LeapAi.SDK.Providers.xAI 2.0.1

<p align="center"> <img src="assets/leap_logo.png" alt="Leap AI Logo" width="120"> </p>

LeapAi.SDK.Providers.xAI

<p align="center"> <a href="https://www.nuget.org/packages/LeapAi.SDK.Providers.xAI"><img src="https://img.shields.io/nuget/v/LeapAi.SDK.Providers.xAI.svg" alt="NuGet" /></a> </p>

xAI Grok provider for the Leap AI SDK — a provider-agnostic .NET toolkit for building AI-powered applications, chatbots, and agents against a unified LeapClient.

Installation

dotnet add package LeapAi.SDK.Providers.xAI

You will also need the Core package (pulled in automatically as a transitive dependency):

dotnet add package LeapAi.SDK.Core

Quick Start

using Leap.AI.Core;
using Leap.AI.Providers.xAI;

var leap = LeapClient.Create()
    .UseXAi("xai-...")          // grok-3 by default
    .UseLogging()
    .UseRetry(maxRetries: 3)
    .Build();

string result = await leap.GenerateTextAsync("Explain the Fermi paradox.");
Console.WriteLine(result);

Supported Models

Override the model at build time:

var leap = LeapClient.Create()
    .UseXAi("xai-...", defaultModel: "grok-3-fast")
    .Build();

Or via the options delegate:

var leap = LeapClient.Create()
    .UseXAi(o =>
    {
        o.ApiKey       = "xai-...";
        o.DefaultModel = "grok-3-mini";
    })
    .Build();

Usage

Text Generation

string answer = await leap.GenerateTextAsync("What is a transformer model?");

Conversation History

var messages = new List<ChatMessage>
{
    ChatMessage.System("You are a concise technical assistant."),
    ChatMessage.User("What is xAI's Grok?")
};

var response = await leap.GenerateAsync(messages);
Console.WriteLine(response.Text);

Streaming

await foreach (var chunk in leap.StreamAsync("Write a haiku about AGI."))
{
    Console.Write(chunk.Text);
}

Structured Output

public record MarketSummary(string Ticker, double Price, string Sentiment);

var summary = await leap.GenerateObjectAsync<MarketSummary>(
    "Summarise the current state of $TSLA in one concise JSON object."
);

Console.WriteLine($"{summary.Ticker}: ${summary.Price} — {summary.Sentiment}");

Tool Calling (Agents)

using Leap.AI.Core.Tools;

public record SearchArgs(string Query);

var searchTool = FunctionTool<SearchArgs>.Create(
    name: "web_search",
    description: "Searches the web and returns a summary of the top results.",
    handler: args => $"Top result for '{args.Query}': ..."
);

var leap = LeapClient.Create()
    .UseXAi("xai-...")
    .UseTool(searchTool)
    .Build();

var response = await leap.GenerateTextAsync("Search for the latest xAI news.");

ASP.NET Core / Dependency Injection

Install the DI extension package:

dotnet add package LeapAi.SDK.Extensions.DependencyInjection

using Leap.AI.Extensions.DependencyInjection;

builder.Services.AddLeap(leap => leap
    .UseXAi(builder.Configuration["xAI:ApiKey"]!)
    .UseLogging()
    .UseRetry(3));

Inject LeapClient anywhere in your app:

public class MyService(LeapClient ai)
{
    public Task<string> AskAsync(string q) => ai.GenerateTextAsync(q);
}

Available Packages

Community & Contributing

The Leap AI SDK community lives on our GitHub repository. Questions, ideas, and contributions are all welcome!