fix: type error

Merge

README.md

@@ -1,16 +1,17 @@
 <div align="center">
 
-<a href='#企业版'>
+<a href='https://nextchat.dev/chat'>
-  <img src="./docs/images/ent.svg" alt="icon"/>
+  <img src="https://github.com/user-attachments/assets/287c510f-f508-478e-ade3-54d30453dc18" width="1000" alt="icon"/>
 </a>
 
+
 <h1 align="center">NextChat (ChatGPT Next Web)</h1>
 
 English / [简体中文](./README_CN.md)
 
-One-Click to get a well-designed cross-platform ChatGPT web UI, with GPT3, GPT4 & Gemini Pro support.
+One-Click to get a well-designed cross-platform ChatGPT web UI, with Claude, GPT4 & Gemini Pro support.
 
-一键免费部署你的跨平台私人 ChatGPT 应用, 支持 GPT3, GPT4 & Gemini Pro 模型。
+一键免费部署你的跨平台私人 ChatGPT 应用, 支持 Claude, GPT4 & Gemini Pro 模型。
 
 [![Saas][Saas-image]][saas-url]
 [![Web][Web-image]][web-url]
@@ -31,7 +32,7 @@ One-Click to get a well-designed cross-platform ChatGPT web UI, with GPT3, GPT4
 [MacOS-image]: https://img.shields.io/badge/-MacOS-black?logo=apple
 [Linux-image]: https://img.shields.io/badge/-Linux-333?logo=ubuntu
 
-[<img src="https://vercel.com/button" alt="Deploy on Vercel" height="30">](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FChatGPTNextWeb%2FChatGPT-Next-Web&env=OPENAI_API_KEY&env=CODE&project-name=nextchat&repository-name=NextChat) [<img src="https://zeabur.com/button.svg" alt="Deploy on Zeabur" height="30">](https://zeabur.com/templates/ZBUEFA)  [<img src="https://gitpod.io/button/open-in-gitpod.svg" alt="Open in Gitpod" height="30">](https://gitpod.io/#https://github.com/Yidadaa/ChatGPT-Next-Web) [<img src="https://img.shields.io/badge/BT_Deploy-Install-20a53a" alt="BT Deply Install" height="30">](https://www.bt.cn/new/download.html) [<img src="https://svgshare.com/i/1AVg.svg" alt="Deploy to Alibaba Cloud" height="30">](https://computenest.aliyun.com/market/service-f1c9b75e59814dc49d52)
+[<img src="https://vercel.com/button" alt="Deploy on Vercel" height="30">](https://vercel.com/new/clone?repository-url=https://github.com/Dogtiti/ChatGPT-Next-Web-EarlyBird&env=OPENAI_API_KEY&env=CODE&project-name=nextchat-earlyBird&repository-name=NextChat-EarlyBird) [<img src="https://zeabur.com/button.svg" alt="Deploy on Zeabur" height="30">](https://zeabur.com/templates/ZBUEFA)  [<img src="https://gitpod.io/button/open-in-gitpod.svg" alt="Open in Gitpod" height="30">](https://gitpod.io/#https://github.com/Yidadaa/ChatGPT-Next-Web) [<img src="https://img.shields.io/badge/BT_Deploy-Install-20a53a" alt="BT Deply Install" height="30">](https://www.bt.cn/new/download.html)
 
 [<img src="https://github.com/user-attachments/assets/903482d4-3e87-4134-9af1-f2588fa90659" height="60" width="288" >](https://monica.im/?utm=nxcrp)
 
@@ -96,10 +97,11 @@ For enterprise inquiries, please contact: **business@nextchat.dev**
 - [x] Artifacts: Easily preview, copy and share generated content/webpages through a separate window [#5092](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/pull/5092)
 - [x] Plugins: support network search, calculator, any other apis etc. [#165](https://github.com/Yidadaa/ChatGPT-Next-Web/issues/165) [#5353](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5353)
   - [x] network search, calculator, any other apis etc. [#165](https://github.com/Yidadaa/ChatGPT-Next-Web/issues/165) [#5353](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5353)
+- [x] Supports Realtime Chat [#5672](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5672)
 - [ ] local knowledge base
 
 ## What's New
-
+- 🚀 v2.15.8 Now supports Realtime Chat [#5672](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5672)
 - 🚀 v2.15.4 The Application supports using Tauri fetch LLM API, MORE SECURITY! [#5379](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5379)
 - 🚀 v2.15.0 Now supports Plugins! Read this: [NextChat-Awesome-Plugins](https://github.com/ChatGPTNextWeb/NextChat-Awesome-Plugins)
 - 🚀 v2.14.0 Now supports  Artifacts & SD 
@@ -134,10 +136,11 @@ For enterprise inquiries, please contact: **business@nextchat.dev**
 - [x] Artifacts: 通过独立窗口，轻松预览、复制和分享生成的内容/可交互网页 [#5092](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/pull/5092)
 - [x] 插件机制，支持`联网搜索`、`计算器`、调用其他平台 api [#165](https://github.com/Yidadaa/ChatGPT-Next-Web/issues/165) [#5353](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5353)
    - [x] 支持联网搜索、计算器、调用其他平台 api [#165](https://github.com/Yidadaa/ChatGPT-Next-Web/issues/165) [#5353](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5353)
+ - [x] 支持 Realtime Chat [#5672](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5672)
  - [ ] 本地知识库
 
 ## 最新动态
-
+- 🚀 v2.15.8 现在支持Realtime Chat [#5672](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5672)
 - 🚀 v2.15.4 客户端支持Tauri本地直接调用大模型API，更安全！[#5379](https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/issues/5379)
 - 🚀 v2.15.0 现在支持插件功能了！了解更多：[NextChat-Awesome-Plugins](https://github.com/ChatGPTNextWeb/NextChat-Awesome-Plugins)
 - 🚀 v2.14.0 现在支持 Artifacts & SD 了。
@@ -353,6 +356,13 @@ For ByteDance: use `modelName@bytedance=deploymentName` to customize model name
 
 Change default model
 
+### `VISION_MODELS` (optional)
+
+> Default: Empty
+> Example: `gpt-4-vision,claude-3-opus,my-custom-model` means add vision capabilities to these models in addition to the default pattern matches (which detect models containing keywords like "vision", "claude-3", "gemini-1.5", etc).
+
+Add additional models to have vision capabilities, beyond the default pattern matching. Multiple models should be separated by commas.
+
 ### `WHITE_WEBDAV_ENDPOINTS` (optional)
 
 You can use this option if you want to increase the number of webdav service addresses you are allowed to access, as required by the format：
@@ -467,7 +477,7 @@ If you want to add a new translation, read this [document](./docs/translation.md
 
 ## Donation
 
-[Buy Me a Coffee](https://www.buymeacoffee.com/yidadaa)
+[Buy Me a Coffee](https://1kafei.com/dogtiti)
 
 ## Special Thanks
 

README_CN.md

@@ -33,7 +33,7 @@
 
 1. 准备好你的 [OpenAI API Key](https://platform.openai.com/account/api-keys);
 2. 点击右侧按钮开始部署：
-   [![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FYidadaa%2FChatGPT-Next-Web&env=OPENAI_API_KEY&env=CODE&env=GOOGLE_API_KEY&project-name=chatgpt-next-web&repository-name=ChatGPT-Next-Web)，直接使用 Github 账号登录即可，记得在环境变量页填入 API Key 和[页面访问密码](#配置页面访问密码) CODE；
+   [![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/Dogtiti/ChatGPT-Next-Web-EarlyBird&env=OPENAI_API_KEY&env=CODE&project-name=nextchat-earlyBird&repository-name=NextChat-EarlyBird)，直接使用 Github 账号登录即可，记得在环境变量页填入 API Key 和[页面访问密码](#配置页面访问密码) CODE；
 3. 部署完毕后，即可开始使用；
 4. （可选）[绑定自定义域名](https://vercel.com/docs/concepts/projects/domains/add-a-domain)：Vercel 分配的域名 DNS 在某些区域被污染了，绑定自定义域名即可直连。
 
@@ -235,6 +235,13 @@ ChatGLM Api Url.
 
 更改默认模型
 
+### `VISION_MODELS` (可选)
+
+> 默认值：空
+> 示例：`gpt-4-vision,claude-3-opus,my-custom-model` 表示为这些模型添加视觉能力，作为对默认模式匹配的补充（默认会检测包含"vision"、"claude-3"、"gemini-1.5"等关键词的模型）。
+
+在默认模式匹配之外，添加更多具有视觉能力的模型。多个模型用逗号分隔。
+
 ### `DEFAULT_INPUT_TEMPLATE` （可选）
 
 自定义默认的 template，用于初始化『设置』中的『用户输入预处理』配置项

README_JA.md

@@ -30,7 +30,7 @@
 
 1. [OpenAI API Key](https://platform.openai.com/account/api-keys)を準備する;
 2. 右側のボタンをクリックしてデプロイを開始：
-   [![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2FYidadaa%2FChatGPT-Next-Web&env=OPENAI_API_KEY&env=CODE&env=GOOGLE_API_KEY&project-name=chatgpt-next-web&repository-name=ChatGPT-Next-Web) 、GitHubアカウントで直接ログインし、環境変数ページにAPI Keyと[ページアクセスパスワード](#設定ページアクセスパスワード) CODEを入力してください;
+   [![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/Dogtiti/ChatGPT-Next-Web-EarlyBird&env=OPENAI_API_KEY&env=CODE&project-name=nextchat-earlyBird&repository-name=NextChat-EarlyBird) 、GitHubアカウントで直接ログインし、環境変数ページにAPI Keyと[ページアクセスパスワード](#設定ページアクセスパスワード) CODEを入力してください;
 3. デプロイが完了したら、すぐに使用を開始できます;
 4. （オプション）[カスタムドメインをバインド](https://vercel.com/docs/concepts/projects/domains/add-a-domain)：Vercelが割り当てたドメインDNSは一部の地域で汚染されているため、カスタムドメインをバインドすると直接接続できます。
 
@@ -217,6 +217,13 @@ ByteDance モードでは、`modelName@bytedance=deploymentName` 形式でモデ
 
 デフォルトのモデルを変更します。
 
+### `VISION_MODELS` (オプション)
+
+> デフォルト：空
+> 例：`gpt-4-vision,claude-3-opus,my-custom-model` は、これらのモデルにビジョン機能を追加します。これはデフォルトのパターンマッチング（"vision"、"claude-3"、"gemini-1.5"などのキーワードを含むモデルを検出）に加えて適用されます。
+
+デフォルトのパターンマッチングに加えて、追加のモデルにビジョン機能を付与します。複数のモデルはカンマで区切ります。
+
 ### `DEFAULT_INPUT_TEMPLATE` （オプション）
 
