Skip to content

LangSmith SDK for .NET

Nuget package dotnet License: MIT Discord

Features 🔥

  • Fully generated C# SDK based on official OpenAPI specification using AutoSDK
  • Automatic releases of new preview versions if there was an update to the OpenAPI specification
  • All modern .NET features - nullability, trimming, NativeAOT, etc.
  • Support .Net Framework/.Net Standard 2.0

Usage

Initializing

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
using var client = new LangSmithClient();
using var openAiClient = new OpenAiClient();

client.JsonSerializerContext = new SpecialJsonSerializerContext(tryAGI.OpenAI.SourceGenerationContext.Default);

// This can be a user input to your app
var question = "Can you summarize this morning's meetings?";

// This can be retrieved in a retrieval step
const string context = "During this morning's meeting, we solved all world conflict.";
var messages = new[]
{
    "You are a helpful assistant. Please respond to the user's request only based on the given context."
        .AsSystemMessage(),
    $"Question: {question}\\nContext: {context}",
};

// Create parent run
var parentRunId = Guid.NewGuid();
await client.Run.CreateRunAsync(
    name: "Chat Pipeline",
    runType: CreateRunRequestRunType.Chain,
    id: parentRunId,
    inputs: new Dictionary<string, object>
    {
        ["question"] = question,
    });

// Create child run
var childRunId = Guid.NewGuid();
await client.Run.CreateRunAsync(
    name: "OpenAI Call",
    runType: CreateRunRequestRunType.Llm,
    id: childRunId,
    parentRunId: parentRunId,
    inputs: new Dictionary<string, object>
    {
        ["messages"] = messages,
    });

// Generate a completion
var chatCompletion = await openAiClient.Chat.CreateChatCompletionAsync(
    model: CreateChatCompletionRequestModel.Gpt35Turbo,
    messages: messages);

// End runs
await client.Run.UpdateRunAsync(
    runId: childRunId,
    outputs: new Dictionary<string, object>
    {
        ["chatCompletion"] = chatCompletion,
    },
    endTime: DateTime.UtcNow.ToString("O"));
await client.Run.UpdateRunAsync(
    runId: parentRunId,
    outputs: new Dictionary<string, object>
    {
        ["answer"] = chatCompletion.Choices[0].Message.Content ?? string.Empty,
    },
    endTime: DateTime.UtcNow.ToString("O"));

Create Dataset

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
using var client = GetAuthorizedClient();

var response = await client.Datasets.CreateDatasetAsync(name: "test3");
try
{
    var response2 = await client.Datasets.DeleteDatasetAsync(datasetId: response.Id);
}
catch (Exception e)
{
    Console.WriteLine(e);
}

Ok

1
2
using var client = GetAuthorizedClient();
var response = await client.OkAsync();

Trace

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
using var client = GetAuthorizedClient();
using var openAiClient = GetAuthorizedOpenAiClient();

client.JsonSerializerContext = new SpecialJsonSerializerContext(tryAGI.OpenAI.SourceGenerationContext.Default);

// This can be a user input to your app
var question = "Can you summarize this morning's meetings?";

// This can be retrieved in a retrieval step
const string context = "During this morning's meeting, we solved all world conflict.";
var messages = new[]
{
    "You are a helpful assistant. Please respond to the user's request only based on the given context."
        .AsSystemMessage(),
    $"Question: {question}\\nContext: {context}",
};

// Create parent run
// var parentRunId = Guid.NewGuid();
// await client.Run.CreateRunAsync(
//     name: "Chat Pipeline",
//     runType: CreateRunRequestRunType.Chain,
//     id: parentRunId,
//     inputs: new Dictionary<string, object>
//     {
//         ["question"] = question,
//     });
//
// // Create child run
// var childRunId = Guid.NewGuid();
// await client.Run.CreateRunAsync(
//     name: "OpenAI Call",
//     runType: CreateRunRequestRunType.Llm,
//     id: childRunId,
//     parentRunId: parentRunId,
//     inputs: new Dictionary<string, object>
//     {
//         ["messages"] = messages,
//     });

// Generate a completion
var chatCompletion = await openAiClient.Chat.CreateChatCompletionAsync(
    model: CreateChatCompletionRequestModel.Gpt35Turbo,
    messages: messages);

// End runs
// await client.Run.UpdateRunAsync(
//     runId: childRunId,
//     outputs: new Dictionary<string, object>
//     {
//         ["chatCompletion"] = chatCompletion,
//     },
//     endTime: DateTime.UtcNow.ToString("O"));
// await client.Run.UpdateRunAsync(
//     runId: parentRunId,
//     outputs: new Dictionary<string, object>
//     {
//         ["answer"] = chatCompletion.Choices[0].Message.Content ?? string.Empty,
//     },
//     endTime: DateTime.UtcNow.ToString("O"));

Support

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

Acknowledgments

JetBrains logo

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