OpenAI

Features 🔥

• Fully generated C# SDK based on official OpenAI OpenAPI specification using AutoSDK
• Same day update to support new features
• Updated and supported automatically if there are no breaking changes
• Contains a supported list of constants such as current prices, models, and other
• Source generator to define functions natively through C# interfaces
• All modern .NET features - nullability, trimming, NativeAOT, etc.
• Support .Net Framework/.Net Standard 2.0
• Support all OpenAI API endpoints including completions, chat, embeddings, images, assistants and more.
• Regularly tested for compatibility with popular custom providers like OpenRouter/DeepSeek/Ollama/LM Studio and many others
• Microsoft.Extensions.AI IChatClient and IEmbeddingGenerator support for OpenAI and all CustomProviders
• Alias-aware routed IChatClient builder with provider fallback and manual 429 cooldown tracking
• FreeLLM package for free-first chat routing across OpenAI-compatible providers and Gemini with OpenAI-compatible and MEAI surfaces

Documentation

Examples and documentation can be found here: https://tryagi.github.io/OpenAI/

Usage

usingvarapi=newOpenAiApi("API_KEY");stringresponse=awaitapi.Chat.CreateChatCompletionAsync(messages:["Generate five random words."],model:ModelIdsSharedEnum.Gpt4oMini);Console.WriteLine(response);// "apple, banana, cherry, date, elderberry"varenumerable=api.Chat.CreateChatCompletionAsStreamAsync(messages:["Generate five random words."],model:ModelIdsSharedEnum.Gpt4oMini);awaitforeach(stringresponseinenumerable){Console.WriteLine(response);}

It uses three implicit conversions:

• from string to ChatCompletionRequestUserMessage. It will always be converted to the user message.
• from ChatCompletionResponseMessage to string . It will always contain the first choice message content.
• from CreateChatCompletionStreamResponse to string . It will always contain the first delta content.

You still can use the full response objects if you need more information, just replace string response to var response.

Tools

usingOpenAI;usingCSharpToJsonSchema;publicenumUnit{Celsius,Fahrenheit,}publicclassWeather{publicstringLocation{get;set;}=string.Empty;publicdoubleTemperature{get;set;}publicUnitUnit{get;set;}publicstringDescription{get;set;}=string.Empty;}[GenerateJsonSchema(Strict=true)]// false by default. You can't use parameters with default values in Strict mode.publicinterfaceIWeatherFunctions{[Description("Get the current weather in a given location")]publicTask<Weather>GetCurrentWeatherAsync([Description("The city and state, e.g. San Francisco, CA")]stringlocation,Unitunit,CancellationTokencancellationToken=default);}publicclassWeatherService:IWeatherFunctions{publicTask<Weather>GetCurrentWeatherAsync(stringlocation,Unitunit=Unit.Celsius,CancellationTokencancellationToken=default){returnTask.FromResult(newWeather{Location=location,Temperature=22.0,Unit=unit,Description="Sunny",});}}usingvarapi=newOpenAiApi("API_KEY");varservice=newWeatherService();vartools=service.AsTools().AsOpenAiTools();varmessages=newList<ChatCompletionRequestMessage>{"You are a helpful weather assistant.".AsSystemMessage(),"What is the current temperature in Dubai, UAE in Celsius?".AsUserMessage(),};varmodel=ModelIdsSharedEnum.Gpt4oMini;varresult=awaitapi.Chat.CreateChatCompletionAsync(messages,model:model,tools:tools);varresultMessage=result.Choices.First().Message;messages.Add(resultMessage.AsRequestMessage());foreach(varcallinresultMessage.ToolCalls){varjson=awaitservice.CallAsync(functionName:call.Function.Name,argumentsAsJson:call.Function.Arguments);messages.Add(json.AsToolMessage(call.Id));}varresult=awaitapi.Chat.CreateChatCompletionAsync(messages,model:model,tools:tools);varresultMessage=result.Choices.First().Message;messages.Add(resultMessage.AsRequestMessage());

> System: You are a helpful weather assistant.
> User: What is the current temperature in Dubai, UAE in Celsius?
> Assistant: call_3sptsiHzKnaxF8bs8BWxPo0B:
GetCurrentWeather({"location":"Dubai, UAE","unit":"celsius"})
> Tool(call_3sptsiHzKnaxF8bs8BWxPo0B):
{"location":"Dubai, UAE","temperature":22,"unit":"celsius","description":"Sunny"}
> Assistant: The current temperature in Dubai, UAE is 22°C with sunny weather.

