Learning .NET MAUI Blazor (6), pseudo ChatGPT based on OpenAI interface

ChatGPT needs no introduction. Since January to the present, it has been extremely popular. There are also various tutorials on the Internet, and there are even ones that claim to be the domestic version of ChatGPT. So does ChatGPT have an open API interface, and how do those applications that use ChatGPT realize it?
In fact, although the domestic environment cannot access ChatGPT, it can access openai. Open openai to see the introduction of ChatGPT.

Digression: openai spent a lot of money to buy ai.com. Visit ai.com and jump directly to the ChatGPT page, which is very convenient.

About ChatGPT

About ChatGPT: ChatGPT is fine-tuned from models in the GPT-3.5 series, which completed training in early 2022. However, the interface of ChatGPT has not been opened yet, so the vast majority of such applications on the market should be counterfeit.
So is there a way to implement an application similar to ChatGPT? The answer is yes, that is to use OpenAI's open interface, the latest interface is GPT-3, and the model will be trained by the end of 2021. In the text model of GPT-3, there are 4 models available, namely:

• text-davinci-003
• text-curie-001
• text-babbage-001
• text-ada-001
  has different functions,
  text-davinci-003 has the strongest comprehension ability, what it is good at: complex intentions, causal relationship, summarizing the
  advantage of text-curie-001 is that it is very fast, what it is good at is : language translation, complex categorization, text sentiment, summary
  text-babbage-001 is also very capable in terms of how well semantic search ranks documents to search queries, what it excels at are: medium categorization, semantic search categorization
  text-ada- 001 is good at: parsing text, simple classification, address correction, keywords

Therefore, according to actual needs, different models can be used to build the best ChatGPT application. Then the shortcomings of the ChatGPT application built using the API provided by GPT-3 are also obvious:

• The accuracy of the reply content is for reference only
• More suitable for open questions.
• Context is not supported.
• It is difficult to have both speed and reply length.

demo

Having written so much, let's take a look at the simple pseudo-ChatGPT application created using MUUI Blazor + MASA + OpenAI.
注意：为什么叫伪ChatGPT呢？因为ChatGPT是基于GPT-3.5，API方式是调用GPT-3，有很大的差距，因为GPT-3.5接口还未开放

Implementation code

create project

Create a .NET MAUI Blazor application, give the project a name, let's call it OpenAITest

the target framework and choose .NET 7.0. Click Create.

Quote from MASA.Blazor

In order to develop quickly, we need a UI framework. MASA.Blazor is a UI framework developed by Chinese people, and we must support it.
Installation dependencies:

NuGet\Install-Package Masa.Blazor -Version 1.0.0-preview.3

Add resource files:
Open wwwrootthe directory index.htmland add the following content:

<link href="_content/Masa.Blazor/css/masa-blazor.min.css" rel="stylesheet" />
<link href="https://cdn.masastack.com/npm/@mdi/[email protected]/css/materialdesignicons.min.css" rel="stylesheet">
<link href="https://cdn.masastack.com/npm/materialicons/materialicons.css" rel="stylesheet">
<link href="https://cdn.masastack.com/npm/fontawesome/v5.0.13/css/all.css" rel="stylesheet">
<script src="_content/BlazorComponent/js/blazor-component.js"></script>

Injection service:
open the root directory MauiProgram.cs, add builder.Services.AddMasaBlazor();.

Global reference:
Open the root directory _Imports.razorand add the application:

@using Masa.Blazor
@using BlazorComponent

Open it Shared/MainLayout.razorand change it to the following code:

@inherits LayoutComponentBase
<MApp>
    <div class="page">
        <div class="sidebar">
            <NavMenu />
        </div>
        <main>
            @Body
        </main>
    </div>
</MApp>

The MASA.Blazor UI framework is installed

Install OpenAI SDK dependencies

Now there are many packaged OpenAI SDKs. According to the comparison, I chose OpenAI-DotNetthis library. Of course, you can also choose other SDK libraries according to your needs. It is used more on nuget

• OpenAI
• Betalgo.OpenAI.GPT3
• AI.Dev.OpenAI.GPT
• OpenAI.Net

Installation OpenAI-DotNetdependencies:

NuGet\Install-Package OpenAI-DotNet -Version 4.4.4

Create a new directory under the root directory Modeland add OpenAIConfig.csfiles. Write the information obtained on openai keyto the file. There are a lot of online tutorials on how to get the key of openai. Very simple.

internal class OpenAIConfig
    {
    
    
        /// <summary>
        /// OpenAI申请的Key
        /// </summary>
        public const string api_key = "openai key";
        /// <summary>
        /// OpenAI所填写的组织
        /// </summary>
        public const string api_org = "";

    }

Build another one QA.cs. Used to record questions and answers.

