Table of Contents

Getting started

OllamaSharp provides .NET bindings for the Ollama API, simplifying interactions with Ollama both locally and remotely. It provides asynchronous streaming, progress reporting and convenience classes and functions to simplify common use cases.

Getting started with OllamaSharp only requires a running Ollama server and a supported version of .NET.

Prerequisites

Using Ollama with OllamaSharp

To use Ollama from your code base, you'll need to create and initialize an instance of the OllamaApiClient. This client wraps each Ollama API endpoint in awaitable methods that fully support response streaming.

// set up the client
var uri = new Uri("http://localhost:11434");
var ollama = new OllamaApiClient(uri);

// select a model which should be used for further operations
ollama.SelectedModel = "qwen3.5:35b-a3b";

Or use the shorter constructor that sets the model directly:

var ollama = new OllamaApiClient("http://localhost:11434", "qwen3.5:35b-a3b");

For advanced scenarios (e.g. Native AOT), you can use the Configuration class:

var ollama = new OllamaApiClient(new OllamaApiClient.Configuration
{
    Uri = new Uri("http://localhost:11434"),
    Model = "qwen3.5:35b-a3b",
});

Once your client is initialized, you can list local models, pull new models from the Ollama model hub and build interactive chats with them.

Listing all models that are available locally

var models = await ollama.ListLocalModelsAsync();

Pulling a model and reporting progress

await foreach (var status in ollama.PullModelAsync("qwen3.5:35b-a3b"))
    Console.WriteLine($"{status.Percent}% {status.Status}");

Generating a completion directly into the console

await foreach (var stream in ollama.GenerateAsync("How are you today?"))
    Console.Write(stream.Response);

Building interactive chats

The Chat class automatically tracks message history across turns so the model always has full context. You can optionally pass a system prompt to shape the model's behaviour.

// messages including their roles and tool calls will automatically be tracked within the chat object
// and are accessible via the Messages property

var chat = new Chat(ollama);

while (true)
{
    var message = Console.ReadLine();
    await foreach (var answerToken in chat.SendAsync(message))
        Console.Write(answerToken);
}

For a complete guide to the Chat class — including system prompts, multi-modal inputs, structured output, tool calls and reasoning models — see the Chat and Generate page.

For managing models (pull, push, copy, delete, embeddings and more) see the Model Management page.

Custom headers

OllamaSharp supports two levels of custom HTTP headers — useful for authentication, proxies or any other middleware between your app and the Ollama server.

Default headers (sent with every request)

var ollama = new OllamaApiClient("http://localhost:11434");
ollama.DefaultRequestHeaders["Authorization"] = "Bearer your-api-key";

Per-request headers

Every request model inherits from OllamaRequest and exposes a CustomHeaders dictionary for one-off headers:

var request = new GenerateRequest
{
    Prompt = "Hello!",
};
request.CustomHeaders["X-Request-Id"] = Guid.NewGuid().ToString();

await foreach (var chunk in ollama.GenerateAsync(request))
    Console.Write(chunk?.Response);

Cloud models (Ollama Turbo)

OllamaSharp can talk to Ollama cloud models as well. Use the constructor that accepts an HttpClient so you can attach your API key:

var client = new HttpClient();
client.BaseAddress = new Uri("http://localhost:11434");
client.DefaultRequestHeaders.Add(/* your api key here */);

var ollama = new OllamaApiClient(client);
Tip

You can also use DefaultRequestHeaders on OllamaApiClient instead of configuring the HttpClient directly.

Usage with Microsoft.Extensions.AI

Microsoft built an abstraction library to streamline the usage of different AI providers. This is a really interesting concept if you plan to build apps that might use different providers, like ChatGPT, Claude and local models with Ollama.

I encourage you to read their accouncement Introducing Microsoft.Extensions.AI Preview – Unified AI Building Blocks for .NET.

OllamaSharp is the first full implementation of their IChatClient and IEmbeddingGenerator that makes it possible to use Ollama just like every other chat provider.

To do this, simply use the OllamaApiClient as IChatClient instead of IOllamaApiClient.

// install package Microsoft.Extensions.AI.Abstractions

private static IChatClient CreateChatClient(Arguments arguments)
{
  if (arguments.Provider.Equals("ollama", StringComparison.OrdinalIgnoreCase))
    return new OllamaApiClient(arguments.Uri, arguments.Model);
  else
    return new OpenAIChatClient(new OpenAI.OpenAIClient(arguments.ApiKey), arguments.Model); // ChatGPT or compatible
}

Note that IOllamaApiClient provides many Ollama specific methods that IChatClient and IEmbeddingGenerator miss.

Because these are abstractions, IChatClient and IEmbeddingGenerator will never implement the full Ollama API specification. However, OllamaApiClient implements three interfaces: the native IOllamaApiClient and Microsoft IChatClient and IEmbeddingGenerator<string, Embedding<float>> which allows you to cast it to any of these interfaces as you need them at any time.

Note

For a comparison of OllamaSharp vs. Microsoft.Extensions.AI vs. Semantic Kernel, see the README.

Edit this page