钦娅芬 发表于 2025-6-2 23:51:39

使用C#构建一个同时问多个LLM并总结的小工具

前言

在AI编程时代,如果自己能够知道一些可行的解决方案,那么描述清楚交给AI,可以有很大的帮助。
但是我们往往不知道真正可行的解决方案是什么?
我自己有过这样的经历,遇到一个需求,我不知道有哪些解决方案,就去问AI,然后AI输出一大堆东西,我一个个去试,然后再换个AI问,又提出了不同的解决方案。
在换AI问与一个个试的过程中好像浪费了很多时间。
突然出现了一个想法,不是可以一下子把问题丢给多个AI,然后再总结一下出现最多的三个方案。那么这三个方案可行的概率会大一点。然后再丢给Cursor或者Cline等AI编程工具帮我们实现一下。
这样做的缺点是比起直接在网页上问,调用API需要耗费Token,但是硅基流动给我赠送了很多额度还没用完,随便玩一下。
实现效果:


实现方案

实现方案也很简单,如下图所示:

先设计一下布局:
<UserControl xmlns="https://github.com/avaloniaui"
             xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
             xmlns:d="http://schemas.microsoft.com/expression/blend/2008"
             xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
             mc:Ignorable="d" d:DesignWidth="800" d:DesignHeight="450"
                       xmlns:vm="using:AIE_Studio.ViewModels"
                       x:DataType="vm:DuoWenViewModel"
             x:>
        <StackPanel>
                <TextBox Text="{Binding Question}"></TextBox>
                <Button Content="提问" Command="{Binding DuoWenStreamingParallelCommand}" Margin="5"/>
                <ScrollViewer VerticalScrollBarVisibility="Auto">
                <Grid>
                        <Grid.RowDefinitions>
                                <RowDefinition Height="*"/>
                                <RowDefinition Height="*"/>
                        </Grid.RowDefinitions>
                        <Grid.ColumnDefinitions>
                                <ColumnDefinition Width="*"/>
                                <ColumnDefinition Width="*"/>
                                <ColumnDefinition Width="*"/>
                        </Grid.ColumnDefinitions>
                       
                        <StackPanel Grid.Row="0" Grid.Column="0">
                                <TextBlock Text="{Binding Title1}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result1}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                       
                        <StackPanel Grid.Row="0" Grid.Column="1">
                                <TextBlock Text="{Binding Title2}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result2}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                       
                        <StackPanel Grid.Row="0" Grid.Column="2">
                                <TextBlock Text="{Binding Title3}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result3}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                       
                        <StackPanel Grid.Row="1" Grid.Column="0">
                                <TextBlock Text="{Binding Title4}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result4}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                       
                        <StackPanel Grid.Row="1" Grid.Column="1">
                                <TextBlock Text="{Binding Title5}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result5}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                       
                        <StackPanel Grid.Row="1" Grid.Column="2">
                                <TextBlock Text="{Binding Title6}" Margin="5"/>
                                <ScrollViewer VerticalScrollBarVisibility="Auto">
                                        <TextBox Text="{Binding Result6}" AcceptsReturn="True" Margin="5" Height="300"/>
                                </ScrollViewer>
                        </StackPanel>
                </Grid>
                </ScrollViewer>
        </StackPanel>
</UserControl>
在ViewModel中先来看一下最原始的显示结果的方式:

private async Task DuoWen()
{
   ApiKeyCredential apiKeyCredential = new ApiKeyCredential("your api key");

   OpenAIClientOptions openAIClientOptions = new OpenAIClientOptions();
   openAIClientOptions.Endpoint = new Uri("https://api.siliconflow.cn/v1");
   
   IChatClient client1 =
   new OpenAI.Chat.ChatClient("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result1 = await client1.GetResponseAsync(Question);

   Result1 = result1.ToString();

   IChatClient client2 =
   new OpenAI.Chat.ChatClient("Qwen/Qwen3-235B-A22B", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result2 = await client2.GetResponseAsync(Question);

   Result2 = result2.ToString();

   IChatClient client3 =
   new OpenAI.Chat.ChatClient("THUDM/GLM-Z1-32B-0414", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result3 = await client3.GetResponseAsync(Question);

   Result3 = result3.ToString();

   IChatClient client4 =
   new OpenAI.Chat.ChatClient("THUDM/GLM-4-32B-0414", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result4 = await client4.GetResponseAsync(Question);

   Result4 = result4.ToString();

   IChatClient client5 =
    new OpenAI.Chat.ChatClient("deepseek-ai/DeepSeek-R1", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result5 = await client5.GetResponseAsync(Question);

