microsoft / microsoft/semantic-kernel

NET: How to disable think mode when calling the ollama model

Open
#13,733 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C#
Stars
28.6k
Forks
4.8k
Avg merge
14h 13m
Merged PRs (30d)
18

Description

Microsoft.SemanticKernel:

public static async Task Main()
{
    Console.OutputEncoding = System.Text.Encoding.UTF8;
    var builder = Kernel.CreateBuilder();
    var modelId = "gemma4:e4b-it-q4_K_M";
    var endpoint = new Uri("http://localhost:11434");

    builder.Services.AddOllamaChatCompletion(modelId, endpoint);
    var kernel = builder.Build();
    var chatCompletionService = kernel.GetRequiredService<IChatCompletionService>();

    string prompt = "c#写一个计算当前月天数的方法";
    var executionSettings = new OllamaPromptExecutionSettings
    {
        FunctionChoiceBehavior = FunctionChoiceBehavior.Auto(autoInvoke: false),
    };
    var response = await chatCompletionService.GetChatMessageContentAsync(prompt, kernel: kernel, executionSettings: executionSettings);
    Console.Write(response.Content);
}

How to disable think mode when calling the ollama model?

It is recommended to add a method like chatOptions.AddOllamaOption(OllamaOption.Think, false);
Microsoft.Extensions.AI:

public async static Task Main()
{
    OllamaApiClient ollamaClient = new OllamaApiClient(new Uri("http://localhost:11434/"), "gemma4:e4b-it-q4_K_M");

    // 3. 调用 Ollama API
    var client = new Microsoft.Extensions.AI.ChatClientBuilder(ollamaClient)
        .UseFunctionInvocation()
        .Build();

    ChatOptions chatOptions = new()
    {
        Tools = [AIFunctionFactory.Create(QueryDateTime), AIFunctionFactory.Create(SaveFile), AIFunctionFactory.Create(ReadFile)],
    };
    chatOptions.AddOllamaOption(OllamaOption.Think, false);//Disable think
    await foreach (var res in client.GetStreamingResponseAsync("c#写一个计算当前月天数的方法", chatOptions))//,
    {
        Console.Write(res.Text);
    }
}

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the Ollama chat-completion integration used by Microsoft.SemanticKernel and the OllamaPromptExecutionSettings shown in the report. Compare its options with the Microsoft.Extensions.AI example, then verify that callers can set the Ollama think option to false and that the model request disables thinking.

Written by the indexing model from the issue text.

Assessment

Tech stack
csharp, ollama
Domain
ai
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
55/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.