Structured Outputs

usingOpenAI;usingvarapi=newOpenAiApi("API_KEY");varresponse=awaitapi.Chat.CreateChatCompletionAsAsync<Weather>(messages:["Generate random weather."],model:ModelIdsSharedEnum.Gpt4oMini,jsonSerializerOptions:newJsonSerializerOptions{Converters={newJsonStringEnumConverter()},});// or (if you need trimmable/NativeAOT version)varresponse=awaitapi.Chat.CreateChatCompletionAsAsync(jsonTypeInfo:SourceGeneratedContext.Default.Weather,messages:["Generate random weather."],model:ModelIdsSharedEnum.Gpt4oMini);// response.Value1 contains the structured output// response.Value2 contains the CreateChatCompletionResponse object

Weather:
Location: San Francisco, CA
Temperature: 65
Unit: Fahrenheit
Description: Partly cloudy with a light breeze and occasional sunshine.
Raw Response:
{"Location":"San Francisco, CA","Temperature":65,"Unit":"Fahrenheit","Description":"Partly cloudy with a light breeze and occasional sunshine."}

Additional code for trimmable/NativeAOT version:

[JsonSourceGenerationOptions(Converters=[typeof(JsonStringEnumConverter<Unit>)])][JsonSerializable(typeof(Weather))]publicpartialclassSourceGeneratedContext:JsonSerializerContext;

Custom providers

usingOpenAI;usingvarapi=CustomProviders.GitHubModels("GITHUB_TOKEN");usingvarapi=CustomProviders.Azure("API_KEY","ENDPOINT");usingvarapi=CustomProviders.DeepInfra("API_KEY");usingvarapi=CustomProviders.Groq("API_KEY");usingvarapi=CustomProviders.XAi("API_KEY");usingvarapi=CustomProviders.DeepSeek("API_KEY");usingvarapi=CustomProviders.Fireworks("API_KEY");usingvarapi=CustomProviders.OpenRouter("API_KEY");usingvarapi=CustomProviders.Together("API_KEY");usingvarapi=CustomProviders.GonkaGate("API_KEY");usingvarapi=CustomProviders.Perplexity("API_KEY");usingvarapi=CustomProviders.SambaNova("API_KEY");usingvarapi=CustomProviders.Mistral("API_KEY");usingvarapi=CustomProviders.Codestral("API_KEY");usingvarapi=CustomProviders.Cerebras("API_KEY");usingvarapi=CustomProviders.Cohere("API_KEY");usingvarapi=CustomProviders.Hyperbolic("API_KEY");usingvarapi=CustomProviders.Nebius("API_KEY");usingvarapi=CustomProviders.Nvidia("API_KEY");usingvarapi=CustomProviders.OllamaCloud("API_KEY");usingvarapi=CustomProviders.Minimax("API_KEY");usingvarapi=CustomProviders.NovitaAI("API_KEY");usingvarapi=CustomProviders.Qwen("API_KEY");usingvarapi=CustomProviders.LeptonAI("API_KEY");usingvarapi=CustomProviders.Cleanlab("API_KEY");usingvarapi=CustomProviders.SiliconFlow("API_KEY");usingvarapi=CustomProviders.Inworld("API_KEY_OR_JWT");usingvarapi=CustomProviders.NousPortal("API_KEY");usingvarapi=CustomProviders.VercelAIGateway("API_KEY");usingvarapi=CustomProviders.HuggingFaceRouter("API_KEY");usingvarapi=CustomProviders.GoogleAIStudio("API_KEY");usingvarapi=CustomProviders.Gemini("API_KEY");usingvarapi=CustomProviders.XiaomiMiMo("API_KEY");usingvarapi=CustomProviders.TencentTokenHub("API_KEY");usingvarapi=CustomProviders.TencentTokenHubIntl("API_KEY");usingvarapi=CustomProviders.ZAi("API_KEY");usingvarapi=CustomProviders.Moonshot("API_KEY");usingvarapi=CustomProviders.KimiForCoding("API_KEY");usingvarapi=CustomProviders.MoonshotChina("API_KEY");usingvarapi=CustomProviders.StepFun("API_KEY");usingvarapi=CustomProviders.StepFunStepPlan("API_KEY");usingvarapi=CustomProviders.MiniMaxChina("API_KEY");usingvarapi=CustomProviders.DashScope("API_KEY");usingvarapi=CustomProviders.DashScopeChina("API_KEY");usingvarapi=CustomProviders.DashScopeUnitedStates("API_KEY");usingvarapi=CustomProviders.DashScopeCodingPlan("API_KEY");usingvarapi=CustomProviders.ArceeAI("API_KEY");usingvarapi=CustomProviders.ArceeConductor("API_KEY");usingvarapi=CustomProviders.GmiCloud("API_KEY");usingvarapi=CustomProviders.KiloGateway("API_KEY");usingvarapi=CustomProviders.OpenCodeZen("API_KEY");usingvarapi=CustomProviders.OpenCodeGo("API_KEY");usingvarapi=CustomProviders.Audra("API_KEY");usingvarapi=CustomProviders.Ollama();usingvarapi=CustomProviders.LmStudio();