   Result5 = result5.ToString();

   IChatClient client6 =
   new OpenAI.Chat.ChatClient("deepseek-ai/DeepSeek-V3", apiKeyCredential, openAIClientOptions).AsChatClient();

   var result6 = await client6.GetResponseAsync(Question);

   Result6 = result6.ToString();这种最简单的方式是非流式的并且也不是并行的,你会发现一个结束了才会继续向下一个提问。
但至少已经成功显示结果了,现在想要实现的是有一个窗体进行总结。
窗体设计:
<Window xmlns="https://github.com/avaloniaui"
      xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
      xmlns:d="http://schemas.microsoft.com/expression/blend/2008"
      xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
      mc:Ignorable="d" d:DesignWidth="600" d:DesignHeight="450"
                xmlns:vm="using:AIE_Studio.ViewModels"
      x:
                x:DataType="vm:ShowResultWindowViewModel"
      Title="ShowResultWindow">
        <StackPanel>
                <TextBlock Text="最终结果:" Margin="5" />
                <ScrollViewer VerticalScrollBarVisibility="Auto">
                        <TextBox Text="{Binding ReceivedValue}" AcceptsReturn="True" Margin="5" Height="400"/>
                </ScrollViewer>
        </StackPanel>
</Window>窗体的ViewModel:
public partial class ShowResultWindowViewModel : ViewModelBase
{
   
    private string? receivedValue;      
}然后只要在全部都有结果之后,再进行一下总结即可。
IChatClient client7 =
new OpenAI.Chat.ChatClient("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions).AsChatClient();
List<Microsoft.Extensions.AI.ChatMessage> messages = new List<Microsoft.Extensions.AI.ChatMessage>();

string prompt = $"""
      请分析以下各个助手给出的方案,选择其中提到最多的3种方案。
      助手1:{result1}
      助手2:{result2}
      助手3:{result3}
      助手4:{result4}
      助手5:{result5}
      助手6:{result6}
      """;
messages.Add(new Microsoft.Extensions.AI.ChatMessage(ChatRole.User, prompt));
var result7 = await client7.GetResponseAsync(messages);

var showWindow = _serviceProvider.GetRequiredService<ShowResultWindow>();
var showWindowViewModel = _serviceProvider.GetRequiredService<ShowResultWindowViewModel>();
showWindowViewModel.ReceivedValue = result7.ToString();
showWindow.DataContext = showWindowViewModel;
showWindow.Show();以上就成功实现了。
但是还是有可以改进的地方,首先是并行,一个一个问不如同时问。

private async Task DuoWenParallel()
{
   ApiKeyCredential apiKeyCredential = new ApiKeyCredential("your api key");

   OpenAIClientOptions openAIClientOptions = new OpenAIClientOptions();
   openAIClientOptions.Endpoint = new Uri("https://api.siliconflow.cn/v1");

   // 创建一个列表来存储所有的任务
   var tasks = new List<Task<string>>();

   // 向每个助手发送请求并将任务添加到列表中
   tasks.Add(GetResponseFromClient("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions));
   tasks.Add(GetResponseFromClient("Qwen/Qwen3-235B-A22B", apiKeyCredential, openAIClientOptions));
   tasks.Add(GetResponseFromClient("THUDM/GLM-Z1-32B-0414", apiKeyCredential, openAIClientOptions));
   tasks.Add(GetResponseFromClient("THUDM/GLM-4-32B-0414", apiKeyCredential, openAIClientOptions));
   tasks.Add(GetResponseFromClient("deepseek-ai/DeepSeek-R1", apiKeyCredential, openAIClientOptions));
   tasks.Add(GetResponseFromClient("deepseek-ai/DeepSeek-V3", apiKeyCredential, openAIClientOptions));

   // 等待所有任务完成
   var results = await Task.WhenAll(tasks);

   // 将结果分配给相应的属性
   Result1 = results;
   Result2 = results;
   Result3 = results;
   Result4 = results;
   Result5 = results;
   Result6 = results;
}

private async Task<string> GetResponseFromClient(string model, ApiKeyCredential apiKeyCredential, OpenAIClientOptions options)
{
      IChatClient client = new OpenAI.Chat.ChatClient(model, apiKeyCredential, options).AsChatClient();
      var result = await client.GetResponseAsync(Question);
      return result.ToString();
}现在虽然是并行了,但是只有等到所有助手都回答了之后,才会统一显示,用户体验也不好。
改成流式:

private async Task DuoWenStreaming()
{
    ApiKeyCredential apiKeyCredential = new ApiKeyCredential("your api key");

