API key from OpenAI. The API can be initialized directly or accessed using the static shared property.

import OpenAI
OpenAI.shared.connect(with: "your-key")
// or
let openai = OpenAI(credentials: "your-key")

An example using several of the available roles (assistant, system and user) with content.

let chats = try await openai.chats(
 for: .init(
 model: .gpt3(.turbo),
 messages: [
 .system("You are a helpful assistant."),
 .user("Who won the world series in 2020?"),
 .assistant("The Los Angeles Dodgers won the World Series in 2020."),
 .user("Where was it played?")
 ]
 )
)
print(chats)

This request supplies nil values by default for many of the available parameters, which can be supplied for added flexibility.

A simple request where the prompt is echoed back.

let completions = try await openai.completions(
 for: .init(
 model: .gpt3(.davinci),
 prompt: "Say this is a test"
 )
)
print(completions)

This request supplies nil values by default for many of the available parameters, which can be supplied for added flexibility.

A simple request for creating an image of a cat.

let images = try await openai.images(for: "A white siamese cat")
print(images) // images[0].url

This request supplies nil values by default for many of the available parameters, which can be supplied for added flexibility.

License: MIT

Please feel free to open a PR with desired changes and a brief description.

This package is not endorsed by, directly affiliated with, maintained, authorized, or sponsored by OpenAI.