internal class QA
    {
    
    
        public string question {
    
     get; set; }
        public string answer {
    
     get; set; }
    }

implement the interface

With the basic settings and tools in place, start designing the interface you need.
Delete unnecessary files:

• Open the root directory MauiProgram.csand delete it builder.Services.AddSingleton<WeatherForecastService>();.
• Delete the directory under the root directory Dataand all files under the directory.
• Open it Shared/NavMenu.razorand modify it as follows:

<div class="top-row ps-3 navbar navbar-dark">
    <div class="container-fluid">
        <a class="navbar-brand" href="">OpenAI</a>
        <button title="Navigation menu" class="navbar-toggler" @onclick="ToggleNavMenu">
            <span class="navbar-toggler-icon"></span>
        </button>
    </div>
</div>

<div class="@NavMenuCssClass" @onclick="ToggleNavMenu">
    <nav class="flex-column">
        <div class="nav-item px-3">
            <NavLink class="nav-link" href="" Match="NavLinkMatch.All">
                <span class="oi oi-home" aria-hidden="true"></span> 文本
            </NavLink>
        </div>
        <div class="nav-item px-3">
            <NavLink class="nav-link" href="code">
                <span class="oi oi-plus" aria-hidden="true"></span> 代码
            </NavLink>
        </div>
        <div class="nav-item px-3">
            <NavLink class="nav-link" href="image">
                <span class="oi oi-list-rich" aria-hidden="true"></span> 图像
            </NavLink>
        </div>
    </nav>
</div>

@code {
    private bool collapseNavMenu = true;

    private string NavMenuCssClass => collapseNavMenu ? "collapse" : null;

    private void ToggleNavMenu()
    {
        collapseNavMenu = !collapseNavMenu;
    }
}

• Delete Sharedthe directory SurveyPrompt.razor.
• Delete Pagesthe Counter.razor, FetchData.razor, add Code.razorandImage.razor

The modified Index.razorhtml code snippet is changed to:

