Cledev.OpenAI

Unofficial .NET SDK for OpenAI with a Blazor Server Playground.

Install-Package Cledev.OpenAI

API

• Models
  • List Models
  • Retrieve Model
• Completions
  • Create Completion
• ChatGPT
  • Create Chat Completion
• ChatGPT-4
  • (coming soon)
• Edits
  • Create Edit
• Images
  • Create Image
  • Create Image Edit
  • Create Image Variation
• Embeddings
  • Create Embeddings
• Audio
  • Create Transcription
  • Create Translation
• Files
  • List Files
  • Upload File
  • Delete File
  • Retrieve File
• Fine-tunes
  • Create Fine-tune
  • List Fine-tunes
  • Retrieve Fine-tune
  • Cancel Fine-tune
  • List Fine-tune Events
  • Delete Fine-tune Model
• Moderations
  • Create Moderation

Utilities

• Azure OpenAI Support
  • (coming soon)

Configuration

Option 1

services.AddOpenAIClient();

This option requires an appsettings.json file

{
  "OpenAI": {
    "ApiKey": "YOUR_API_KEY",
    "Organization": "OUR_ORGANIZATION"
  }
}

Option 2

services.AddOpenAIClient(options =>
{
     options.ApiKey = Environment.GetEnvironmentVariable("YOUR_API_KEY");
     options.Organization = Environment.GetEnvironmentVariable("YOUR_ORGANIZATION");
 });

Important: use environment variables to load the OpenAI API key, do not hardcode it in the source code.

Usage

Inject IOpenAIClient interface into your service

ChatGPT Stream Example

var request = new CreateChatCompletionRequest
{
    Model = ChatModel.Gpt35Turbo.ToStringModel(),
    Stream = true,
    MaxTokens = 500,
    Messages = new List<ChatCompletionMessage>
    {
        new("system", "You are a helpful assistant."),
        new("user", "Who won the world series in 2020?"),
        new("assistant", "The Los Angeles Dodgers won the World Series in 2020."),
        new("user", "Where was it played?")
    }
};

var completions = client.CreateChatCompletionAsStream(request);

await foreach (var completion in completions)
{
    Console.Write(completion.Choices[0].Message?.Content);
}

Completion Stream Example

var request = new CreateCompletionRequest
{
    Model = CompletionsModel.TextDavinciV3.ToStringModel(),
    Stream = true,
    Prompt = "Please write a 1000 word assay about differences between functional programming and object oriented programming",
    MaxTokens = 500
};

var completions = client.CreateCompletionAsStream(request);

await foreach (var completion in completions)
{
    Console.Write(completion.Choices[0].Text);
}

Audio Transcription Example (Whisper)

const string fileName = "YOUR_RECORDING.m4a";
var fileBytes = await File.ReadAllBytesAsync($"Data/{fileName}");

var request = new CreateAudioTranslationRequest
{
    Model = AudioModel.Whisper1.ToStringModel(),
    File = fileBytes,
    FileName = fileName,
    ResponseFormat = AudioResponseFormat.VerboseJson.ToStringFormat()
};

var response = await client.CreateAudioTranslation(request);

Console.Write($"{JsonSerializer.Serialize(response, jsonSerializerOptions)}");

Create Image Example (Dall-E)

var request = new CreateImageRequest
{
    Prompt = "Once upon a time",
    Size = ImageSize.Size512x512.ToStringSize(),
    ResponseFormat = ImageResponseFormat.B64Json.ToStringFormat(),
    N = 1
};

var response = await client.CreateImage(Request);

<img src="@response.Data[0].Url" />

Other Examples

Please, have a look at the Blazor Server Playground for more examples on how to use the client.

Index

ChapGPT

Completions

Edits

Images

Audio Transcriptions