 『設定』の『ユーザー入力前処理』の初期設定に使用するテンプレートをカスタマイズします。

app/api/common.ts

@@ -1,8 +1,8 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getServerSideConfig } from "../config/server";
 import { OPENAI_BASE_URL, ServiceProvider } from "../constant";
-import { isModelAvailableInServer } from "../utils/model";
 import { cloudflareAIGatewayUrl } from "../utils/cloudflare";
+import { getModelProvider, isModelAvailableInServer } from "../utils/model";
 
 const serverConfig = getServerSideConfig();
 
@@ -71,7 +71,7 @@ export async function requestOpenai(req: NextRequest) {
         .filter((v) => !!v && !v.startsWith("-") && v.includes(modelName))
         .forEach((m) => {
           const [fullName, displayName] = m.split("=");
-          const [_, providerName] = fullName.split("@");
+          const [_, providerName] = getModelProvider(fullName);
           if (providerName === "azure" && !displayName) {
             const [_, deployId] = (serverConfig?.azureUrl ?? "").split(
               "deployments/",

app/api/openai.ts

@@ -14,7 +14,7 @@ function getModels(remoteModelRes: OpenAIListModelResponse) {
   if (config.disableGPT4) {
     remoteModelRes.data = remoteModelRes.data.filter(
       (m) =>
-        !(m.id.startsWith("gpt-4") || m.id.startsWith("chatgpt-4o")) ||
+        !(m.id.startsWith("gpt-4") || m.id.startsWith("chatgpt-4o") || m.id.startsWith("o1")) ||
         m.id.startsWith("gpt-4o-mini"),
     );
   }

app/api/proxy.ts

@@ -1,4 +1,5 @@
 import { NextRequest, NextResponse } from "next/server";
+import { getServerSideConfig } from "@/app/config/server";
 
 export async function handle(
   req: NextRequest,
@@ -9,6 +10,7 @@ export async function handle(
   if (req.method === "OPTIONS") {
     return NextResponse.json({ body: "OK" }, { status: 200 });
   }
+  const serverConfig = getServerSideConfig();
 
   // remove path params from searchParams
   req.nextUrl.searchParams.delete("path");
@@ -31,6 +33,18 @@ export async function handle(
       return true;
     }),
   );
+  // if dalle3 use openai api key
+    const baseUrl = req.headers.get("x-base-url");
+    if (baseUrl?.includes("api.openai.com")) {
+      if (!serverConfig.apiKey) {
+        return NextResponse.json(
+          { error: "OpenAI API key not configured" },
+          { status: 500 },
+        );
+      }
+      headers.set("Authorization", `Bearer ${serverConfig.apiKey}`);
+    }
+
   const controller = new AbortController();
   const fetchOptions: RequestInit = {
     headers,

app/client/platforms/glm.ts

@@ -25,12 +25,103 @@ import { getMessageTextContent } from "@/app/utils";
 import { RequestPayload } from "./openai";
 import { fetch } from "@/app/utils/stream";
 
+interface BasePayload {
+  model: string;
+}
+
+interface ChatPayload extends BasePayload {
+  messages: ChatOptions["messages"];
+  stream?: boolean;
+  temperature?: number;
+  presence_penalty?: number;
+  frequency_penalty?: number;
+  top_p?: number;
+}
+
+interface ImageGenerationPayload extends BasePayload {
+  prompt: string;
+  size?: string;
+  user_id?: string;
+}
+
+interface VideoGenerationPayload extends BasePayload {
+  prompt: string;
+  duration?: number;
+  resolution?: string;
+  user_id?: string;
+}
+
+type ModelType = "chat" | "image" | "video";
+
 export class ChatGLMApi implements LLMApi {
   private disableListModels = true;
 
+  private getModelType(model: string): ModelType {
+    if (model.startsWith("cogview-")) return "image";
+    if (model.startsWith("cogvideo-")) return "video";
+    return "chat";
+  }
+
+  private getModelPath(type: ModelType): string {
+    switch (type) {
+      case "image":
+        return ChatGLM.ImagePath;
+      case "video":
+        return ChatGLM.VideoPath;
+      default:
+        return ChatGLM.ChatPath;
+    }
+  }
+
+  private createPayload(
+    messages: ChatOptions["messages"],
+    modelConfig: any,
+    options: ChatOptions,
+  ): BasePayload {
+    const modelType = this.getModelType(modelConfig.model);
+    const lastMessage = messages[messages.length - 1];
+    const prompt =
+      typeof lastMessage.content === "string"
+        ? lastMessage.content
+        : lastMessage.content.map((c) => c.text).join("\n");
+
+    switch (modelType) {
+      case "image":
+        return {
+          model: modelConfig.model,
+          prompt,
+          size: options.config.size,
+        } as ImageGenerationPayload;
+      default:
+        return {
+          messages,
+          stream: options.config.stream,
+          model: modelConfig.model,
+          temperature: modelConfig.temperature,
+          presence_penalty: modelConfig.presence_penalty,
+          frequency_penalty: modelConfig.frequency_penalty,
+          top_p: modelConfig.top_p,
+        } as ChatPayload;
+    }
+  }
+
+  private parseResponse(modelType: ModelType, json: any): string {
+    switch (modelType) {
+      case "image": {
+        const imageUrl = json.data?.[0]?.url;
+        return imageUrl ? `![Generated Image](${imageUrl})` : "";
+      }
+      case "video": {
+        const videoUrl = json.data?.[0]?.url;
+        return videoUrl ? `<video controls src="${videoUrl}"></video>` : "";
+      }
+      default:
+        return this.extractMessage(json);
+    }
+  }
+
   path(path: string): string {
     const accessStore = useAccessStore.getState();
-
     let baseUrl = "";
 
     if (accessStore.useCustomConfig) {
@@ -51,7 +142,6 @@ export class ChatGLMApi implements LLMApi {
     }
 
     console.log("[Proxy Endpoint] ", baseUrl, path);
-
     return [baseUrl, path].join("/");
   }
 
@@ -79,24 +169,16 @@ export class ChatGLMApi implements LLMApi {
       },
     };
 
-    const requestPayload: RequestPayload = {
+    const modelType = this.getModelType(modelConfig.model);
-      messages,
+    const requestPayload = this.createPayload(messages, modelConfig, options);
-      stream: options.config.stream,
+    const path = this.path(this.getModelPath(modelType));
-      model: modelConfig.model,
-      temperature: modelConfig.temperature,
-      presence_penalty: modelConfig.presence_penalty,
-      frequency_penalty: modelConfig.frequency_penalty,
-      top_p: modelConfig.top_p,
-    };
 
-    console.log("[Request] glm payload: ", requestPayload);
+    console.log(`[Request] glm ${modelType} payload: `, requestPayload);
 
-    const shouldStream = !!options.config.stream;
     const controller = new AbortController();
     options.onController?.(controller);
 
     try {
-      const chatPath = this.path(ChatGLM.ChatPath);
       const chatPayload = {
         method: "POST",
         body: JSON.stringify(requestPayload),
@@ -104,12 +186,23 @@ export class ChatGLMApi implements LLMApi {
         headers: getHeaders(),
       };
 
-      // make a fetch request
       const requestTimeoutId = setTimeout(
         () => controller.abort(),
         REQUEST_TIMEOUT_MS,
       );
 
+      if (modelType === "image" || modelType === "video") {
+        const res = await fetch(path, chatPayload);
+        clearTimeout(requestTimeoutId);
+
+        const resJson = await res.json();
+        console.log(`[Response] glm ${modelType}:`, resJson);
+        const message = this.parseResponse(modelType, resJson);
+        options.onFinish(message, res);
+        return;
+      }
+
+      const shouldStream = !!options.config.stream;
       if (shouldStream) {
         const [tools, funcs] = usePluginStore
           .getState()
@@ -117,7 +210,7 @@ export class ChatGLMApi implements LLMApi {
             useChatStore.getState().currentSession().mask?.plugin || [],
           );
         return stream(
-          chatPath,
+          path,
           requestPayload,
           getHeaders(),
           tools as any,
@@ -125,7 +218,6 @@ export class ChatGLMApi implements LLMApi {
           controller,
           // parseSSE
           (text: string, runTools: ChatMessageTool[]) => {
-            // console.log("parseSSE", text, runTools);
             const json = JSON.parse(text);
             const choices = json.choices as Array<{
               delta: {
@@ -154,7 +246,7 @@ export class ChatGLMApi implements LLMApi {
             }
             return choices[0]?.delta?.content;
           },
-          // processToolMessage, include tool_calls message and tool call results
+          // processToolMessage
           (
             requestPayload: RequestPayload,
             toolCallMessage: any,
@@ -172,7 +264,7 @@ export class ChatGLMApi implements LLMApi {
           options,
         );
       } else {
-        const res = await fetch(chatPath, chatPayload);
+        const res = await fetch(path, chatPayload);
         clearTimeout(requestTimeoutId);
 
         const resJson = await res.json();
@@ -184,6 +276,7 @@ export class ChatGLMApi implements LLMApi {
       options.onError?.(e as Error);
     }
   }
+
   async usage() {
     return {
       used: 0,

app/client/platforms/google.ts

@@ -29,7 +29,7 @@ import { RequestPayload } from "./openai";
 import { fetch } from "@/app/utils/stream";
 
 export class GeminiProApi implements LLMApi {
-  path(path: string): string {
+  path(path: string, shouldStream = false): string {
     const accessStore = useAccessStore.getState();
 
     let baseUrl = "";
@@ -51,8 +51,10 @@ export class GeminiProApi implements LLMApi {
     console.log("[Proxy Endpoint] ", baseUrl, path);
 
     let chatPath = [baseUrl, path].join("/");
+    if (shouldStream) {
+      chatPath += chatPath.includes("?") ? "&alt=sse" : "?alt=sse";
+    }
 
