LightResults Extensions

Extensions for LightResults, an extremely light and modern Operation Result Pattern library for .NET.

OpenAI

Provides comprehensive OpenAI (v2.9.0) integration with the Result pattern, offering extension methods that return Result<T> instead of throwing exceptions.

Documentation

Make sure to read the docs for the full API.

Try overloads

This package provides a Try extension method version of all public methods for all OpenAI clients, wrapping operations in a try { } catch { } block. If an exception occurs, a failed result will be returned and the Exception will be added to the result as metadata.

Supported Clients:

Assistants - AssistantClient extensions for assistant operations
Audio Processing - AudioClient extensions for speech-to-text and text-to-speech
Batch Processing - BatchClient extensions for batch operations
Chat Completions - ChatClient extensions for chat completions and streaming
Containers - ContainerClient extensions for container operations
Conversations - ConversationClient extensions for conversation operations
Embeddings - EmbeddingClient extensions for text embedding operations
Evaluations - EvaluationClient extensions for evaluation operations
File Operations - OpenAIFileClient extensions for file and upload operations
Fine-tuning - FineTuningClient extensions for fine-tuning operations
Graders - GraderClient extensions for grader operations
Image Generation - ImageClient extensions for image generation and editing
Moderation - ModerationClient extensions for moderation operations
OpenAI Root Client - OpenAIClient extensions for client factory helpers
Model Operations - OpenAIModelClient extensions for model operations
Realtime - RealtimeClient extensions for realtime operations
Realtime Sessions - RealtimeSessionClient extensions for realtime session operations
Responses - ResponsesClient extensions for responses and streaming, including token-count and compact helpers
Vector Stores - VectorStoreClient extensions for vector store operations
Video - VideoClient extensions for video generation operations

Usage Examples

Chat Completions

using OpenAI.Chat;
using LightResults.Extensions.OpenAI;

var client = new ChatClient("gpt-4o", Environment.GetEnvironmentVariable("OPENAI_API_KEY"));

// Simple chat completion
var result = await client.TryCompleteChatAsync("What is the meaning of life?");
if (result.IsSuccess(out var completion))
{
    Console.WriteLine($"Response: {completion.Content[0].Text}");
}
else if (result.IsFailure(out var error))
{
    Console.WriteLine($"Error: {error.Message}");
}

// Chat with multiple messages
var messages = new List<ChatMessage>
{
    new SystemChatMessage("You are a helpful assistant."),
    new UserChatMessage("Explain quantum computing in simple terms.")
};

var chatResult = await client.TryCompleteChatAsync(messages);
if (chatResult.IsSuccess(out var chatCompletion))
{
    Console.WriteLine($"Response: {chatCompletion.Content[0].Text}");
}

Embeddings

using OpenAI.Embeddings;
using LightResults.Extensions.OpenAI;

var embeddingClient = new EmbeddingClient("text-embedding-3-small", apiKey);

// Generate single embedding
var embeddingResult = await embeddingClient.TryGenerateEmbeddingAsync("Hello, world!");
if (embeddingResult.IsSuccess(out var embedding))
{
    var vector = embedding.ToFloats();
    Console.WriteLine($"Generated embedding with {vector.Length} dimensions");
}

// Generate multiple embeddings
var texts = new[] { "First text", "Second text", "Third text" };
var embeddingsResult = await embeddingClient.TryGenerateEmbeddingsAsync(texts);
if (embeddingsResult.IsSuccess(out var embeddings))
{
    Console.WriteLine($"Generated {embeddings.Count} embeddings");
}

Image Generation

using OpenAI.Images;
using LightResults.Extensions.OpenAI;

var imageClient = new ImageClient("dall-e-3", apiKey);

// Generate image
var imageResult = await imageClient.TryGenerateImageAsync(
    "A serene landscape with mountains and a lake at sunset");

if (imageResult.IsSuccess(out var image))
{
    Console.WriteLine($"Generated image URL: {image.ImageUri}");
}

// Generate multiple images
var imagesResult = await imageClient.TryGenerateImagesAsync(
    "A cute cat wearing a hat", imageCount: 2);

if (imagesResult.IsSuccess(out var images))
{
    foreach (var generatedImage in images)
    {
        Console.WriteLine($"Image URL: {generatedImage.ImageUri}");
    }
}

Audio Processing

using OpenAI.Audio;
using LightResults.Extensions.OpenAI;

var audioClient = new AudioClient("whisper-1", apiKey);

// Transcribe audio
var audioData = File.ReadAllBytes("audio.mp3");
var transcriptionResult = await audioClient.TryTranscribeAudioAsync(audioData, "audio.mp3");
if (transcriptionResult.IsSuccess(out var transcription))
{
    Console.WriteLine($"Transcription: {transcription.Text}");
}