    OpenAIClientOptions openAIClientOptions = new OpenAIClientOptions();
    openAIClientOptions.Endpoint = new Uri("https://api.siliconflow.cn/v1");

    //string question = "C#如何获取鼠标滑动选中的值?请告诉我一些可能的方案,每个方案只需用一句话描述即可,不用展开说明。";

    IChatClient client1 =
    new OpenAI.Chat.ChatClient("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions).AsChatClient();

    await foreach (var item in client1.GetStreamingResponseAsync(Question))
    {
       Result1 += item.ToString();
    }         
}现在查看效果:

最后再改造成流式+并行就好了。

private async Task DuoWenStreamingParallel()
{
   ApiKeyCredential apiKeyCredential = new ApiKeyCredential("your api key");

   OpenAIClientOptions openAIClientOptions = new OpenAIClientOptions();
   openAIClientOptions.Endpoint = new Uri("https://api.siliconflow.cn/v1");

   // Clear previous results
   Result1 = Result2 = Result3 = Result4 = Result5 = Result6 = string.Empty;

   // Create a list of tasks for parallel processing
   var tasks = new List<Task>
   {
         ProcessStreamingResponse("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions, (text) => Result1 += text),
         ProcessStreamingResponse("Qwen/Qwen3-235B-A22B", apiKeyCredential, openAIClientOptions, (text) => Result2 += text),
         ProcessStreamingResponse("THUDM/GLM-Z1-32B-0414", apiKeyCredential, openAIClientOptions, (text) => Result3 += text),
         ProcessStreamingResponse("THUDM/GLM-4-32B-0414", apiKeyCredential, openAIClientOptions, (text) => Result4 += text),
         ProcessStreamingResponse("deepseek-ai/DeepSeek-R1", apiKeyCredential, openAIClientOptions, (text) => Result5 += text),
         ProcessStreamingResponse("deepseek-ai/DeepSeek-V3", apiKeyCredential, openAIClientOptions, (text) => Result6 += text)
   };

   // Wait for all streaming responses to complete
   await Task.WhenAll(tasks);

   IChatClient client7 =
   new OpenAI.Chat.ChatClient("Qwen/Qwen2.5-72B-Instruct", apiKeyCredential, openAIClientOptions).AsChatClient();
   List<Microsoft.Extensions.AI.ChatMessage> messages = new List<Microsoft.Extensions.AI.ChatMessage>();

   string prompt = $"""
         请分析以下各个助手给出的方案,选择其中提到最多的3种方案。
         助手1:{Result1}
         助手2:{Result2}
         助手3:{Result3}
         助手4:{Result4}
         助手5:{Result5}
         助手6:{Result6}
         """;
   messages.Add(new Microsoft.Extensions.AI.ChatMessage(ChatRole.User, prompt));
   var result7 = await client7.GetResponseAsync(messages);

   var showWindow = _serviceProvider.GetRequiredService<ShowResultWindow>();
   var showWindowViewModel = _serviceProvider.GetRequiredService<ShowResultWindowViewModel>();
   showWindowViewModel.ReceivedValue = result7.ToString();
   showWindow.DataContext = showWindowViewModel;
   showWindow.Show();
}

private async Task ProcessStreamingResponse(string model, ApiKeyCredential apiKeyCredential, OpenAIClientOptions options, Action<string> updateResult)
{
    IChatClient client = new OpenAI.Chat.ChatClient(model, apiKeyCredential, options).AsChatClient();
   
    await foreach (var item in client.GetStreamingResponseAsync(Question))
    {
      updateResult(item.ToString());
    }
}这里使用了一个带有一个参数的委托来更新每个助手回复的结果。
现在再查看效果:

Qwen/Qwen3-235B-A22B、THUDM/GLM-Z1-32B-0414、deepseek-ai/DeepSeek-R1有思考过程,返回结果比较慢。
目前Microsoft.Extensions.AI.OpenAI好像还无法获取思考内容。

等待久一会之后,可以看到结果都出来了:

然后总结窗口会显示最终的总结内容:

确定方案之后可以让Cursor或者Cline帮我们写一下试试。

来源:程序园用户自行投稿发布,如果侵权,请联系站长删除
免责声明:如果侵犯了您的权益,请联系站长,我们会及时删除侵权内容,谢谢合作!
页: [1]
查看完整版本: 使用C#构建一个同时问多个LLM并总结的小工具