-    chatPath += chatPath.includes("?") ? "&alt=sse" : "?alt=sse";
     return chatPath;
   }
   extractMessage(res: any) {
@@ -60,6 +62,7 @@ export class GeminiProApi implements LLMApi {
 
     return (
       res?.candidates?.at(0)?.content?.parts.at(0)?.text ||
+      res?.at(0)?.candidates?.at(0)?.content?.parts.at(0)?.text ||
       res?.error?.message ||
       ""
     );
@@ -166,7 +169,10 @@ export class GeminiProApi implements LLMApi {
     options.onController?.(controller);
     try {
       // https://github.com/google-gemini/cookbook/blob/main/quickstarts/rest/Streaming_REST.ipynb
-      const chatPath = this.path(Google.ChatPath(modelConfig.model));
+      const chatPath = this.path(
+        Google.ChatPath(modelConfig.model),
+        shouldStream,
+      );
 
       const chatPayload = {
         method: "POST",

app/client/platforms/openai.ts

@@ -24,7 +24,7 @@ import {
   stream,
 } from "@/app/utils/chat";
 import { cloudflareAIGatewayUrl } from "@/app/utils/cloudflare";
-import { DalleSize, DalleQuality, DalleStyle } from "@/app/typing";
+import { ModelSize, DalleQuality, DalleStyle } from "@/app/typing";
 
 import {
   ChatOptions,
@@ -65,6 +65,7 @@ export interface RequestPayload {
   frequency_penalty: number;
   top_p: number;
   max_tokens?: number;
+  max_completion_tokens?: number;
 }
 
 export interface DalleRequestPayload {
@@ -72,7 +73,7 @@ export interface DalleRequestPayload {
   prompt: string;
   response_format: "url" | "b64_json";
   n: number;
-  size: DalleSize;
+  size: ModelSize;
   quality: DalleQuality;
   style: DalleStyle;
 }
@@ -223,7 +224,7 @@ export class ChatGPTApi implements LLMApi {
       // O1 not support image, tools (plugin in ChatGPTNextWeb) and system, stream, logprobs, temperature, top_p, n, presence_penalty, frequency_penalty yet.
       requestPayload = {
         messages,
-        stream: !isO1 ? options.config.stream : false,
+        stream: options.config.stream,
         model: modelConfig.model,
         temperature: !isO1 ? modelConfig.temperature : 1,
         presence_penalty: !isO1 ? modelConfig.presence_penalty : 0,
@@ -233,6 +234,11 @@ export class ChatGPTApi implements LLMApi {
         // Please do not ask me why not send max_tokens, no reason, this param is just shit, I dont want to explain anymore.
       };
 
+      // O1 使用 max_completion_tokens 控制token数 (https://platform.openai.com/docs/guides/reasoning#controlling-costs)
+      if (isO1) {
+        requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
+      }
+
       // add max_tokens to vision model
       if (visionModel) {
         requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
@@ -241,7 +247,7 @@ export class ChatGPTApi implements LLMApi {
 
     console.log("[Request] openai payload: ", requestPayload);
 
-    const shouldStream = !isDalle3 && !!options.config.stream && !isO1;
+    const shouldStream = !isDalle3 && !!options.config.stream;
     const controller = new AbortController();
     options.onController?.(controller);
 

app/components/auth.tsx

@@ -18,6 +18,8 @@ import {
   trackSettingsPageGuideToCPaymentClick,
   trackAuthorizationPageButtonToCPaymentClick,
 } from "../utils/auth-settings-events";
+import clsx from "clsx";
+
 const storage = safeLocalStorage();
 
 export function AuthPage() {
@@ -54,7 +56,7 @@ export function AuthPage() {
           onClick={() => navigate(Path.Home)}
         ></IconButton>
       </div>
-      <div className={`no-dark ${styles["auth-logo"]}`}>
+      <div className={clsx("no-dark", styles["auth-logo"])}>
         <BotIcon />
       </div>
 
@@ -163,7 +165,7 @@ function TopBanner() {
       onMouseEnter={handleMouseEnter}
       onMouseLeave={handleMouseLeave}
     >
-      <div className={`${styles["top-banner-inner"]} no-dark`}>
+      <div className={clsx(styles["top-banner-inner"], "no-dark")}>
         <Logo className={styles["top-banner-logo"]}></Logo>
         <span>
           {Locale.Auth.TopTips}

app/components/button.tsx

@@ -2,6 +2,7 @@ import * as React from "react";
 
 import styles from "./button.module.scss";
 import { CSSProperties } from "react";
+import clsx from "clsx";
 
 export type ButtonType = "primary" | "danger" | null;
 
@@ -22,12 +23,16 @@ export function IconButton(props: {
 }) {
   return (
     <button
-      className={
+      className={clsx(
-        styles["icon-button"] +
+        "clickable",
-        ` ${props.bordered && styles.border} ${props.shadow && styles.shadow} ${
+        styles["icon-button"],
-          props.className ?? ""
+        {
-        } clickable ${styles[props.type ?? ""]}`
+          [styles.border]: props.bordered,
-      }
+          [styles.shadow]: props.shadow,
+        },
+        styles[props.type ?? ""],
+        props.className,
+      )}
       onClick={props.onClick}
       title={props.title}
       disabled={props.disabled}
@@ -40,10 +45,9 @@ export function IconButton(props: {
       {props.icon && (
         <div
           aria-label={props.text || props.title}
-          className={
+          className={clsx(styles["icon-button-icon"], {
-            styles["icon-button-icon"] +
+            "no-dark": props.type === "primary",
-            ` ${props.type === "primary" && "no-dark"}`
+          })}
-          }
         >
           {props.icon}
         </div>

app/components/chat-list.tsx

@@ -18,6 +18,7 @@ import { Mask } from "../store/mask";
 import { useRef, useEffect } from "react";
 import { showConfirm } from "./ui-lib";
 import { useMobileScreen } from "../utils";
+import clsx from "clsx";
 
 export function ChatItem(props: {
   onClick?: () => void;
@@ -45,11 +46,11 @@ export function ChatItem(props: {
     <Draggable draggableId={`${props.id}`} index={props.index}>
       {(provided) => (
         <div
-          className={`${styles["chat-item"]} ${
+          className={clsx(styles["chat-item"], {
-            props.selected &&
+            [styles["chat-item-selected"]]:
-            (currentPath === Path.Chat || currentPath === Path.Home) &&
+              props.selected &&
-            styles["chat-item-selected"]
+              (currentPath === Path.Chat || currentPath === Path.Home),
-          }`}
+          })}
           onClick={props.onClick}
           ref={(ele) => {
             draggableRef.current = ele;
@@ -63,7 +64,7 @@ export function ChatItem(props: {
         >
           {props.narrow ? (
             <div className={styles["chat-item-narrow"]}>
-              <div className={styles["chat-item-avatar"] + " no-dark"}>
+              <div className={clsx(styles["chat-item-avatar"], "no-dark")}>
                 <MaskAvatar
                   avatar={props.mask.avatar}
                   model={props.mask.modelConfig.model}

app/components/chat.module.scss

@@ -45,6 +45,14 @@
 .chat-input-actions {
   display: flex;
   flex-wrap: wrap;
+  justify-content: space-between;
+  gap: 5px;
+
+  &-end {
+    display: flex;
+    margin-left: auto;
+    gap: 5px;
+  }
 
   .chat-input-action {
     display: inline-flex;
@@ -62,10 +70,6 @@
     width: var(--icon-width);
     overflow: hidden;
 
-    &:not(:last-child) {
-      margin-right: 5px;
-    }
-
     .text {
       white-space: nowrap;
       padding-left: 5px;
@@ -231,10 +235,12 @@
 
   animation: slide-in ease 0.3s;
 
-  $linear: linear-gradient(to right,
+  $linear: linear-gradient(
-      rgba(0, 0, 0, 0),
+    to right,
-      rgba(0, 0, 0, 1),
+    rgba(0, 0, 0, 0),
-      rgba(0, 0, 0, 0));
+    rgba(0, 0, 0, 1),
+    rgba(0, 0, 0, 0)
+  );
   mask-image: $linear;
 
   @mixin show {
@@ -373,7 +379,7 @@
   }
 }
 
-.chat-message-user>.chat-message-container {
+.chat-message-user > .chat-message-container {
   align-items: flex-end;
 }
 
@@ -443,6 +449,25 @@
   transition: all ease 0.3s;
 }
 
+.chat-message-audio {
+  display: flex;
+  align-items: center;
+  justify-content: space-between;
+  border-radius: 10px;
+  background-color: rgba(0, 0, 0, 0.05);
+  border: var(--border-in-light);
+  position: relative;
+  transition: all ease 0.3s;
+  margin-top: 10px;
+  font-size: 14px;
+  user-select: text;
+  word-break: break-word;
+  box-sizing: border-box;
+  audio {
+    height: 30px; /* 调整高度 */
+  }
+}
+
 .chat-message-item-image {
   width: 100%;
   margin-top: 10px;
@@ -471,23 +496,27 @@
   border: rgba($color: #888, $alpha: 0.2) 1px solid;
 }
 
-
 @media only screen and (max-width: 600px) {
-  $calc-image-width: calc(100vw/3*2/var(--image-count));
+  $calc-image-width: calc(100vw / 3 * 2 / var(--image-count));
 
   .chat-message-item-image-multi {
     width: $calc-image-width;
     height: $calc-image-width;
   }
-  
+
   .chat-message-item-image {
-    max-width: calc(100vw/3*2);
+    max-width: calc(100vw / 3 * 2);
   }
 }
 
 @media screen and (min-width: 600px) {
-  $max-image-width: calc(calc(1200px - var(--sidebar-width))/3*2/var(--image-count));
+  $max-image-width: calc(
-  $image-width: calc(calc(var(--window-width) - var(--sidebar-width))/3*2/var(--image-count));
+    calc(1200px - var(--sidebar-width)) / 3 * 2 / var(--image-count)
+  );
+  $image-width: calc(
+    calc(var(--window-width) - var(--sidebar-width)) / 3 * 2 /
+      var(--image-count)
+  );
 
   .chat-message-item-image-multi {
     width: $image-width;
@@ -497,7 +526,7 @@
   }
 
   .chat-message-item-image {
-    max-width: calc(calc(1200px - var(--sidebar-width))/3*2);
+    max-width: calc(calc(1200px - var(--sidebar-width)) / 3 * 2);
   }
 }
 
@@ -515,7 +544,7 @@
   z-index: 1;
 }
 
-.chat-message-user>.chat-message-container>.chat-message-item {
+.chat-message-user > .chat-message-container > .chat-message-item {
   background-color: var(--second);
 
