Azure.AI.OpenAI.Assistants
1.0.0-beta.4
Prefix Reserved
dotnet add package Azure.AI.OpenAI.Assistants --version 1.0.0-beta.4
NuGet\Install-Package Azure.AI.OpenAI.Assistants -Version 1.0.0-beta.4
<PackageReference Include="Azure.AI.OpenAI.Assistants" Version="1.0.0-beta.4" />
paket add Azure.AI.OpenAI.Assistants --version 1.0.0-beta.4
#r "nuget: Azure.AI.OpenAI.Assistants, 1.0.0-beta.4"
// Install Azure.AI.OpenAI.Assistants as a Cake Addin #addin nuget:?package=Azure.AI.OpenAI.Assistants&version=1.0.0-beta.4&prerelease // Install Azure.AI.OpenAI.Assistants as a Cake Tool #tool nuget:?package=Azure.AI.OpenAI.Assistants&version=1.0.0-beta.4&prerelease
Azure OpenAI: OpenAI Assistants client library for .NET
The Azure OpenAI Assistants client library for .NET is an adaptation of OpenAI's REST APIs that provides an idiomatic interface and rich integration with the rest of the Azure SDK ecosystem. It will connect to Azure OpenAI resources or to the non-Azure OpenAI inference endpoint, making it a great choice for even non-Azure OpenAI development.
Use this library to:
- Create and manage assistants, threads, messages, and runs
- Configure and use tools with assistants
- Upload and manage files for use with assistants
Getting started
Prerequisites
To use Assistants capabilities, you'll need service API access through OpenAI or Azure OpenAI:
- To use OpenAI (api.openai.com), you'll need an API key obtained from a developer account at https://platform.openai.com
- To use an Azure OpenAI resource, you must have an Azure subscription and Azure OpenAI access. This will allow you to create an Azure OpenAI resource and get both a connection URL as well as API keys. For more information, see Quickstart: Get started generating text using Azure OpenAI Service.
Install the package
Install the client library for .NET with NuGet:
dotnet add package Azure.AI.OpenAI.Assistants --prerelease
Authenticate the client
See OpenAI's "how assistants work" documentation for an overview of the concepts and relationships used with assistants. This overview closely follows OpenAI's overview example to demonstrate the basics of creating, running, and using assistants and threads.
To get started, create an AssistantsClient
:
AssistantsClient client = isAzureOpenAI
? new AssistantsClient(new Uri(azureResourceUrl), new AzureKeyCredential(azureApiKey))
: new AssistantsClient(nonAzureApiKey);
NOTE: The Assistants API should always be used from a trusted device. Because the same authentication mechanism for running threads also allows changing persistent resources like Assistant instructions, a malicious user could extract an API key and modify Assistant behavior for other customers.
Key concepts
Overview
For an overview of Assistants and the pertinent key concepts like Threads, Messages, Runs, and Tools, please see OpenAI's Assistants API overview.
Usage
Examples
With an authenticated client, an assistant can be created:
Response<Assistant> assistantResponse = await client.CreateAssistantAsync(
new AssistantCreationOptions("gpt-4-1106-preview")
{
Name = "Math Tutor",
Instructions = "You are a personal math tutor. Write and run code to answer math questions.",
Tools = { new CodeInterpreterToolDefinition() }
});
Assistant assistant = assistantResponse.Value;
Next, create a thread:
Response<AssistantThread> threadResponse = await client.CreateThreadAsync();
AssistantThread thread = threadResponse.Value;
With a thread created, messages can be created on it:
Response<ThreadMessage> messageResponse = await client.CreateMessageAsync(
thread.Id,
MessageRole.User,
"I need to solve the equation `3x + 11 = 14`. Can you help me?");
ThreadMessage message = messageResponse.Value;
A run can then be started that evaluates the thread against an assistant:
Response<ThreadRun> runResponse = await client.CreateRunAsync(
thread.Id,
new CreateRunOptions(assistant.Id)
{
AdditionalInstructions = "Please address the user as Jane Doe. The user has a premium account.",
});
ThreadRun run = runResponse.Value;
Once the run has started, it should then be polled until it reaches a terminal status:
do
{
await Task.Delay(TimeSpan.FromMilliseconds(500));
runResponse = await client.GetRunAsync(thread.Id, runResponse.Value.Id);
}
while (runResponse.Value.Status == RunStatus.Queued
|| runResponse.Value.Status == RunStatus.InProgress);
Assuming the run successfully completed, listing messages from the thread that was run will now reflect new information added by the assistant:
Response<PageableList<ThreadMessage>> afterRunMessagesResponse
= await client.GetMessagesAsync(thread.Id);
IReadOnlyList<ThreadMessage> messages = afterRunMessagesResponse.Value.Data;
// Note: messages iterate from newest to oldest, with the messages[0] being the most recent
foreach (ThreadMessage threadMessage in messages)
{
Console.Write($"{threadMessage.CreatedAt:yyyy-MM-dd HH:mm:ss} - {threadMessage.Role,10}: ");
foreach (MessageContent contentItem in threadMessage.ContentItems)
{
if (contentItem is MessageTextContent textItem)
{
Console.Write(textItem.Text);
}
else if (contentItem is MessageImageFileContent imageFileItem)
{
Console.Write($"<image from ID: {imageFileItem.FileId}");
}
Console.WriteLine();
}
}
Example output from this sequence:
2023-11-14 20:21:23 - assistant: The solution to the equation \(3x + 11 = 14\) is \(x = 1\).