Microsoft.Extensions.AI

The client natively implements IChatClient and IEmbeddingGenerator<string, Embedding<float>> from Microsoft.Extensions.AI, providing a unified interface across 40+ providers:

usingOpenAI;usingMicrosoft.Extensions.AI;// Works with OpenAI and all CustomProviders (Azure, DeepSeek, Groq, etc.)usingvarclient=newOpenAiClient("API_KEY");// or: using var client = CustomProviders.Groq("API_KEY");// IChatClientIChatClientchatClient=client;varresponse=awaitchatClient.GetResponseAsync("Say hello!",newChatOptions{ModelId="gpt-4o-mini"});Console.WriteLine(response.Messages[0].Text);// Streamingawaitforeach(varupdateinchatClient.GetStreamingResponseAsync("Count to 5.",newChatOptions{ModelId="gpt-4o-mini"})){Console.Write(string.Concat(update.Contents.OfType<TextContent>().Select(c =>c.Text)));}// IEmbeddingGeneratorIEmbeddingGenerator<string,Embedding<float>>generator=client;varembeddings=awaitgenerator.GenerateAsync(["Hello, world!"],newEmbeddingGenerationOptions{ModelId="text-embedding-3-small"});

Routed chat client

For OpenAI-compatible providers, you can build one routed IChatClient with aliases like smart, smart-any, fast, and cheap.

usingMicrosoft.Extensions.AI;usingtryAGI.OpenAI;usingvarrouted=newOpenAiRoutedChatClientBuilder().AddProvider("cerebras",CustomProviders.Cerebras("CEREBRAS_API_KEY"), provider =>provider.AddModel("qwen-3-235b-a22b-instruct-2507", model =>model.AsSmart(priority:100).AsSmartAny(priority:100).SupportsToolCalls().SupportsStructuredOutputs().IsRecurringFree())).AddProvider("groq",CustomProviders.Groq("GROQ_API_KEY"), provider =>provider.AddModel("llama-3.3-70b-versatile", model =>model.AsFast(priority:100).AsCheap(priority:70).AsSmartAny(priority:60).SupportsToolCalls())).AddProvider("openrouter",CustomProviders.OpenRouter("OPENROUTER_API_KEY"), provider =>provider.WithRateLimitCooldown(TimeSpan.FromMinutes(2)).AddModel("openrouter/free", model =>model.AsCheap(priority:100).AsSmartAny(priority:40))).Build();IChatClientchatClient=routed;varresponse=awaitchatClient.GetResponseAsync("Explain the tradeoffs of vector search vs keyword search.",newChatOptions{ModelId=OpenAiModelAliases.Smart});Console.WriteLine(response.Messages[0].Text);

Notes:

• This router is for providers exposed through CustomProviders, i.e. OpenAI-compatible endpoints.
• Provider cooldowns are tracked automatically from 429 responses and common rate-limit headers.
• If smart is exhausted, the router also considers models tagged with smart-any.

FreeLLM

FreeLLM is a separate package in this repo. It depends on tryAGI.OpenAI and Google_Gemini, and gives you one routed chat client across OpenAI-compatible providers and Gemini.