   &:hover {
@@ -626,7 +655,8 @@
   min-height: 68px;
 }
 
-.chat-input:focus {}
+.chat-input:focus {
+}
 
 .chat-input-send {
   background-color: var(--primary);
@@ -693,4 +723,31 @@
 .shortcut-key span {
   font-size: 12px;
   color: var(--black);
-}
+}
+
+.chat-main {
+  display: flex;
+  height: 100%;
+  width: 100%;
+  position: relative;
+  overflow: hidden;
+  .chat-body-container {
+    height: 100%;
+    display: flex;
+    flex-direction: column;
+    flex: 1;
+    width: 100%;
+  }
+  .chat-side-panel {
+    position: absolute;
+    inset: 0;
+    background: var(--white);
+    overflow: hidden;
+    z-index: 10;
+    transform: translateX(100%);
+    transition: all ease 0.3s;
+    &-show {
+      transform: translateX(0);
+    }
+  }
+}

app/components/emoji.tsx

@@ -37,7 +37,8 @@ export function Avatar(props: { model?: ModelType; avatar?: string }) {
     return (
       <div className="no-dark">
         {props.model?.startsWith("gpt-4") ||
-        props.model?.startsWith("chatgpt-4o") ? (
+        props.model?.startsWith("chatgpt-4o") ||
+        props.model?.startsWith("o1") ? (
           <BlackBotIcon className="user-avatar" />
         ) : (
           <BotIcon className="user-avatar" />

app/components/exporter.tsx

@@ -40,6 +40,7 @@ import { EXPORT_MESSAGE_CLASS_NAME } from "../constant";
 import { getClientConfig } from "../config/client";
 import { type ClientApi, getClientApi } from "../client/api";
 import { getMessageTextContent } from "../utils";
+import clsx from "clsx";
 
 const Markdown = dynamic(async () => (await import("./markdown")).Markdown, {
   loading: () => <LoadingIcon />,
@@ -118,9 +119,10 @@ function Steps<
           return (
             <div
               key={i}
-              className={`${styles["step"]} ${
+              className={clsx("clickable", styles["step"], {
-                styles[i <= props.index ? "step-finished" : ""]
+                [styles["step-finished"]]: i <= props.index,
-              } ${i === props.index && styles["step-current"]} clickable`}
+                [styles["step-current"]]: i === props.index,
+              })}
               onClick={() => {
                 props.onStepChange?.(i);
               }}
@@ -525,11 +527,11 @@ export function ImagePreviewer(props: {
         messages={props.messages}
       />
       <div
-        className={`${styles["preview-body"]} ${styles["default-theme"]}`}
+        className={clsx(styles["preview-body"], styles["default-theme"])}
         ref={previewRef}
       >
         <div className={styles["chat-info"]}>
-          <div className={styles["logo"] + " no-dark"}>
+          <div className={clsx(styles["logo"], "no-dark")}>
             <NextImage
               src={ChatGptIcon.src}
               alt="logo"
@@ -570,7 +572,7 @@ export function ImagePreviewer(props: {
         {props.messages.map((m, i) => {
           return (
             <div
-              className={styles["message"] + " " + styles["message-" + m.role]}
+              className={clsx(styles["message"], styles["message-" + m.role])}
               key={i}
             >
               <div className={styles["avatar"]}>

app/components/home.tsx

@@ -3,7 +3,6 @@
 require("../polyfill");
 
 import { useState, useEffect } from "react";
-
 import styles from "./home.module.scss";
 
 import BotIcon from "../icons/bot.svg";
@@ -29,10 +28,11 @@ import { AuthPage } from "./auth";
 import { getClientConfig } from "../config/client";
 import { type ClientApi, getClientApi } from "../client/api";
 import { useAccessStore } from "../store";
+import clsx from "clsx";
 
 export function Loading(props: { noLogo?: boolean }) {
   return (
-    <div className={styles["loading-content"] + " no-dark"}>
+    <div className={clsx("no-dark", styles["loading-content"])}>
       {!props.noLogo && <BotIcon />}
       <LoadingIcon />
     </div>
@@ -179,7 +179,11 @@ function Screen() {
     if (isSdNew) return <Sd />;
     return (
       <>
-        <SideBar className={isHome ? styles["sidebar-show"] : ""} />
+        <SideBar
+          className={clsx({
+            [styles["sidebar-show"]]: isHome,
+          })}
+        />
         <WindowContent>
           <Routes>
             <Route path={Path.Home} element={<Chat />} />
@@ -197,9 +201,10 @@ function Screen() {
 
   return (
     <div
-      className={`${styles.container} ${
+      className={clsx(styles.container, {
-        shouldTightBorder ? styles["tight-container"] : styles.container
+        [styles["tight-container"]]: shouldTightBorder,
-      } ${getLang() === "ar" ? styles["rtl-screen"] : ""}`}
+        [styles["rtl-screen"]]: getLang() === "ar",
+      })}
     >
       {renderContent()}
     </div>

app/components/input-range.tsx

@@ -1,5 +1,6 @@
 import * as React from "react";
 import styles from "./input-range.module.scss";
+import clsx from "clsx";
 
 interface InputRangeProps {
   onChange: React.ChangeEventHandler<HTMLInputElement>;
@@ -23,7 +24,7 @@ export function InputRange({
   aria,
 }: InputRangeProps) {
   return (
-    <div className={styles["input-range"] + ` ${className ?? ""}`}>
+    <div className={clsx(styles["input-range"], className)}>
       {title || value}
       <input
         aria-label={aria}

app/components/markdown.tsx

@@ -23,6 +23,7 @@ import { useChatStore } from "../store";
 import { IconButton } from "./button";
 
 import { useAppConfig } from "../store/config";
+import clsx from "clsx";
 
 export function Mermaid(props: { code: string }) {
   const ref = useRef<HTMLDivElement>(null);
@@ -57,7 +58,7 @@ export function Mermaid(props: { code: string }) {
 
   return (
     <div
-      className="no-dark mermaid"
+      className={clsx("no-dark", "mermaid")}
       style={{
         cursor: "pointer",
         overflow: "auto",
@@ -89,7 +90,11 @@ export function PreCode(props: { children: any }) {
     const refText = ref.current.querySelector("code")?.innerText;
     if (htmlDom) {
       setHtmlCode((htmlDom as HTMLElement).innerText);
-    } else if (refText?.startsWith("<!DOCTYPE")) {
+    } else if (
+      refText?.startsWith("<!DOCTYPE") ||
+      refText?.startsWith("<svg") ||
+      refText?.startsWith("<?xml")
+    ) {
       setHtmlCode(refText);
     }
   }, 600);
@@ -193,7 +198,12 @@ function CustomCode(props: { children: any; className?: string }) {
   const renderShowMoreButton = () => {
     if (showToggle && enableCodeFold && collapsed) {
       return (
-        <div className={`show-hide-button ${collapsed ? "collapsed" : "expanded"}`}>
+        <div
+          className={clsx("show-hide-button", {
+            collapsed,
+            expanded: !collapsed,
+          })}
+        >
           <button onClick={toggleCollapsed}>{Locale.NewChat.More}</button>
         </div>
       );
@@ -203,7 +213,7 @@ function CustomCode(props: { children: any; className?: string }) {
   return (
     <>
       <code
-        className={props?.className}
+        className={clsx(props?.className)}
         ref={ref}
         style={{
           maxHeight: enableCodeFold && collapsed ? "400px" : "none",
@@ -238,6 +248,10 @@ function escapeBrackets(text: string) {
 
 function tryWrapHtmlCode(text: string) {
   // try add wrap html code (fixed: html codeblock include 2 newline)
+  // ignore embed codeblock
+  if (text.includes("```")) {
+    return text;
+  }
   return text
     .replace(
       /([`]*?)(\w*?)([\n\r]*?)(<!DOCTYPE html>)/g,

app/components/mask.tsx

@@ -55,6 +55,7 @@ import {
   OnDragEndResponder,
 } from "@hello-pangea/dnd";
 import { getMessageTextContent } from "../utils";
+import clsx from "clsx";
 
 // drag and drop helper function
 function reorder<T>(list: T[], startIndex: number, endIndex: number): T[] {
@@ -588,7 +589,7 @@ export function MaskPage() {
                   </div>
                   <div className={styles["mask-title"]}>
                     <div className={styles["mask-name"]}>{m.name}</div>
-                    <div className={styles["mask-info"] + " one-line"}>
+                    <div className={clsx(styles["mask-info"], "one-line")}>
                       {`${Locale.Mask.Item.Info(m.context.length)} / ${
                         ALL_LANG_OPTIONS[m.lang]
                       } / ${m.modelConfig.model}`}

app/components/message-selector.tsx

@@ -8,6 +8,7 @@ import Locale from "../locales";
 
 import styles from "./message-selector.module.scss";
 import { getMessageTextContent } from "../utils";
+import clsx from "clsx";
 
 function useShiftRange() {
   const [startIndex, setStartIndex] = useState<number>();
@@ -71,6 +72,7 @@ export function MessageSelector(props: {
   defaultSelectAll?: boolean;
   onSelected?: (messages: ChatMessage[]) => void;
 }) {
+  const LATEST_COUNT = 4;
   const chatStore = useChatStore();
   const session = chatStore.currentSession();
   const isValid = (m: ChatMessage) => m.content && !m.isError && !m.streaming;
@@ -141,15 +143,13 @@ export function MessageSelector(props: {
     // eslint-disable-next-line react-hooks/exhaustive-deps
   }, [startIndex, endIndex]);
 