<MContainer Class="spacing-playground pa-6 pr h-100" Fluid>
    <MSkeletonLoader Class="mb-12 h-full w-100" Boilerplate="@Boilerplate" Elevation="2" Loading="@Boilerplate" Type="article, actions">
        <ChildContent>
            <MRow Class="h-full body_row">
                <MCol Class="d-flex answers_box" Cols="24" Sm="12">
                    <MVirtualScroll Class="answer_list_box" Items="_items" @ref="_vs">
                        <ItemContent>
                            <MListItem Class=" d-flex flex-column">
                                <div class="question_items d-flex flex-row w-100">
                                    <div class="quest_img"><img src="image/user.png" /></div>
                                    <div class="question_txt">@context.question</div>
                                </div>
                                <div class="answers_items d-flex flex-row  w-100">
                                    <div class="answer-img"><svg width="41" height="41" viewBox="0 0 41 41" fill="none" xmlns="http://www.w3.org/2000/svg" stroke-width="1.5" class="h-6 w-6"><path d="M37.5324 16.8707C37.9808 15.5241 38.1363 14.0974 37.9886 12.6859C37.8409 11.2744 37.3934 9.91076 36.676 8.68622C35.6126 6.83404 33.9882 5.3676 32.0373 4.4985C30.0864 3.62941 27.9098 3.40259 25.8215 3.85078C24.8796 2.7893 23.7219 1.94125 22.4257 1.36341C21.1295 0.785575 19.7249 0.491269 18.3058 0.500197C16.1708 0.495044 14.0893 1.16803 12.3614 2.42214C10.6335 3.67624 9.34853 5.44666 8.6917 7.47815C7.30085 7.76286 5.98686 8.3414 4.8377 9.17505C3.68854 10.0087 2.73073 11.0782 2.02839 12.312C0.956464 14.1591 0.498905 16.2988 0.721698 18.4228C0.944492 20.5467 1.83612 22.5449 3.268 24.1293C2.81966 25.4759 2.66413 26.9026 2.81182 28.3141C2.95951 29.7256 3.40701 31.0892 4.12437 32.3138C5.18791 34.1659 6.8123 35.6322 8.76321 36.5013C10.7141 37.3704 12.8907 37.5973 14.9789 37.1492C15.9208 38.2107 17.0786 39.0587 18.3747 39.6366C19.6709 40.2144 21.0755 40.5087 22.4946 40.4998C24.6307 40.5054 26.7133 39.8321 28.4418 38.5772C30.1704 37.3223 31.4556 35.5506 32.1119 33.5179C33.5027 33.2332 34.8167 32.6547 35.9659 31.821C37.115 30.9874 38.0728 29.9178 38.7752 28.684C39.8458 26.8371 40.3023 24.6979 40.0789 22.5748C39.8556 20.4517 38.9639 18.4544 37.5324 16.8707ZM22.4978 37.8849C20.7443 37.8874 19.0459 37.2733 17.6994 36.1501C17.7601 36.117 17.8666 36.0586 17.936 36.0161L25.9004 31.4156C26.1003 31.3019 26.2663 31.137 26.3813 30.9378C26.4964 30.7386 26.5563 30.5124 26.5549 30.2825V19.0542L29.9213 20.998C29.9389 21.0068 29.9541 21.0198 29.9656 21.0359C29.977 21.052 29.9842 21.0707 29.9867 21.0902V30.3889C29.9842 32.375 29.1946 34.2791 27.7909 35.6841C26.3872 37.0892 24.4838 37.8806 22.4978 37.8849ZM6.39227 31.0064C5.51397 29.4888 5.19742 27.7107 5.49804 25.9832C5.55718 26.0187 5.66048 26.0818 5.73461 26.1244L13.699 30.7248C13.8975 30.8408 14.1233 30.902 14.3532 30.902C14.583 30.902 14.8088 30.8408 15.0073 30.7248L24.731 25.1103V28.9979C24.7321 29.0177 24.7283 29.0376 24.7199 29.0556C24.7115 29.0736 24.6988 29.0893 24.6829 29.1012L16.6317 33.7497C14.9096 34.7416 12.8643 35.0097 10.9447 34.4954C9.02506 33.9811 7.38785 32.7263 6.39227 31.0064ZM4.29707 13.6194C5.17156 12.0998 6.55279 10.9364 8.19885 10.3327C8.19885 10.4013 8.19491 10.5228 8.19491 10.6071V19.808C8.19351 20.0378 8.25334 20.2638 8.36823 20.4629C8.48312 20.6619 8.64893 20.8267 8.84863 20.9404L18.5723 26.5542L15.206 28.4979C15.1894 28.5089 15.1703 28.5155 15.1505 28.5173C15.1307 28.5191 15.1107 28.516 15.0924 28.5082L7.04046 23.8557C5.32135 22.8601 4.06716 21.2235 3.55289 19.3046C3.03862 17.3858 3.30624 15.3413 4.29707 13.6194ZM31.955 20.0556L22.2312 14.4411L25.5976 12.4981C25.6142 12.4872 25.6333 12.4805 25.6531 12.4787C25.6729 12.4769 25.6928 12.4801 25.7111 12.4879L33.7631 17.1364C34.9967 17.849 36.0017 18.8982 36.6606 20.1613C37.3194 21.4244 37.6047 22.849 37.4832 24.2684C37.3617 25.6878 36.8382 27.0432 35.9743 28.1759C35.1103 29.3086 33.9415 30.1717 32.6047 30.6641C32.6047 30.5947 32.6047 30.4733 32.6047 30.3889V21.188C32.6066 20.9586 32.5474 20.7328 32.4332 20.5338C32.319 20.3348 32.154 20.1698 31.955 20.0556ZM35.3055 15.0128C35.2464 14.9765 35.1431 14.9142 35.069 14.8717L27.1045 10.2712C26.906 10.1554 26.6803 10.0943 26.4504 10.0943C26.2206 10.0943 25.9948 10.1554 25.7963 10.2712L16.0726 15.8858V11.9982C16.0715 11.9783 16.0753 11.9585 16.0837 11.9405C16.0921 11.9225 16.1048 11.9068 16.1207 11.8949L24.1719 7.25025C25.4053 6.53903 26.8158 6.19376 28.2383 6.25482C29.6608 6.31589 31.0364 6.78077 32.2044 7.59508C33.3723 8.40939 34.2842 9.53945 34.8334 10.8531C35.3826 12.1667 35.5464 13.6095 35.3055 15.0128ZM14.2424 21.9419L10.8752 19.9981C10.8576 19.9893 10.8423 19.9763 10.8309 19.9602C10.8195 19.9441 10.8122 19.9254 10.8098 19.9058V10.6071C10.8107 9.18295 11.2173 7.78848 11.9819 6.58696C12.7466 5.38544 13.8377 4.42659 15.1275 3.82264C16.4173 3.21869 17.8524 2.99464 19.2649 3.1767C20.6775 3.35876 22.0089 3.93941 23.1034 4.85067C23.0427 4.88379 22.937 4.94215 22.8668 4.98473L14.9024 9.58517C14.7025 9.69878 14.5366 9.86356 14.4215 10.0626C14.3065 10.2616 14.2466 10.4877 14.2479 10.7175L14.2424 21.9419ZM16.071 17.9991L20.4018 15.4978L24.7325 17.9975V22.9985L20.4018 25.4983L16.071 22.9985V17.9991Z" fill="currentColor"></path></svg></div>
                                    <div class="answer-txt">@context.answer</div>
                                </div>
                            </MListItem>
                        </ItemContent>
                    </MVirtualScroll>
                </MCol>
            </MRow>
            <div class="footer_box">
                <div class="d-flex justify-content-center">
                    <div class="col-18 col-sm-9">
                        <div class="send_box">
                            <textarea class="send_input" @bind="question"></textarea>
                            <MButton Class="send_btn" Fab Small Color="primary" OnClick="RequestAnswers"><MIcon Size="14">mdi-send-outline</MIcon></MButton>
                        </div>
                    </div>
                </div>