<PackageReferenceInclude="FreeLLM"Version="x.y.z" />

usingMicrosoft.Extensions.AI;usingFreeLLM;usingtryAGI.OpenAI;usingvarclient=newFreeLlmClientBuilder().WithCatalogRefresh(TimeSpan.FromHours(6),routeDiscoveredModels:true)// Curated defaults are applied automatically for popular providers..AddCerebras("CEREBRAS_API_KEY").AddGemini("GEMINI_API_KEY", provider =>provider.WithPriority(320).AddModel("gemini-2.5-flash", model =>model.AsSmart(priority:190).AsSmartAny(priority:190).AsFast(priority:140)).AddModel("gemini-2.5-flash-lite", model =>model.AsCheap(priority:220))).AddOpenRouter("OPENROUTER_API_KEY", provider =>provider.WithPriority(90).RemoveModel("openrouter/free").AddModel("openrouter/free", model =>model.AsCheap(priority:250))).Build();// OpenAI-compatible chat completions APIvarraw=awaitclient.Chat.CreateChatCompletionAsync(newCreateChatCompletionRequest{Value2=newCreateChatCompletionRequestVariant2{Model=FreeLlmModelAliases.Smart,Messages=["Explain vector search vs keyword search."],},});// Microsoft.Extensions.AI APIIChatClientchatClient=client;varmeai=awaitchatClient.GetResponseAsync("Explain vector search vs keyword search.",newChatOptions{ModelId=FreeLlmModelAliases.SmartAny});Console.WriteLine(raw.Choices[0].Message.Content);Console.WriteLine(meai.Messages[0].Text);varcatalogs=client.GetProviderCatalogs();foreach(varcatalogincatalogs){Console.WriteLine($"{catalog.ProviderId}: refreshed={catalog.RefreshedAt:O}, missing={string.Join(", ",catalog.MissingConfiguredModelIds)}");}

Notes:

• FreeLlmModelAliases includes smart, smart-any, fast, and cheap.
• Convenience methods for Gemini, Cerebras, SambaNova, OpenRouter, GitHub Models, Groq, and NVIDIA register curated default models and priorities.
• Use provider.WithPriority(...) to bias whole providers, and model.AsSmart(...), model.AsCheap(...), model.AsFast(...), and model.AsSmartAny(...) to tune alias-specific model priority.
• Use builder.WithCatalogRefresh(...) or provider.WithCatalogRefresh(...) to periodically call provider model catalogs, mark configured models as missing when they disappear, and optionally route newly discovered models when routeDiscoveredModels: true.
• Use client.RefreshProviderCatalogsAsync() to force a catalog refresh and client.GetProviderCatalogs() to inspect configured, discovered, present, missing, and routable models per provider.
• Use provider.ClearModels() or pass useDefaultModels: false to a convenience method if you want a fully manual model list.
• provider.AddModel("existing-model", ...) updates preset models in place, so you can override defaults without duplicating registrations.
• client.Chat preserves raw OpenAI-compatible requests for OpenAI-compatible providers and translates supported chat-completions requests to Gemini when a Gemini model wins routing.
• Gemini translation currently supports CreateChatCompletionRequestVariant2, single-choice text chat, JSON response formats, and data-URI images.
• Raw OpenAI tool schemas/functions, audio/modalities, logprobs, web search, prediction, and remote image URLs are not translated to Gemini; use the MEAI surface for Gemini tool calling.
• Provider cooldowns and last-seen rate-limit data are available through client.GetProviderStatuses().

Constants

All tryGetXXX methods return null if the value is not found.
There also non-try methods that throw an exception if the value is not found.

usingOpenAI;// You can try to get the enum from string using:varmodel=ModelIdsSharedEnumExtensions.ToEnum("gpt-4o")??thrownewException("Invalid model");// Chatvarmodel=ModelIdsSharedEnum.Gpt4oMini;double?priceInUsd=model.TryGetPriceInUsd(inputTokens:500,outputTokens:500)
double?priceInUsd=model.TryGetFineTunePriceInUsd(trainingTokens:500,inputTokens:500,outputTokens:500)
int contextLength = model.TryGetContextLength()// 128_000
int outputLength = model.TryGetOutputLength()// 16_000// Embeddings
var model = CreateEmbeddingRequestModel.TextEmbedding3Small;int?maxInputTokens=model.TryGetMaxInputTokens()// 8191
double?priceInUsd=model.TryGetPriceInUsd(tokens:500)// Images
double?priceInUsd=CreateImageRequestModel.DallE3.TryGetPriceInUsd(size:CreateImageRequestSize.x1024x1024,quality:CreateImageRequestQuality.Hd)// Speech to Text
double?priceInUsd=CreateTranscriptionRequestModel.Whisper1.TryGetPriceInUsd(seconds:60)// Text to Speech
double?priceInUsd=CreateSpeechRequestModel.Tts1Hd.TryGetPriceInUsd(characters:1000)