-  const LATEST_COUNT = 4;
-
   return (
     <div className={styles["message-selector"]}>
       <div className={styles["message-filter"]}>
         <input
           type="text"
           placeholder={Locale.Select.Search}
-          className={styles["filter-item"] + " " + styles["search-bar"]}
+          className={clsx(styles["filter-item"], styles["search-bar"])}
           value={searchInput}
           onInput={(e) => {
             setSearchInput(e.currentTarget.value);
@@ -196,9 +196,9 @@ export function MessageSelector(props: {
 
           return (
             <div
-              className={`${styles["message"]} ${
+              className={clsx(styles["message"], {
-                props.selection.has(m.id!) && styles["message-selected"]
+                [styles["message-selected"]]: props.selection.has(m.id!),
-              }`}
+              })}
               key={i}
               onClick={() => {
                 props.updateSelection((selection) => {
@@ -221,7 +221,7 @@ export function MessageSelector(props: {
                 <div className={styles["date"]}>
                   {new Date(m.date).toLocaleString()}
                 </div>
-                <div className={`${styles["content"]} one-line`}>
+                <div className={clsx(styles["content"], "one-line")}>
                   {getMessageTextContent(m)}
                 </div>
               </div>

app/components/model-config.tsx

@@ -7,6 +7,7 @@ import { ListItem, Select } from "./ui-lib";
 import { useAllModels } from "../utils/hooks";
 import { groupBy } from "lodash-es";
 import styles from "./model-config.module.scss";
+import { getModelProvider } from "../utils/model";
 
 export function ModelConfigList(props: {
   modelConfig: ModelConfig;
@@ -28,7 +29,9 @@ export function ModelConfigList(props: {
           value={value}
           align="left"
           onChange={(e) => {
-            const [model, providerName] = e.currentTarget.value.split("@");
+            const [model, providerName] = getModelProvider(
+              e.currentTarget.value,
+            );
             props.updateConfig((config) => {
               config.model = ModalConfigValidator.model(model);
               config.providerName = providerName as ServiceProvider;
@@ -247,7 +250,9 @@ export function ModelConfigList(props: {
           aria-label={Locale.Settings.CompressModel.Title}
           value={compressModelValue}
           onChange={(e) => {
-            const [model, providerName] = e.currentTarget.value.split("@");
+            const [model, providerName] = getModelProvider(
+              e.currentTarget.value,
+            );
             props.updateConfig((config) => {
               config.compressModel = ModalConfigValidator.model(model);
               config.compressProviderName = providerName as ServiceProvider;

app/components/new-chat.tsx

@@ -16,6 +16,7 @@ import { MaskAvatar } from "./mask";
 import { useCommand } from "../command";
 import { showConfirm } from "./ui-lib";
 import { BUILTIN_MASK_STORE } from "../masks";
+import clsx from "clsx";
 
 function MaskItem(props: { mask: Mask; onClick?: () => void }) {
   return (
@@ -24,7 +25,9 @@ function MaskItem(props: { mask: Mask; onClick?: () => void }) {
         avatar={props.mask.avatar}
         model={props.mask.modelConfig.model}
       />
-      <div className={styles["mask-name"] + " one-line"}>{props.mask.name}</div>
+      <div className={clsx(styles["mask-name"], "one-line")}>
+        {props.mask.name}
+      </div>
     </div>
   );
 }

app/components/plugin.tsx

@@ -28,6 +28,7 @@ import {
 import Locale from "../locales";
 import { useNavigate } from "react-router-dom";
 import { useState } from "react";
+import clsx from "clsx";
 
 export function PluginPage() {
   const navigate = useNavigate();
@@ -199,7 +200,7 @@ export function PluginPage() {
                     <div className={styles["mask-name"]}>
                       {m.title}@<small>{m.version}</small>
                     </div>
-                    <div className={styles["mask-info"] + " one-line"}>
+                    <div className={clsx(styles["mask-info"], "one-line")}>
                       {Locale.Plugin.Item.Info(
                         FunctionToolService.add(m).length,
                       )}
@@ -335,7 +336,10 @@ export function PluginPage() {
               <ListItem
                 subTitle={
                   <div
-                    className={`markdown-body ${pluginStyles["plugin-content"]}`}
+                    className={clsx(
+                      "markdown-body",
+                      pluginStyles["plugin-content"],
+                    )}
                     dir="auto"
                   >
                     <pre>

app/components/realtime-chat/index.ts

@@ -0,0 +1 @@
+export * from "./realtime-chat";

app/components/realtime-chat/realtime-chat.module.scss

@@ -0,0 +1,74 @@
+.realtime-chat {
+  width: 100%;
+  justify-content: center;
+  align-items: center;
+  position: relative;
+  display: flex;
+  flex-direction: column;
+  height: 100%;
+  padding: 20px;
+  box-sizing: border-box;
+  .circle-mic {
+    width: 150px;
+    height: 150px;
+    border-radius: 50%;
+    background: linear-gradient(to bottom right, #a0d8ef, #f0f8ff);
+    display: flex;
+    justify-content: center;
+    align-items: center;
+  }
+  .icon-center {
+    font-size: 24px;
+  }
+
+  .bottom-icons {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    width: 100%;
+    position: absolute;
+    bottom: 20px;
+    box-sizing: border-box;
+    padding: 0 20px;
+  }
+
+  .icon-left,
+  .icon-right {
+    width: 46px;
+    height: 46px;
+    font-size: 36px;
+    background: var(--second);
+    border-radius: 50%;
+    padding: 2px;
+    display: flex;
+    justify-content: center;
+    align-items: center;
+    cursor: pointer;
+    &:hover {
+      opacity: 0.8;
+    }
+  }
+
+  &.mobile {
+    display: none;
+  }
+}
+
+.pulse {
+  animation: pulse 1.5s infinite;
+}
+
+@keyframes pulse {
+  0% {
+    transform: scale(1);
+    opacity: 0.7;
+  }
+  50% {
+    transform: scale(1.1);
+    opacity: 1;
+  }
+  100% {
+    transform: scale(1);
+    opacity: 0.7;
+  }
+}

app/components/realtime-chat/realtime-chat.tsx

@@ -0,0 +1,359 @@
+import VoiceIcon from "@/app/icons/voice.svg";
+import VoiceOffIcon from "@/app/icons/voice-off.svg";
+import PowerIcon from "@/app/icons/power.svg";
+
+import styles from "./realtime-chat.module.scss";
+import clsx from "clsx";
+
+import { useState, useRef, useEffect } from "react";
+
+import { useChatStore, createMessage, useAppConfig } from "@/app/store";
+
+import { IconButton } from "@/app/components/button";
+
+import {
+  Modality,
+  RTClient,
+  RTInputAudioItem,
+  RTResponse,
+  TurnDetection,
+} from "rt-client";
+import { AudioHandler } from "@/app/lib/audio";
+import { uploadImage } from "@/app/utils/chat";
+import { VoicePrint } from "@/app/components/voice-print";
+
+interface RealtimeChatProps {
+  onClose?: () => void;
+  onStartVoice?: () => void;
+  onPausedVoice?: () => void;
+}
+
+export function RealtimeChat({
+  onClose,
+  onStartVoice,
+  onPausedVoice,
+}: RealtimeChatProps) {
+  const chatStore = useChatStore();
+  const session = chatStore.currentSession();
+  const config = useAppConfig();
+  const [status, setStatus] = useState("");
+  const [isRecording, setIsRecording] = useState(false);
+  const [isConnected, setIsConnected] = useState(false);
+  const [isConnecting, setIsConnecting] = useState(false);
+  const [modality, setModality] = useState("audio");
+  const [useVAD, setUseVAD] = useState(true);
+  const [frequencies, setFrequencies] = useState<Uint8Array | undefined>();
+
+  const clientRef = useRef<RTClient | null>(null);
+  const audioHandlerRef = useRef<AudioHandler | null>(null);
+  const initRef = useRef(false);
+
+  const temperature = config.realtimeConfig.temperature;
+  const apiKey = config.realtimeConfig.apiKey;
+  const model = config.realtimeConfig.model;
+  const azure = config.realtimeConfig.provider === "Azure";
+  const azureEndpoint = config.realtimeConfig.azure.endpoint;
+  const azureDeployment = config.realtimeConfig.azure.deployment;
+  const voice = config.realtimeConfig.voice;
+
+  const handleConnect = async () => {
+    if (isConnecting) return;
+    if (!isConnected) {
+      try {
+        setIsConnecting(true);
+        clientRef.current = azure
+          ? new RTClient(
+              new URL(azureEndpoint),
+              { key: apiKey },
+              { deployment: azureDeployment },
+            )
+          : new RTClient({ key: apiKey }, { model });
+        const modalities: Modality[] =
+          modality === "audio" ? ["text", "audio"] : ["text"];
+        const turnDetection: TurnDetection = useVAD
+          ? { type: "server_vad" }
+          : null;
+        await clientRef.current.configure({
+          instructions: "",
+          voice,
+          input_audio_transcription: { model: "whisper-1" },
+          turn_detection: turnDetection,
+          tools: [],
+          temperature,
+          modalities,
+        });
+        startResponseListener();
+
+        setIsConnected(true);
+        // TODO
+        // try {
+        //   const recentMessages = chatStore.getMessagesWithMemory();
+        //   for (const message of recentMessages) {
+        //     const { role, content } = message;
+        //     if (typeof content === "string") {
+        //       await clientRef.current.sendItem({
+        //         type: "message",
+        //         role: role as any,
+        //         content: [
+        //           {
+        //             type: (role === "assistant" ? "text" : "input_text") as any,
+        //             text: content as string,
+        //           },
+        //         ],
+        //       });
+        //     }
+        //   }
+        //   // await clientRef.current.generateResponse();
+        // } catch (error) {
+        //   console.error("Set message failed:", error);
+        // }
+      } catch (error) {
+        console.error("Connection failed:", error);
+        setStatus("Connection failed");
+      } finally {
+        setIsConnecting(false);
+      }
+    } else {
+      await disconnect();
+    }
+  };
+
+  const disconnect = async () => {
+    if (clientRef.current) {
+      try {
+        await clientRef.current.close();
+        clientRef.current = null;
+        setIsConnected(false);
+      } catch (error) {
+        console.error("Disconnect failed:", error);
+      }
+    }
+  };
+
+  const startResponseListener = async () => {
+    if (!clientRef.current) return;
+
+    try {
+      for await (const serverEvent of clientRef.current.events()) {
+        if (serverEvent.type === "response") {
+          await handleResponse(serverEvent);
+        } else if (serverEvent.type === "input_audio") {
+          await handleInputAudio(serverEvent);
+        }
+      }
+    } catch (error) {
+      if (clientRef.current) {
+        console.error("Response iteration error:", error);
+      }
+    }
+  };
+
+  const handleResponse = async (response: RTResponse) => {
+    for await (const item of response) {
+      if (item.type === "message" && item.role === "assistant") {
+        const botMessage = createMessage({
+          role: item.role,
+          content: "",
+        });
+        // add bot message first
+        chatStore.updateTargetSession(session, (session) => {
+          session.messages = session.messages.concat([botMessage]);
+        });
+        let hasAudio = false;
+        for await (const content of item) {
+          if (content.type === "text") {
+            for await (const text of content.textChunks()) {
+              botMessage.content += text;
+            }
+          } else if (content.type === "audio") {
+            const textTask = async () => {
+              for await (const text of content.transcriptChunks()) {
+                botMessage.content += text;
+              }
+            };
+            const audioTask = async () => {
+              audioHandlerRef.current?.startStreamingPlayback();
+              for await (const audio of content.audioChunks()) {
+                hasAudio = true;
+                audioHandlerRef.current?.playChunk(audio);
+              }
+            };
+            await Promise.all([textTask(), audioTask()]);
+          }
+          // update message.content
+          chatStore.updateTargetSession(session, (session) => {
+            session.messages = session.messages.concat();
+          });
+        }
+        if (hasAudio) {
+          // upload audio get audio_url
+          const blob = audioHandlerRef.current?.savePlayFile();
+          uploadImage(blob!).then((audio_url) => {
+            botMessage.audio_url = audio_url;
+            // update text and audio_url
+            chatStore.updateTargetSession(session, (session) => {
+              session.messages = session.messages.concat();
+            });
+          });
+        }
+      }
+    }
+  };
+
+  const handleInputAudio = async (item: RTInputAudioItem) => {
+    await item.waitForCompletion();
+    if (item.transcription) {
+      const userMessage = createMessage({
+        role: "user",
+        content: item.transcription,
+      });
+      chatStore.updateTargetSession(session, (session) => {
+        session.messages = session.messages.concat([userMessage]);
+      });
+      // save input audio_url, and update session
+      const { audioStartMillis, audioEndMillis } = item;
+      // upload audio get audio_url
+      const blob = audioHandlerRef.current?.saveRecordFile(
+        audioStartMillis,
+        audioEndMillis,
+      );
+      uploadImage(blob!).then((audio_url) => {
+        userMessage.audio_url = audio_url;
+        chatStore.updateTargetSession(session, (session) => {
+          session.messages = session.messages.concat();
+        });
+      });
+    }
+    // stop streaming play after get input audio.
+    audioHandlerRef.current?.stopStreamingPlayback();
+  };
+
+  const toggleRecording = async () => {
+    if (!isRecording && clientRef.current) {
+      try {
+        if (!audioHandlerRef.current) {
+          audioHandlerRef.current = new AudioHandler();
+          await audioHandlerRef.current.initialize();
+        }
+        await audioHandlerRef.current.startRecording(async (chunk) => {
+          await clientRef.current?.sendAudio(chunk);
+        });
+        setIsRecording(true);
+      } catch (error) {
+        console.error("Failed to start recording:", error);
+      }
+    } else if (audioHandlerRef.current) {
+      try {
+        audioHandlerRef.current.stopRecording();
+        if (!useVAD) {
+          const inputAudio = await clientRef.current?.commitAudio();
+          await handleInputAudio(inputAudio!);
+          await clientRef.current?.generateResponse();
+        }
+        setIsRecording(false);
+      } catch (error) {
+        console.error("Failed to stop recording:", error);
+      }
+    }
+  };
+
+  useEffect(() => {
+    // 防止重复初始化
+    if (initRef.current) return;
+    initRef.current = true;
+
+    const initAudioHandler = async () => {
+      const handler = new AudioHandler();
+      await handler.initialize();
+      audioHandlerRef.current = handler;
+      await handleConnect();
+      await toggleRecording();
+    };
+
+    initAudioHandler().catch((error) => {
+      setStatus(error);
+      console.error(error);
+    });
+
+    return () => {
+      if (isRecording) {
+        toggleRecording();
+      }
+      audioHandlerRef.current?.close().catch(console.error);
+      disconnect();
+    };
+  }, []);
+
+  useEffect(() => {
+    let animationFrameId: number;
+
+    if (isConnected && isRecording) {
+      const animationFrame = () => {
+        if (audioHandlerRef.current) {
+          const freqData = audioHandlerRef.current.getByteFrequencyData();
+          setFrequencies(freqData);
+        }
+        animationFrameId = requestAnimationFrame(animationFrame);
+      };
+
+      animationFrameId = requestAnimationFrame(animationFrame);
+    } else {
+      setFrequencies(undefined);
+    }
+
+    return () => {
+      if (animationFrameId) {
+        cancelAnimationFrame(animationFrameId);
+      }
+    };
+  }, [isConnected, isRecording]);
+
+  // update session params
+  useEffect(() => {
+    clientRef.current?.configure({ voice });
+  }, [voice]);
+  useEffect(() => {
+    clientRef.current?.configure({ temperature });
+  }, [temperature]);
+
+  const handleClose = async () => {
+    onClose?.();
+    if (isRecording) {
+      await toggleRecording();
+    }
+    disconnect().catch(console.error);
+  };
+
+  return (
+    <div className={styles["realtime-chat"]}>
+      <div
+        className={clsx(styles["circle-mic"], {
+          [styles["pulse"]]: isRecording,
+        })}
+      >
+        <VoicePrint frequencies={frequencies} isActive={isRecording} />
+      </div>
+
+      <div className={styles["bottom-icons"]}>
+        <div>
+          <IconButton
+            icon={isRecording ? <VoiceIcon /> : <VoiceOffIcon />}
+            onClick={toggleRecording}
+            disabled={!isConnected}
+            shadow
+            bordered
+          />
+        </div>
+        <div className={styles["icon-center"]}>{status}</div>
+        <div>
+          <IconButton
+            icon={<PowerIcon />}
+            onClick={handleClose}
+            shadow
+            bordered
+          />
+        </div>
+      </div>
+    </div>
+  );
+}