            </div>
        </ChildContent>
    </MSkeletonLoader>
    
    
</MContainer>

Add citation:

@using System.Text;
@using Mode;

For convenience, Code.razoryou Image.razorcan change it like this. Image.razorNeed to pay attention, because the picture link returned by the api, so it needs to be <div class="answer-txt">@context.answer</div>changed <div class="answer-txt"><img src="@context.answer" style="width:512px; height:512px;" /></div>.
We set the image size to 512px, because OpenAI's Image interface returns images of three sizes, which are 256x256, 512x512, and 1024x1024.
The returned image type can be set to urlor b64_json.
For the sake of simplicity, only the creation of pictures is realized, and the picture editing and picture changes are not realized for the time being.

Realize Q&A

Open Index.razor, modified Index.razorcode codesnippet:

@code {
    
    
@code{
    
    

    public string question = "";
    private string answer = "";
    private bool Boilerplate = false;

    private MVirtualScroll<QA> _vs;
    private List<QA> _items = new();

    private async Task RequestAnswers()
    {
    
    
        Boilerplate = true;
        answer = "";
        var api = new OpenAI.OpenAIClient(new OpenAI.OpenAIAuthentication(OpenAIConfig.api_key));

        QA qa = new();
        qa.question = question;

        await api.CompletionsEndpoint.StreamCompletionAsync(result =>
        {
    
    
            foreach (var choice in result.Completions)
            {
    
    
                answer += choice;
            }
        }, question, maxTokens: 500, temperature: 0.5, presencePenalty: 0.1, frequencyPenalty: 0.1, model: OpenAI.Models.Model.Davinci);
        
        answer = answer.ReplaceFirst("\r","").ReplaceFirst("\n","");
        qa.answer = answer;
        _items.Add(qa);
        Boilerplate = false;
    }

}

This completes a question and answer.
Modified Code.razorcode codesnippet:

@code {
    
    

    public string question = "";
    private string answer = "";
    private bool Boilerplate = false;

    private MVirtualScroll<QA> _vs;
    private List<QA> _items = new();

    private async Task RequestAnswers()
    {
    
    
        Boilerplate = true;
        answer = "";
        var api = new OpenAI.OpenAIClient(new OpenAI.OpenAIAuthentication(OpenAIConfig.api_key));
        QA qa = new();
        qa.question = question;
        OpenAI.Models.Model model = new OpenAI.Models.Model("code-davinci-002");
        await api.CompletionsEndpoint.StreamCompletionAsync(result =>
        {
    
    
            foreach (var choice in result.Completions)
            {
    
    
                answer += choice;
            }
        }, question, maxTokens: 500, temperature: 0.5, presencePenalty: 0.1, frequencyPenalty: 0.1, model: model);
        
        answer = answer.ReplaceFirst("\r","").ReplaceFirst("\n","");
        qa.answer = answer;
        _items.Add(qa);
        Boilerplate = false;
    }

}

OpenAI's Codex model is very powerful, and it is by no means as simple as the above. The Codex model can generate corresponding codes based on comments and descriptions. Existing code can also be automatically completed. The example simply implements the basic functions of the Codex model, and the more advanced gameplay is not implemented yet!

Modified code snippet Image.razor:code

@code {
    
    

    public string question = "";
    private string answer = "";
    private bool Boilerplate = false;

    private MVirtualScroll<QA> _vs;
    private List<QA> _items = new();

    private async Task RequestAnswers()
    {
    
    
        Boilerplate = true;
        answer = "";
        var api = new OpenAI.OpenAIClient(new OpenAI.OpenAIAuthentication(OpenAIConfig.api_key));

        QA qa = new();
        qa.question = question;

        var results = await api.ImagesEndPoint.GenerateImageAsync(question, 1, OpenAI.Images.ImageSize.Small);
        foreach(var item in results)
        {
    
    
            answer += item;
        }

        answer = answer.ReplaceFirst("\r", "").ReplaceFirst("\n", "");
        qa.answer = answer;
        _items.Add(qa);
        Boilerplate = false;
    }

}

OpenAI's Image model is also very powerful and can be created, edited and changed. The example only implements the creation of pictures. More advanced gameplay is not yet implemented! You can look at the example of OpenAI, the function of image editing and changing.
Picture editing function:

Picture changing function:

Full file download

Click on the official account card below, follow me, and reply 1003to download!