Chat Completion

Send a simple chat completion request.

usingvarclient=newOpenAiClient(apiKey);stringresponse=awaitclient.Chat.CreateChatCompletionAsync(newCreateChatCompletionRequest{Value2=newCreateChatCompletionRequestVariant2{Messages=["Generate five random words."],Model="gpt-4o-mini",}});Console.WriteLine(response);

Chat Completion Streaming

Stream a chat completion response token by token.

usingvarclient=newOpenAiClient(apiKey);varenumerable=client.Chat.CreateChatCompletionAsStreamAsync(newCreateChatCompletionRequest{Value2=newCreateChatCompletionRequestVariant2{Messages=["Generate five random words."],Model="gpt-4o-mini",}});awaitforeach(stringresponseinenumerable){Console.Write(response);}

Chat With Vision

Send an image to the model for analysis.

usingvarclient=newOpenAiClient(apiKey);CreateChatCompletionResponseresponse=awaitclient.Chat.CreateChatCompletionAsync(newCreateChatCompletionRequest{Value2=newCreateChatCompletionRequestVariant2{Messages=["Please describe the following image.",H.Resources.images_dog_and_cat_png.AsBytes().AsUserMessage(mimeType:"image/png"),],Model="gpt-4o-mini",}});Console.WriteLine(response.Choices[0].Message.Content);

JSON Response Format

Request a response in JSON format.

usingvarclient=newOpenAiClient(apiKey);stringresponse=awaitclient.Chat.CreateChatCompletionAsync(newCreateChatCompletionRequest{Value2=newCreateChatCompletionRequestVariant2{Messages=["Generate five random words as json."],Model="gpt-4o-mini",ResponseFormat=newResponseFormatJsonObject{Type=ResponseFormatJsonObjectType.JsonObject,},}});Console.WriteLine(response);

Structured Outputs

Get structured JSON responses using a C# type as the schema.

usingvarclient=newOpenAiClient(apiKey);varresponse=awaitclient.Chat.CreateChatCompletionAsAsync<WordsResponse>(messages:["Generate five random words as json."],model:"gpt-4o-mini");Console.WriteLine("Words:");foreach(varwordinresponse.Value1!.Words){Console.WriteLine(word);}

Structured Outputs (AOT)

Get structured JSON responses using a JsonTypeInfo for AOT/trimming compatibility.

usingvarclient=newOpenAiClient(apiKey);varresponse=awaitclient.Chat.CreateChatCompletionAsAsync(jsonTypeInfo:SourceGeneratedContext.Default.WordsResponse,messages:["Generate five random words."],model:"gpt-4o-mini");Console.WriteLine("Words:");foreach(varwordinresponse.Value1!.Words){Console.WriteLine(word);}

Embeddings

Create a text embedding vector.

usingvarclient=newOpenAiClient(apiKey);varresponse=awaitclient.Embeddings.CreateEmbeddingAsync(input:"Hello, world",model:CreateEmbeddingRequestModel.TextEmbedding3Small);foreach(vardatainresponse.Data.ElementAt(0).Embedding1){Console.WriteLine($"{data}");}

Image Generation

Generate an image from a text prompt.

usingvarclient=newOpenAiClient(apiKey);varresponse=awaitclient.Images.CreateImageAsync(prompt:"a white siamese cat",model:CreateImageRequestModel.GptImage1Mini,n:1,quality:CreateImageRequestQuality.Low,size:CreateImageRequestSize.x1024x1024,outputFormat:CreateImageRequestOutputFormat.Png);varbase64=response.Data?.ElementAt(0).B64Json;Console.WriteLine($"Generated image ({base64?.Length} base64 chars)");

Text To Speech

Convert text to speech audio using streaming.

usingvarclient=newOpenAiClient(apiKey);usingvarmemoryStream=newMemoryStream();awaitforeach(varstreamEventinclient.Audio.CreateSpeechAsync(model:CreateSpeechRequestModel.Gpt4oMiniTts,input:"Hello! This is a text-to-speech test.",voice:(VoiceIdsShared)VoiceIdsSharedEnum.Alloy,responseFormat:CreateSpeechRequestResponseFormat.Mp3,speed:1.0,streamFormat:CreateSpeechRequestStreamFormat.Sse)){if(streamEvent.SpeechAudioDeltais{}delta){byte[]chunk=Convert.FromBase64String(delta.Audio);memoryStream.Write(chunk,0,chunk.Length);}}byte[]audio=memoryStream.ToArray();Console.WriteLine($"Generated {audio.Length} bytes of audio.");