// Generate speech
var speechClient = new AudioClient("tts-1", apiKey);
var speechResult = await speechClient.TryGenerateSpeechAsync("Hello, world!", GeneratedSpeechVoice.Alloy);
if (speechResult.IsSuccess(out var speechData))
{
    await File.WriteAllBytesAsync("output.mp3", speechData.ToArray());
}

File Operations

using OpenAI.Files;
using LightResults.Extensions.OpenAI;

var fileClient = new OpenAIFileClient(apiKey);

// Upload file
var fileData = File.ReadAllBytes("document.txt");
var uploadResult = await fileClient.TryUploadFileAsync(fileData, "document.txt", FileUploadPurpose.Assistants);
if (uploadResult.IsSuccess(out var uploadedFile))
{
    Console.WriteLine($"Uploaded file ID: {uploadedFile.Id}");
}

// List files
var filesResult = await fileClient.TryGetFilesAsync();
if (filesResult.IsSuccess(out var files))
{
    foreach (var file in files)
    {
        Console.WriteLine($"File: {file.Filename} (ID: {file.Id})");
    }
}

Error Handling

All extension methods wrap operations in a try { } catch { } block and return Result<T> types for safe error handling:

var result = await client.TryCompleteChatAsync("Hello!");
if (result.IsFailure(out var error))
{
    var ex = error.Exception;

    // Handle specific exception types
    if (ex is ClientResultException clientException)
    {
        Console.WriteLine($"API Error: {clientException.Message}");
    }
    else if (ex is TaskCanceledException)
    {
        Console.WriteLine("Request was cancelled");
    }
    else
    {
        Console.WriteLine($"Unexpected error: {ex?.Message ?? error.Message}");
    }
}

Getting the Exception

var result = await client.TryCompleteChatAsync("Hello!");
if (result.IsFailure(out var error))
{
    var ex = error.Exception;
    // Do something with the base exception type or...

    if (ex is ClientResultException clientResultException)
    {
        // Handle OpenAI-specific errors
        Console.WriteLine($"OpenAI API Error: {clientResultException.Message}");
    }
}

LightResults Extensions

Extensions for LightResults, an extremely light and modern Operation Result Pattern library for .NET.

OpenAI

Provides comprehensive OpenAI (v2.9.0) integration with the Result pattern, offering extension methods that return Result<T> instead of throwing exceptions.

Documentation

Make sure to read the docs for the full API.

Try overloads

Supported Clients:

Assistants - AssistantClient extensions for assistant operations
Audio Processing - AudioClient extensions for speech-to-text and text-to-speech
Batch Processing - BatchClient extensions for batch operations
Chat Completions - ChatClient extensions for chat completions and streaming
Containers - ContainerClient extensions for container operations
Conversations - ConversationClient extensions for conversation operations
Embeddings - EmbeddingClient extensions for text embedding operations
Evaluations - EvaluationClient extensions for evaluation operations
File Operations - OpenAIFileClient extensions for file and upload operations
Fine-tuning - FineTuningClient extensions for fine-tuning operations
Graders - GraderClient extensions for grader operations
Image Generation - ImageClient extensions for image generation and editing
Moderation - ModerationClient extensions for moderation operations
OpenAI Root Client - OpenAIClient extensions for client factory helpers
Model Operations - OpenAIModelClient extensions for model operations
Realtime - RealtimeClient extensions for realtime operations
Realtime Sessions - RealtimeSessionClient extensions for realtime session operations
Responses - ResponsesClient extensions for responses and streaming, including token-count and compact helpers
Vector Stores - VectorStoreClient extensions for vector store operations
Video - VideoClient extensions for video generation operations

Usage Examples

Chat Completions

using OpenAI.Chat;
using LightResults.Extensions.OpenAI;

var client = new ChatClient("gpt-4o", Environment.GetEnvironmentVariable("OPENAI_API_KEY"));

// Simple chat completion
var result = await client.TryCompleteChatAsync("What is the meaning of life?");
if (result.IsSuccess(out var completion))
{
    Console.WriteLine($"Response: {completion.Content[0].Text}");
}
else if (result.IsFailure(out var error))
{
    Console.WriteLine($"Error: {error.Message}");
}

// Chat with multiple messages
var messages = new List<ChatMessage>
{
    new SystemChatMessage("You are a helpful assistant."),
    new UserChatMessage("Explain quantum computing in simple terms.")
};

var chatResult = await client.TryCompleteChatAsync(messages);
if (chatResult.IsSuccess(out var chatCompletion))
{
    Console.WriteLine($"Response: {chatCompletion.Content[0].Text}");
}