app/components/realtime-chat/realtime-config.tsx

@@ -0,0 +1,173 @@
+import { RealtimeConfig } from "@/app/store";
+
+import Locale from "@/app/locales";
+import { ListItem, Select, PasswordInput } from "@/app/components/ui-lib";
+
+import { InputRange } from "@/app/components/input-range";
+import { Voice } from "rt-client";
+import { ServiceProvider } from "@/app/constant";
+
+const providers = [ServiceProvider.OpenAI, ServiceProvider.Azure];
+
+const models = ["gpt-4o-realtime-preview-2024-10-01"];
+
+const voice = ["alloy", "shimmer", "echo"];
+
+export function RealtimeConfigList(props: {
+  realtimeConfig: RealtimeConfig;
+  updateConfig: (updater: (config: RealtimeConfig) => void) => void;
+}) {
+  const azureConfigComponent = props.realtimeConfig.provider ===
+    ServiceProvider.Azure && (
+    <>
+      <ListItem
+        title={Locale.Settings.Realtime.Azure.Endpoint.Title}
+        subTitle={Locale.Settings.Realtime.Azure.Endpoint.SubTitle}
+      >
+        <input
+          value={props.realtimeConfig?.azure?.endpoint}
+          type="text"
+          placeholder={Locale.Settings.Realtime.Azure.Endpoint.Title}
+          onChange={(e) => {
+            props.updateConfig(
+              (config) => (config.azure.endpoint = e.currentTarget.value),
+            );
+          }}
+        />
+      </ListItem>
+      <ListItem
+        title={Locale.Settings.Realtime.Azure.Deployment.Title}
+        subTitle={Locale.Settings.Realtime.Azure.Deployment.SubTitle}
+      >
+        <input
+          value={props.realtimeConfig?.azure?.deployment}
+          type="text"
+          placeholder={Locale.Settings.Realtime.Azure.Deployment.Title}
+          onChange={(e) => {
+            props.updateConfig(
+              (config) => (config.azure.deployment = e.currentTarget.value),
+            );
+          }}
+        />
+      </ListItem>
+    </>
+  );
+
+  return (
+    <>
+      <ListItem
+        title={Locale.Settings.Realtime.Enable.Title}
+        subTitle={Locale.Settings.Realtime.Enable.SubTitle}
+      >
+        <input
+          type="checkbox"
+          checked={props.realtimeConfig.enable}
+          onChange={(e) =>
+            props.updateConfig(
+              (config) => (config.enable = e.currentTarget.checked),
+            )
+          }
+        ></input>
+      </ListItem>
+
+      {props.realtimeConfig.enable && (
+        <>
+          <ListItem
+            title={Locale.Settings.Realtime.Provider.Title}
+            subTitle={Locale.Settings.Realtime.Provider.SubTitle}
+          >
+            <Select
+              aria-label={Locale.Settings.Realtime.Provider.Title}
+              value={props.realtimeConfig.provider}
+              onChange={(e) => {
+                props.updateConfig(
+                  (config) =>
+                    (config.provider = e.target.value as ServiceProvider),
+                );
+              }}
+            >
+              {providers.map((v, i) => (
+                <option value={v} key={i}>
+                  {v}
+                </option>
+              ))}
+            </Select>
+          </ListItem>
+          <ListItem
+            title={Locale.Settings.Realtime.Model.Title}
+            subTitle={Locale.Settings.Realtime.Model.SubTitle}
+          >
+            <Select
+              aria-label={Locale.Settings.Realtime.Model.Title}
+              value={props.realtimeConfig.model}
+              onChange={(e) => {
+                props.updateConfig((config) => (config.model = e.target.value));
+              }}
+            >
+              {models.map((v, i) => (
+                <option value={v} key={i}>
+                  {v}
+                </option>
+              ))}
+            </Select>
+          </ListItem>
+          <ListItem
+            title={Locale.Settings.Realtime.ApiKey.Title}
+            subTitle={Locale.Settings.Realtime.ApiKey.SubTitle}
+          >
+            <PasswordInput
+              aria={Locale.Settings.ShowPassword}
+              aria-label={Locale.Settings.Realtime.ApiKey.Title}
+              value={props.realtimeConfig.apiKey}
+              type="text"
+              placeholder={Locale.Settings.Realtime.ApiKey.Placeholder}
+              onChange={(e) => {
+                props.updateConfig(
+                  (config) => (config.apiKey = e.currentTarget.value),
+                );
+              }}
+            />
+          </ListItem>
+          {azureConfigComponent}
+          <ListItem
+            title={Locale.Settings.TTS.Voice.Title}
+            subTitle={Locale.Settings.TTS.Voice.SubTitle}
+          >
+            <Select
+              value={props.realtimeConfig.voice}
+              onChange={(e) => {
+                props.updateConfig(
+                  (config) => (config.voice = e.currentTarget.value as Voice),
+                );
+              }}
+            >
+              {voice.map((v, i) => (
+                <option value={v} key={i}>
+                  {v}
+                </option>
+              ))}
+            </Select>
+          </ListItem>
+          <ListItem
+            title={Locale.Settings.Realtime.Temperature.Title}
+            subTitle={Locale.Settings.Realtime.Temperature.SubTitle}
+          >
+            <InputRange
+              aria={Locale.Settings.Temperature.Title}
+              value={props.realtimeConfig?.temperature?.toFixed(1)}
+              min="0.6"
+              max="1"
+              step="0.1"
+              onChange={(e) => {
+                props.updateConfig(
+                  (config) =>
+                    (config.temperature = e.currentTarget.valueAsNumber),
+                );
+              }}
+            ></InputRange>
+          </ListItem>
+        </>
+      )}
+    </>
+  );
+}

app/components/sd/sd-panel.tsx

@@ -4,6 +4,7 @@ import { Select } from "@/app/components/ui-lib";
 import { IconButton } from "@/app/components/button";
 import Locale from "@/app/locales";
 import { useSdStore } from "@/app/store/sd";
+import clsx from "clsx";
 