2023-11-14 20:21:18 - user: I need to solve the equation `3x + 11 = 14`. Can you help me?
Working with files for retrieval
Files can be uploaded and then referenced by assistants or messages. First, use the generalized upload API with a purpose of 'assistants' to make a file ID available:
File.WriteAllText(
path: "sample_file_for_upload.txt",
contents: "The word 'apple' uses the code 442345, while the word 'banana' uses the code 673457.");
Response<OpenAIFile> uploadAssistantFileResponse = await client.UploadFileAsync(
localFilePath: "sample_file_for_upload.txt",
purpose: OpenAIFilePurpose.Assistants);
OpenAIFile uploadedAssistantFile = uploadAssistantFileResponse.Value;
Once uploaded, the file ID can then be provided to an assistant upon creation. Note that file IDs will only be used if an appropriate tool like Code Interpreter or Retrieval is enabled.
Response<Assistant> assistantResponse = await client.CreateAssistantAsync(
new AssistantCreationOptions("gpt-4-1106-preview")
{
Name = "SDK Test Assistant - Retrieval",
Instructions = "You are a helpful assistant that can help fetch data from files you know about.",
Tools = { new RetrievalToolDefinition() },
FileIds = { uploadedAssistantFile.Id },
});
Assistant assistant = assistantResponse.Value;
With a file ID association and a supported tool enabled, the assistant will then be able to consume the associated data when running threads.
Using function tools and parallel function calling
As described in OpenAI's documentation for assistant tools, tools that reference caller-defined capabilities as functions can be provided to an assistant to allow it to dynamically resolve and disambiguate during a run.
Here, outlined is a simple assistant that "knows how to," via caller-provided functions:
- Get the user's favorite city
- Get a nickname for a given city
- Get the current weather, optionally with a temperature unit, in a city
To do this, begin by defining the functions to use -- the actual implementations here are merely representative stubs.
// Example of a function that defines no parameters
string GetUserFavoriteCity() => "Seattle, WA";
FunctionToolDefinition getUserFavoriteCityTool = new("getUserFavoriteCity", "Gets the user's favorite city.");
// Example of a function with a single required parameter
string GetCityNickname(string location) => location switch
{
"Seattle, WA" => "The Emerald City",
_ => throw new NotImplementedException(),
};
FunctionToolDefinition getCityNicknameTool = new(
name: "getCityNickname",
description: "Gets the nickname of a city, e.g. 'LA' for 'Los Angeles, CA'.",
parameters: BinaryData.FromObjectAsJson(
new
{
Type = "object",
Properties = new
{
Location = new
{
Type = "string",
Description = "The city and state, e.g. San Francisco, CA",
},
},
Required = new[] { "location" },
},
new JsonSerializerOptions() { PropertyNamingPolicy = JsonNamingPolicy.CamelCase }));
// Example of a function with one required and one optional, enum parameter
string GetWeatherAtLocation(string location, string temperatureUnit = "f") => location switch
{
"Seattle, WA" => temperatureUnit == "f" ? "70f" : "21c",
_ => throw new NotImplementedException()
};
FunctionToolDefinition getCurrentWeatherAtLocationTool = new(
name: "getCurrentWeatherAtLocation",
description: "Gets the current weather at a provided location.",
parameters: BinaryData.FromObjectAsJson(
new
{
Type = "object",
Properties = new
{
Location = new
{
Type = "string",
Description = "The city and state, e.g. San Francisco, CA",
},
Unit = new
{
Type = "string",
Enum = new[] { "c", "f" },
},
},
Required = new[] { "location" },
},
new JsonSerializerOptions() { PropertyNamingPolicy = JsonNamingPolicy.CamelCase }));
With the functions defined in their appropriate tools, an assistant can be now created that has those tools enabled:
Response<Assistant> assistantResponse = await client.CreateAssistantAsync(
// note: parallel function calling is only supported with newer models like gpt-4-1106-preview
new AssistantCreationOptions("gpt-4-1106-preview")
