Skip to content

Latest commit

 

History

History
99 lines (76 loc) · 3.67 KB

README.md

File metadata and controls

99 lines (76 loc) · 3.67 KB
Raw

Introduction

liboai is a simple, unofficial C++17 library for the OpenAI API. It allows developers to access OpenAI endpoints through a simple collection of methods and classes. The library can most effectively be thought of as a spiritual port of OpenAI's Python library, simply called openai, due to its similar structure - with few exceptions.

Features

Usage

See below for just how similar in style liboai and its Python alternative are when generating an image using DALL-E.

DALL-E Generation in Python. 
import openai
import os

openai.api_key = os.getenv("OPENAI_API_KEY")
response = openai.Image.create(
    prompt="A snake in the grass!",
    n=1,
    size="256x256"
)
print(response["data"][0]["url"])
DALL-E Generation in C++. 
#include "liboai.h"

using namespace liboai;

int main() {
  OpenAI oai;
  oai.auth.SetKeyEnv("OPENAI_API_KEY");
	
  Response res = oai.Image->create(
    "A snake in the grass!",
    1,
    "256x256"
  );

  std::cout << res["data"][0]["url"] << std::endl;
}

Running the above will print out the URL to the resulting generated image, which may or may not look similar to the one found below.

Example Image

Keep in mind the above C++ example is a minimal example and is not an exception-safe snippet. Please see the documentation for more detailed and exception-safe code snippets.

Dependencies

For the library to work the way it does, it relies on two major dependencies. These dependencies can be found listed below.

If building the library using the provided solution, it is recommended to install these dependencies using vcpkg.

Documentation

For detailed documentation and additional code examples, see the library's documentation here.

Contributing

Artificial intelligence is an exciting and quickly-changing field.

If you'd like to partake in further placing the power of AI in the hands of everyday people, please consider contributing by submitting new code and features via a Pull Request. If you have any issues using the library, or just want to suggest new features, feel free to contact me directly using the info on my profile or open an Issue.