 export const params = [
   {
@@ -136,7 +137,7 @@ export function ControlParamItem(props: {
   className?: string;
 }) {
   return (
-    <div className={styles["ctrl-param-item"] + ` ${props.className || ""}`}>
+    <div className={clsx(styles["ctrl-param-item"], props.className)}>
       <div className={styles["ctrl-param-item-header"]}>
         <div className={styles["ctrl-param-item-title"]}>
           <div>

app/components/sd/sd.tsx

@@ -36,6 +36,7 @@ import { removeImage } from "@/app/utils/chat";
 import { SideBar } from "./sd-sidebar";
 import { WindowContent } from "@/app/components/home";
 import { params } from "./sd-panel";
+import clsx from "clsx";
 
 function getSdTaskStatus(item: any) {
   let s: string;
@@ -104,7 +105,7 @@ export function Sd() {
 
   return (
     <>
-      <SideBar className={isSd ? homeStyles["sidebar-show"] : ""} />
+      <SideBar className={clsx({ [homeStyles["sidebar-show"]]: isSd })} />
       <WindowContent>
         <div className={chatStyles.chat} key={"1"}>
           <div className="window-header" data-tauri-drag-region>
@@ -121,7 +122,10 @@ export function Sd() {
               </div>
             )}
             <div
-              className={`window-header-title ${chatStyles["chat-body-title"]}`}
+              className={clsx(
+                "window-header-title",
+                chatStyles["chat-body-title"],
+              )}
             >
               <div className={`window-header-main-title`}>Stability AI</div>
               <div className="window-header-sub-title">

app/components/settings.tsx

@@ -85,6 +85,7 @@ import { nanoid } from "nanoid";
 import { useMaskStore } from "../store/mask";
 import { ProviderType } from "../utils/cloud";
 import { TTSConfigList } from "./tts-config";
+import { RealtimeConfigList } from "./realtime-chat/realtime-config";
 
 function EditPromptModal(props: { id: string; onClose: () => void }) {
   const promptStore = usePromptStore();
@@ -1770,9 +1771,11 @@ export function Settings() {
           <ListItem
             title={Locale.Settings.Access.CustomModel.Title}
             subTitle={Locale.Settings.Access.CustomModel.SubTitle}
+            vertical={true}
           >
             <input
               aria-label={Locale.Settings.Access.CustomModel.Title}
+              style={{ width: "100%", maxWidth: "unset", textAlign: "left" }}
               type="text"
               value={config.customModels}
               placeholder="model1,model2,model3"
@@ -1799,7 +1802,18 @@ export function Settings() {
         {shouldShowPromptModal && (
           <UserPromptModal onClose={() => setShowPromptModal(false)} />
         )}
-
+        <List>
+          <RealtimeConfigList
+            realtimeConfig={config.realtimeConfig}
+            updateConfig={(updater) => {
+              const realtimeConfig = { ...config.realtimeConfig };
+              updater(realtimeConfig);
+              config.update(
+                (config) => (config.realtimeConfig = realtimeConfig),
+              );
+            }}
+          />
+        </List>
         <List>
           <TTSConfigList
             ttsConfig={config.ttsConfig}

app/components/sidebar.tsx

@@ -30,6 +30,7 @@ import { Link, useNavigate } from "react-router-dom";
 import { isIOS, useMobileScreen } from "../utils";
 import dynamic from "next/dynamic";
 import { showConfirm, Selector } from "./ui-lib";
+import clsx from "clsx";
 
 const ChatList = dynamic(async () => (await import("./chat-list")).ChatList, {
   loading: () => null,
@@ -141,9 +142,9 @@ export function SideBarContainer(props: {
   const { children, className, onDragStart, shouldNarrow } = props;
   return (
     <div
-      className={`${styles.sidebar} ${className} ${
+      className={clsx(styles.sidebar, className, {
-        shouldNarrow && styles["narrow-sidebar"]
+        [styles["narrow-sidebar"]]: shouldNarrow,
-      }`}
+      })}
       style={{
         // #3016 disable transition on ios mobile screen
         transition: isMobileScreen && isIOSMobile ? "none" : undefined,
@@ -171,9 +172,9 @@ export function SideBarHeader(props: {
   return (
     <Fragment>
       <div
-        className={`${styles["sidebar-header"]} ${
+        className={clsx(styles["sidebar-header"], {
-          shouldNarrow ? styles["sidebar-header-narrow"] : ""
+          [styles["sidebar-header-narrow"]]: shouldNarrow,
-        }`}
+        })}
         data-tauri-drag-region
       >
         <div className={styles["sidebar-title-container"]}>
@@ -182,7 +183,7 @@ export function SideBarHeader(props: {
           </div>
           <div className={styles["sidebar-sub-title"]}>{subTitle}</div>
         </div>
-        <div className={styles["sidebar-logo"] + " no-dark"}>{logo}</div>
+        <div className={clsx(styles["sidebar-logo"], "no-dark")}>{logo}</div>
       </div>
       {children}
     </Fragment>
@@ -286,7 +287,7 @@ export function SideBar(props: { className?: string }) {
       <SideBarTail
         primaryAction={
           <>
-            <div className={styles["sidebar-action"] + " " + styles.mobile}>
+            <div className={clsx(styles["sidebar-action"], styles.mobile)}>
               <IconButton
                 icon={<DeleteIcon />}
                 onClick={async () => {

app/components/ui-lib.tsx

@@ -23,6 +23,7 @@ import React, {
   useRef,
 } from "react";
 import { IconButton } from "./button";
+import clsx from "clsx";
 
 export function Popover(props: {
   children: JSX.Element;
@@ -45,7 +46,7 @@ export function Popover(props: {
 
 export function Card(props: { children: JSX.Element[]; className?: string }) {
   return (
-    <div className={styles.card + " " + props.className}>{props.children}</div>
+    <div className={clsx(styles.card, props.className)}>{props.children}</div>
   );
 }
 
@@ -60,11 +61,13 @@ export function ListItem(props: {
 }) {
   return (
     <div
-      className={
+      className={clsx(
-        styles["list-item"] +
+        styles["list-item"],
-        ` ${props.vertical ? styles["vertical"] : ""} ` +
+        {
-        ` ${props.className || ""}`
+          [styles["vertical"]]: props.vertical,
-      }
+        },
+        props.className,
+      )}
       onClick={props.onClick}
     >
       <div className={styles["list-header"]}>
@@ -135,9 +138,9 @@ export function Modal(props: ModalProps) {
 
   return (
     <div
-      className={
+      className={clsx(styles["modal-container"], {
-        styles["modal-container"] + ` ${isMax && styles["modal-container-max"]}`
+        [styles["modal-container-max"]]: isMax,
-      }
+      })}
     >
       <div className={styles["modal-header"]}>
         <div className={styles["modal-title"]}>{props.title}</div>
@@ -260,7 +263,7 @@ export function Input(props: InputProps) {
   return (
     <textarea
       {...props}
-      className={`${styles["input"]} ${props.className}`}
+      className={clsx(styles["input"], props.className)}
     ></textarea>
   );
 }
@@ -301,9 +304,13 @@ export function Select(
   const { className, children, align, ...otherProps } = props;
   return (
     <div
-      className={`${styles["select-with-icon"]} ${
+      className={clsx(
-        align === "left" ? styles["left-align-option"] : ""
+        styles["select-with-icon"],
-      } ${className}`}
+        {
+          [styles["left-align-option"]]: align === "left",
+        },
+        className,
+      )}
     >
       <select className={styles["select-with-icon-select"]} {...otherProps}>
         {children}
@@ -509,9 +516,9 @@ export function Selector<T>(props: {
             const selected = selectedValues.includes(item.value);
             return (
               <ListItem
-                className={`${styles["selector-item"]} ${
+                className={clsx(styles["selector-item"], {
-                  item.disable && styles["selector-item-disabled"]
+                  [styles["selector-item-disabled"]]: item.disable,
-                }`}
+                })}
                 key={i}
                 title={item.title}
                 subTitle={item.subTitle}

app/components/voice-print/index.ts

@@ -0,0 +1 @@
+export * from "./voice-print";

app/components/voice-print/voice-print.module.scss

@@ -0,0 +1,11 @@
+.voice-print {
+  width: 100%;
+  height: 60px;
+  margin: 20px 0;
+
+  canvas {
+    width: 100%;
+    height: 100%;
+    filter: brightness(1.2); // 增加整体亮度
+  }
+}