{
Name = "SDK Test Assistant - Functions",
Instructions = "You are a weather bot. Use the provided functions to help answer questions. "
+ "Customize your responses to the user's preferences as much as possible and use friendly "
+ "nicknames for cities whenever possible.",
Tools =
{
getUserFavoriteCityTool,
getCityNicknameTool,
getCurrentWeatherAtLocationTool,
},
});
Assistant assistant = assistantResponse.Value;
If the assistant calls tools, the calling code will need to resolve ToolCall
instances into matching
ToolOutput
instances. For convenience, a basic example is extracted here:
ToolOutput GetResolvedToolOutput(RequiredToolCall toolCall)
{
if (toolCall is RequiredFunctionToolCall functionToolCall)
{
if (functionToolCall.Name == getUserFavoriteCityTool.Name)
{
return new ToolOutput(toolCall, GetUserFavoriteCity());
}
using JsonDocument argumentsJson = JsonDocument.Parse(functionToolCall.Arguments);
if (functionToolCall.Name == getCityNicknameTool.Name)
{
string locationArgument = argumentsJson.RootElement.GetProperty("location").GetString();
return new ToolOutput(toolCall, GetCityNickname(locationArgument));
}
if (functionToolCall.Name == getCurrentWeatherAtLocationTool.Name)
{
string locationArgument = argumentsJson.RootElement.GetProperty("location").GetString();
if (argumentsJson.RootElement.TryGetProperty("unit", out JsonElement unitElement))
{
string unitArgument = unitElement.GetString();
return new ToolOutput(toolCall, GetWeatherAtLocation(locationArgument, unitArgument));
}
return new ToolOutput(toolCall, GetWeatherAtLocation(locationArgument));
}
}
return null;
}
To handle user input like "what's the weather like right now in my favorite city?", polling the response for completion
should be supplemented by a RunStatus
check for RequiresAction
or, in this case, the presence of the
RequiredAction
property on the run. Then, the collection of ToolOutputSubmissions
should be submitted to the
run via the SubmitRunToolOutputs
method so that the run can continue:
do
{
await Task.Delay(TimeSpan.FromMilliseconds(500));
runResponse = await client.GetRunAsync(thread.Id, runResponse.Value.Id);
if (runResponse.Value.Status == RunStatus.RequiresAction
&& runResponse.Value.RequiredAction is SubmitToolOutputsAction submitToolOutputsAction)
{
List<ToolOutput> toolOutputs = new();
foreach (RequiredToolCall toolCall in submitToolOutputsAction.ToolCalls)
{
toolOutputs.Add(GetResolvedToolOutput(toolCall));
}
runResponse = await client.SubmitToolOutputsToRunAsync(runResponse.Value, toolOutputs);
}
}
while (runResponse.Value.Status == RunStatus.Queued
|| runResponse.Value.Status == RunStatus.InProgress);
Note that, when using supported models, the assistant may request that several functions be called in parallel. Older models may only call one function at a time.
Once all needed function calls have been resolved, the run will proceed normally and the completed messages on the thread will contain model output supplemented by the provided function tool outputs.
Troubleshooting
When you interact with Azure OpenAI using the .NET SDK, errors returned by the service correspond to the same HTTP status codes returned for REST API requests.
For example, if you try to create a client using an endpoint that doesn't match your Azure OpenAI Resource endpoint, a 404
error is returned, indicating Resource Not Found
.
Next steps
- Provide a link to additional code examples, ideally to those sitting alongside the README in the package's
/samples
directory. - If appropriate, point users to other packages that might be useful.
- If you think there's a good chance that developers might stumble across your package in error (because they're searching for specific functionality and mistakenly think the package provides that functionality), point them to the packages they might be looking for.
Contributing
See the Azure SDK CONTRIBUTING.md for details on building, testing, and contributing to this library.
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit cla.microsoft.com.