Embeddings

using OpenAI.Embeddings;
using LightResults.Extensions.OpenAI;

var embeddingClient = new EmbeddingClient("text-embedding-3-small", apiKey);

// Generate single embedding
var embeddingResult = await embeddingClient.TryGenerateEmbeddingAsync("Hello, world!");
if (embeddingResult.IsSuccess(out var embedding))
{
    var vector = embedding.ToFloats();
    Console.WriteLine($"Generated embedding with {vector.Length} dimensions");
}

// Generate multiple embeddings
var texts = new[] { "First text", "Second text", "Third text" };
var embeddingsResult = await embeddingClient.TryGenerateEmbeddingsAsync(texts);
if (embeddingsResult.IsSuccess(out var embeddings))
{
    Console.WriteLine($"Generated {embeddings.Count} embeddings");
}

Image Generation

using OpenAI.Images;
using LightResults.Extensions.OpenAI;

var imageClient = new ImageClient("dall-e-3", apiKey);

// Generate image
var imageResult = await imageClient.TryGenerateImageAsync(
    "A serene landscape with mountains and a lake at sunset");

if (imageResult.IsSuccess(out var image))
{
    Console.WriteLine($"Generated image URL: {image.ImageUri}");
}

// Generate multiple images
var imagesResult = await imageClient.TryGenerateImagesAsync(
    "A cute cat wearing a hat", imageCount: 2);

if (imagesResult.IsSuccess(out var images))
{
    foreach (var generatedImage in images)
    {
        Console.WriteLine($"Image URL: {generatedImage.ImageUri}");
    }
}

Audio Processing

using OpenAI.Audio;
using LightResults.Extensions.OpenAI;

var audioClient = new AudioClient("whisper-1", apiKey);

// Transcribe audio
var audioData = File.ReadAllBytes("audio.mp3");
var transcriptionResult = await audioClient.TryTranscribeAudioAsync(audioData, "audio.mp3");
if (transcriptionResult.IsSuccess(out var transcription))
{
    Console.WriteLine($"Transcription: {transcription.Text}");
}

// Generate speech
var speechClient = new AudioClient("tts-1", apiKey);
var speechResult = await speechClient.TryGenerateSpeechAsync("Hello, world!", GeneratedSpeechVoice.Alloy);
if (speechResult.IsSuccess(out var speechData))
{
    await File.WriteAllBytesAsync("output.mp3", speechData.ToArray());
}

File Operations

using OpenAI.Files;
using LightResults.Extensions.OpenAI;

var fileClient = new OpenAIFileClient(apiKey);

// Upload file
var fileData = File.ReadAllBytes("document.txt");
var uploadResult = await fileClient.TryUploadFileAsync(fileData, "document.txt", FileUploadPurpose.Assistants);
if (uploadResult.IsSuccess(out var uploadedFile))
{
    Console.WriteLine($"Uploaded file ID: {uploadedFile.Id}");
}

// List files
var filesResult = await fileClient.TryGetFilesAsync();
if (filesResult.IsSuccess(out var files))
{
    foreach (var file in files)
    {
        Console.WriteLine($"File: {file.Filename} (ID: {file.Id})");
    }
}

Error Handling

All extension methods wrap operations in a try { } catch { } block and return Result<T> types for safe error handling:

var result = await client.TryCompleteChatAsync("Hello!");
if (result.IsFailure(out var error))
{
    var ex = error.Exception;

    // Handle specific exception types
    if (ex is ClientResultException clientException)
    {
        Console.WriteLine($"API Error: {clientException.Message}");
    }
    else if (ex is TaskCanceledException)
    {
        Console.WriteLine("Request was cancelled");
    }
    else
    {
        Console.WriteLine($"Unexpected error: {ex?.Message ?? error.Message}");
    }
}

Getting the Exception

var result = await client.TryCompleteChatAsync("Hello!");
if (result.IsFailure(out var error))
{
    var ex = error.Exception;
    // Do something with the base exception type or...

    if (ex is ClientResultException clientResultException)
    {
        // Handle OpenAI-specific errors
        Console.WriteLine($"OpenAI API Error: {clientResultException.Message}");
    }
}

jscarle/LightResults.Extensions.OpenAIv10.0.3

Get Started

Readme

LightResults Extensions

OpenAI

Documentation

Try overloads

Usage Examples

Chat Completions

Embeddings

Image Generation

Audio Processing

File Operations

Error Handling

Getting the Exception

jscarle/LightResults.Extensions.OpenAIv10.0.3

Get Started

Readme

LightResults Extensions

OpenAI

Documentation

Try overloads

Usage Examples

Chat Completions

Embeddings

Image Generation

Audio Processing

File Operations

Error Handling

Getting the Exception