List Models

List all available models.

usingvarclient=newOpenAiClient(apiKey);varmodels=awaitclient.Models.ListModelsAsync();foreach(varmodelinmodels.Data){Console.WriteLine(model.Id);}

Moderation

Check text for policy violations using the moderation endpoint.

usingvarclient=newOpenAiClient(apiKey);varresponse=awaitclient.Moderations.CreateModerationAsync(input:"Hello, world",model:CreateModerationRequestModel.OmniModerationLatest);Console.WriteLine($"Flagged: {response.Results.First().Flagged}");

MEAI Chat Completion

Use the Microsoft.Extensions.AI IChatClient interface for chat completions.

usingvarclient=newOpenAiClient(apiKey);// using Meai = Microsoft.Extensions.AI;Meai.IChatClientchatClient=client;varmessages=newList<Meai.ChatMessage>{new(Meai.ChatRole.User,"Say hello in exactly 3 words."),};varresponse=awaitchatClient.GetResponseAsync(messages,newMeai.ChatOptions{ModelId="gpt-4o-mini"});Console.WriteLine(response.Messages[0].Text);

MEAI Chat Streaming

Stream a chat completion using the Microsoft.Extensions.AI IChatClient interface.

usingvarclient=newOpenAiClient(apiKey);// using Meai = Microsoft.Extensions.AI;Meai.IChatClientchatClient=client;varmessages=newList<Meai.ChatMessage>{new(Meai.ChatRole.User,"Count from 1 to 5."),};awaitforeach(varupdateinchatClient.GetStreamingResponseAsync(messages,newMeai.ChatOptions{ModelId="gpt-4o-mini"})){vartext=string.Concat(update.Contents.OfType<Meai.TextContent>().Select(c =>c.Text));if(!string.IsNullOrEmpty(text)){Console.Write(text);}}

MEAI Tool Calling

Use function/tool calling via the Microsoft.Extensions.AI IChatClient interface.

usingvarclient=newOpenAiClient(apiKey);// using Meai = Microsoft.Extensions.AI;Meai.IChatClientchatClient=client;vartool=Meai.AIFunctionFactory.Create((stringcity)=>cityswitch{"Paris"=>"22C, sunny","London"=>"15C, cloudy",
_ =>"Unknown",},name:"GetWeather",description:"Gets the current weather for a city");varchatOptions=newMeai.ChatOptions{ModelId="gpt-4o-mini",Tools=[tool],};varmessages=newList<Meai.ChatMessage>{new(Meai.ChatRole.User,"What's the weather in Paris? Respond with the temperature only."),};// First turn: get tool callvarresponse=awaitchatClient.GetResponseAsync((IEnumerable<Meai.ChatMessage>)messages,chatOptions);varfunctionCall=response.Messages.SelectMany(m =>m.Contents).OfType<Meai.FunctionCallContent>().First();// Execute tool and add resultvartoolResult=awaittool.InvokeAsync(functionCall.Argumentsis{}args?newMeai.AIFunctionArguments(args):null);messages.AddRange(response.Messages);messages.Add(newMeai.ChatMessage(Meai.ChatRole.Tool,newMeai.AIContent[]{newMeai.FunctionResultContent(functionCall.CallId,toolResult),}));// Second turn: get final responsevarfinalResponse=awaitchatClient.GetResponseAsync((IEnumerable<Meai.ChatMessage>)messages,chatOptions);Console.WriteLine(finalResponse.Messages[0].Text);

MEAI Embeddings

Generate embeddings using the Microsoft.Extensions.AI IEmbeddingGenerator interface.

usingvarclient=newOpenAiClient(apiKey);// using Meai = Microsoft.Extensions.AI;Meai.IEmbeddingGenerator<string,Meai.Embedding<float>>generator=client;varresult=awaitgenerator.GenerateAsync(newList<string>{"Hello, world!"},newMeai.EmbeddingGenerationOptions{ModelId="text-embedding-3-small",});Console.WriteLine($"Embedding dimension: {result[0].Vector.Length}");

Support

Priority place for bugs: https://github.com/tryAGI/OpenAI/issues
Priority place for ideas and general questions: https://github.com/tryAGI/OpenAI/discussions
Discord: https://discord.gg/Ca2xhfBf3v

Acknowledgments

This project is supported by JetBrains through the Open Source Support Program.