When you submit a pull request, a CLA-bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., label, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
Product | Versions Compatible and additional computed target framework versions. |
---|---|
.NET | net5.0 was computed. net5.0-windows was computed. net6.0 was computed. net6.0-android was computed. net6.0-ios was computed. net6.0-maccatalyst was computed. net6.0-macos was computed. net6.0-tvos was computed. net6.0-windows was computed. net7.0 was computed. net7.0-android was computed. net7.0-ios was computed. net7.0-maccatalyst was computed. net7.0-macos was computed. net7.0-tvos was computed. net7.0-windows was computed. net8.0 was computed. net8.0-android was computed. net8.0-browser was computed. net8.0-ios was computed. net8.0-maccatalyst was computed. net8.0-macos was computed. net8.0-tvos was computed. net8.0-windows was computed. |
.NET Core | netcoreapp2.0 was computed. netcoreapp2.1 was computed. netcoreapp2.2 was computed. netcoreapp3.0 was computed. netcoreapp3.1 was computed. |
.NET Standard | netstandard2.0 is compatible. netstandard2.1 was computed. |
.NET Framework | net461 was computed. net462 was computed. net463 was computed. net47 was computed. net471 was computed. net472 was computed. net48 was computed. net481 was computed. |
MonoAndroid | monoandroid was computed. |
MonoMac | monomac was computed. |
MonoTouch | monotouch was computed. |
Tizen | tizen40 was computed. tizen60 was computed. |
Xamarin.iOS | xamarinios was computed. |
Xamarin.Mac | xamarinmac was computed. |
Xamarin.TVOS | xamarintvos was computed. |
Xamarin.WatchOS | xamarinwatchos was computed. |
-
.NETStandard 2.0
- Azure.Core (>= 1.39.0)
- System.ClientModel (>= 1.0.0)
- System.Text.Json (>= 4.7.2)
NuGet packages (3)
Showing the top 3 NuGet packages that depend on Azure.AI.OpenAI.Assistants:
Package | Downloads |
---|---|
MedbaseLibrary
Shared code for Medbase projects. |
|
Mythosia.AI
## Package Summary The `Mythosia.AI` library provides an abstraction for various AI models, including support for **OpenAI GPT-4** and **Anthropic Claude** models. This library allows easy interaction with AI services, enabling both synchronous completion and streaming capabilities for models like GPT-3.5, GPT-4, Claude, and more. This library abstracts the HTTP requests required to communicate with the APIs and provides a unified interface for various AI models, making it easier to integrate into any C# or ASP.NET Core project. ### Supported Models - OpenAI GPT-4 and GPT-4 Turbo - OpenAI GPT-3.5 Turbo - Claude 3.x variants (Claude3_5Sonnet, Claude3Opus, Claude3Haiku) ### Key Features - Synchronous completion and streaming support for AI models - Customizable model selection, temperature, and max tokens - Abstraction over the complexity of managing HTTP requests and responses - Extendable structure for adding support for new AI models ## How to Use To use this library, create an instance of either `ChatGptService` or `ClaudeService` depending on the AI model you're working with. Here's an example: |
|
Mythosia.Azure.AI
Mythosia.Azure.AI - AssistantAI Client Library Mythosia.Azure.AI provides an easy-to-use client library for interacting with Azure OpenAI Assistants. This library integrates the power of Azure's AI services, allowing you to create and manage assistant threads, send user inputs, and retrieve assistant-generated responses in an asynchronous manner. It also includes support for managing sensitive credentials using Azure Key Vault. Features: Assistant Management: Easily create and manage Azure OpenAI Assistants and conversation threads. Message Handling: Send user input to the assistant, run conversations asynchronously, and retrieve the assistant’s responses. Azure Key Vault Integration: Safely manage your endpoint and key secrets using Azure Key Vault with SecretFetcher. Lazy Initialization: Efficient resource management with lazy initialization for both assistant and thread objects. Custom Search Index Integration: The SearchIndexClient enables advanced integration with Azure Search if needed. Key Classes: AssistantAI: Core class for interacting with Azure OpenAI Assistants. You can initialize the assistant using either Key Vault secrets or connection strings. SecretFetcher: Utility class for retrieving sensitive information such as API keys and connection strings from Azure Key Vault. Example Usage: // Using Key Vault secrets to initialize the AssistantAI var assistantAI = new AssistantAI(endpointSecretFetcher, keySecretFetcher, assistantOptions); // Using direct connection string and key to initialize the AssistantAI var assistantAI = new AssistantAI("https://my-endpoint", "my-api-key", assistantOptions); // Run a conversation with the assistant string response = await assistantAI.RunAsync("Hello, how can you assist me?"); Console.WriteLine(response); // Start a new thread and run a conversation string newThreadResponse = await assistantAI.RunOnNewThreadAsync("Start a new session"); Console.WriteLine(newThreadResponse); |
GitHub repositories (1)
Showing the top 1 popular GitHub repositories that depend on Azure.AI.OpenAI.Assistants:
Repository | Stars |
---|---|
microsoft/teams-ai
SDK focused on building AI based applications and extensions for Microsoft Teams and other Bot Framework channels
|
Version | Downloads | Last updated |
---|---|---|
1.0.0-beta.4 | 81,582 | 5/1/2024 |
1.0.0-beta.3 | 132,074 | 3/7/2024 |
1.0.0-beta.2 | 9,721 | 2/6/2024 |
1.0.0-beta.1 | 185 | 2/1/2024 |