app/components/voice-print/voice-print.tsx

@@ -0,0 +1,180 @@
+import { useEffect, useRef, useCallback } from "react";
+import styles from "./voice-print.module.scss";
+
+interface VoicePrintProps {
+  frequencies?: Uint8Array;
+  isActive?: boolean;
+}
+
+export function VoicePrint({ frequencies, isActive }: VoicePrintProps) {
+  // Canvas引用，用于获取绘图上下文
+  const canvasRef = useRef<HTMLCanvasElement>(null);
+  // 存储历史频率数据，用于平滑处理
+  const historyRef = useRef<number[][]>([]);
+  // 控制保留的历史数据帧数，影响平滑度
+  const historyLengthRef = useRef(10);
+  // 存储动画帧ID，用于清理
+  const animationFrameRef = useRef<number>();
+
+  /**
+   * 更新频率历史数据
+   * 使用FIFO队列维护固定长度的历史记录
+   */
+  const updateHistory = useCallback((freqArray: number[]) => {
+    historyRef.current.push(freqArray);
+    if (historyRef.current.length > historyLengthRef.current) {
+      historyRef.current.shift();
+    }
+  }, []);
+
+  useEffect(() => {
+    const canvas = canvasRef.current;
+    if (!canvas) return;
+
+    const ctx = canvas.getContext("2d");
+    if (!ctx) return;
+
+    /**
+     * 处理高DPI屏幕显示
+     * 根据设备像素比例调整canvas实际渲染分辨率
+     */
+    const dpr = window.devicePixelRatio || 1;
+    canvas.width = canvas.offsetWidth * dpr;
+    canvas.height = canvas.offsetHeight * dpr;
+    ctx.scale(dpr, dpr);
+
+    /**
+     * 主要绘制函数
+     * 使用requestAnimationFrame实现平滑动画
+     * 包含以下步骤：
+     * 1. 清空画布
+     * 2. 更新历史数据
+     * 3. 计算波形点
+     * 4. 绘制上下对称的声纹
+     */
+    const draw = () => {
+      // 清空画布
+      ctx.clearRect(0, 0, canvas.width, canvas.height);
+
+      if (!frequencies || !isActive) {
+        historyRef.current = [];
+        return;
+      }
+
+      const freqArray = Array.from(frequencies);
+      updateHistory(freqArray);
+
+      // 绘制声纹
+      const points: [number, number][] = [];
+      const centerY = canvas.height / 2;
+      const width = canvas.width;
+      const sliceWidth = width / (frequencies.length - 1);
+
+      // 绘制主波形
+      ctx.beginPath();
+      ctx.moveTo(0, centerY);
+
+      /**
+       * 声纹绘制算法：
+       * 1. 使用历史数据平均值实现平滑过渡
+       * 2. 通过正弦函数添加自然波动
+       * 3. 使用贝塞尔曲线连接点，使曲线更平滑
+       * 4. 绘制对称部分形成完整声纹
+       */
+      for (let i = 0; i < frequencies.length; i++) {
+        const x = i * sliceWidth;
+        let avgFrequency = frequencies[i];
+
+        /**
+         * 波形平滑处理：
+         * 1. 收集历史数据中对应位置的频率值
+         * 2. 计算当前值与历史值的加权平均
+         * 3. 根据平均值计算实际显示高度
+         */
+        if (historyRef.current.length > 0) {
+          const historicalValues = historyRef.current.map((h) => h[i] || 0);
+          avgFrequency =
+            (avgFrequency + historicalValues.reduce((a, b) => a + b, 0)) /
+            (historyRef.current.length + 1);
+        }
+
+        /**
+         * 波形变换：
+         * 1. 归一化频率值到0-1范围
+         * 2. 添加时间相关的正弦变换
+         * 3. 使用贝塞尔曲线平滑连接点
+         */
+        const normalized = avgFrequency / 255.0;
+        const height = normalized * (canvas.height / 2);
+        const y = centerY + height * Math.sin(i * 0.2 + Date.now() * 0.002);
+
+        points.push([x, y]);
+
+        if (i === 0) {
+          ctx.moveTo(x, y);
+        } else {
+          // 使用贝塞尔曲线使波形更平滑
+          const prevPoint = points[i - 1];
+          const midX = (prevPoint[0] + x) / 2;
+          ctx.quadraticCurveTo(
+            prevPoint[0],
+            prevPoint[1],
+            midX,
+            (prevPoint[1] + y) / 2,
+          );
+        }
+      }
+
+      // 绘制对称的下半部分
+      for (let i = points.length - 1; i >= 0; i--) {
+        const [x, y] = points[i];
+        const symmetricY = centerY - (y - centerY);
+        if (i === points.length - 1) {
+          ctx.lineTo(x, symmetricY);
+        } else {
+          const nextPoint = points[i + 1];
+          const midX = (nextPoint[0] + x) / 2;
+          ctx.quadraticCurveTo(
+            nextPoint[0],
+            centerY - (nextPoint[1] - centerY),
+            midX,
+            centerY - ((nextPoint[1] + y) / 2 - centerY),
+          );
+        }
+      }
+
+      ctx.closePath();
+
+      /**
+       * 渐变效果：
+       * 从左到右应用三色渐变，带透明度
+       * 使用蓝色系配色提升视觉效果
+       */
+      const gradient = ctx.createLinearGradient(0, 0, canvas.width, 0);
+      gradient.addColorStop(0, "rgba(100, 180, 255, 0.95)");
+      gradient.addColorStop(0.5, "rgba(140, 200, 255, 0.9)");
+      gradient.addColorStop(1, "rgba(180, 220, 255, 0.95)");
+
+      ctx.fillStyle = gradient;
+      ctx.fill();
+
+      animationFrameRef.current = requestAnimationFrame(draw);
+    };
+
+    // 启动动画循环
+    draw();
+
+    // 清理函数：在组件卸载时取消动画
+    return () => {
+      if (animationFrameRef.current) {
+        cancelAnimationFrame(animationFrameRef.current);
+      }
+    };
+  }, [frequencies, isActive, updateHistory]);
+
+  return (
+    <div className={styles["voice-print"]}>
+      <canvas ref={canvasRef} />
+    </div>
+  );
+}

app/config/build.ts

@@ -40,6 +40,7 @@ export const getBuildConfig = () => {
     buildMode,
     isApp,
     template: process.env.DEFAULT_INPUT_TEMPLATE ?? DEFAULT_INPUT_TEMPLATE,
+    visionModels: process.env.VISION_MODELS || "",
   };
 };
 

app/config/server.ts

@@ -129,14 +129,15 @@ export const getServerSideConfig = () => {
     if (customModels) customModels += ",";
     customModels += DEFAULT_MODELS.filter(
       (m) =>
-        (m.name.startsWith("gpt-4") || m.name.startsWith("chatgpt-4o")) &&
+        (m.name.startsWith("gpt-4") || m.name.startsWith("chatgpt-4o") || m.name.startsWith("o1")) &&
         !m.name.startsWith("gpt-4o-mini"),
     )
       .map((m) => "-" + m.name)
       .join(",");
     if (
       (defaultModel.startsWith("gpt-4") ||
-        defaultModel.startsWith("chatgpt-4o")) &&
+        defaultModel.startsWith("chatgpt-4o") ||
+        defaultModel.startsWith("o1")) &&
       !defaultModel.startsWith("gpt-4o-mini")
     )
       defaultModel = "";

app/constant.ts

@@ -232,7 +232,9 @@ export const XAI = {
 
 export const ChatGLM = {
   ExampleEndpoint: CHATGLM_BASE_URL,
-  ChatPath: "/api/paas/v4/chat/completions",
+  ChatPath: "api/paas/v4/chat/completions",
+  ImagePath: "api/paas/v4/images/generations",
+  VideoPath: "api/paas/v4/videos/generations",
 };
 
 export const DEFAULT_INPUT_TEMPLATE = `{{input}}`; // input / time / model / lang
@@ -264,6 +266,7 @@ export const KnowledgeCutOffDate: Record<string, string> = {
   "gpt-4o": "2023-10",
   "gpt-4o-2024-05-13": "2023-10",
   "gpt-4o-2024-08-06": "2023-10",
+  "gpt-4o-2024-11-20": "2023-10",
   "chatgpt-4o-latest": "2023-10",
   "gpt-4o-mini": "2023-10",
   "gpt-4o-mini-2024-07-18": "2023-10",
@@ -290,6 +293,22 @@ export const DEFAULT_TTS_VOICES = [
   "shimmer",
 ];
 
+export const VISION_MODEL_REGEXES = [
+  /vision/,
+  /gpt-4o/,
+  /claude-3/,
+  /gemini-1\.5/,
+  /gemini-exp/,
+  /gemini-2\.0/,
+  /learnlm/,
+  /qwen-vl/,
+  /qwen2-vl/,
+  /gpt-4-turbo(?!.*preview)/, // Matches "gpt-4-turbo" but not "gpt-4-turbo-preview"
+  /^dall-e-3$/, // Matches exactly "dall-e-3"
+];
+
+export const EXCLUDE_VISION_MODEL_REGEXES = [/claude-3-5-haiku-20241022/];
+
 const openaiModels = [
   "gpt-3.5-turbo",
   "gpt-3.5-turbo-1106",
@@ -303,6 +322,7 @@ const openaiModels = [
   "gpt-4o",
   "gpt-4o-2024-05-13",
   "gpt-4o-2024-08-06",
+  "gpt-4o-2024-11-20",
   "chatgpt-4o-latest",
   "gpt-4o-mini",
   "gpt-4o-mini-2024-07-18",
@@ -315,10 +335,23 @@ const openaiModels = [
 ];
 
 const googleModels = [
-  "gemini-1.0-pro",
+  "gemini-1.0-pro", // Deprecated on 2/15/2025
   "gemini-1.5-pro-latest",
+  "gemini-1.5-pro",
+  "gemini-1.5-pro-002",
+  "gemini-1.5-pro-exp-0827",
   "gemini-1.5-flash-latest",
-  "gemini-pro-vision",
+  "gemini-1.5-flash-8b-latest",
+  "gemini-1.5-flash",
+  "gemini-1.5-flash-8b",
+  "gemini-1.5-flash-002",
+  "gemini-1.5-flash-exp-0827",
+  "learnlm-1.5-pro-experimental",
+  "gemini-exp-1114",
+  "gemini-exp-1121",
+  "gemini-exp-1206",
+  "gemini-2.0-flash-exp",
+  "gemini-2.0-flash-thinking-exp-1219",
 ];
 
 const anthropicModels = [
@@ -329,10 +362,11 @@ const anthropicModels = [
   "claude-3-opus-20240229",
   "claude-3-opus-latest",
   "claude-3-haiku-20240307",
+  "claude-3-5-haiku-20241022",
+  "claude-3-5-haiku-latest",
   "claude-3-5-sonnet-20240620",
   "claude-3-5-sonnet-20241022",
   "claude-3-5-sonnet-latest",
-  "claude-3-5-haiku-latest",
 ];
 
 const baiduModels = [
@@ -399,6 +433,15 @@ const chatglmModels = [
   "glm-4-long",
   "glm-4-flashx",
   "glm-4-flash",
+  "glm-4v-plus",
+  "glm-4v",
+  "glm-4v-flash", // free
+  "cogview-3-plus",
+  "cogview-3",
+  "cogview-3-flash", // free
+  // 目前无法适配轮询任务
+  //   "cogvideox",
+  //   "cogvideox-flash", // free
 ];
 
 let seq = 1000; // 内置的模型序号生成器从1000开始

app/icons/headphone.svg

@@ -0,0 +1,11 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="16" height="16" viewBox="0 0 48 48" fill="none" xmlns="http://www.w3.org/2000/svg">
+    <path d="M4 28C4 26.8954 4.89543 26 6 26H10V38H6C4.89543 38 4 37.1046 4 36V28Z" fill="none" />
+    <path d="M38 26H42C43.1046 26 44 26.8954 44 28V36C44 37.1046 43.1046 38 42 38H38V26Z"
+        fill="none" />
+    <path
+        d="M10 36V24C10 16.268 16.268 10 24 10C31.732 10 38 16.268 38 24V36M10 26H6C4.89543 26 4 26.8954 4 28V36C4 37.1046 4.89543 38 6 38H10V26ZM38 26H42C43.1046 26 44 26.8954 44 28V36C44 37.1046 43.1046 38 42 38H38V26Z"
+        stroke="#333" stroke-width="4" stroke-linecap="round" stroke-linejoin="round" />
+    <path d="M16 32H20L22 26L26 38L28 32H32" stroke="#333" stroke-width="4" stroke-linecap="round"
+        stroke-linejoin="round" />
+</svg>

app/icons/power.svg

@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="24" height="24" viewBox="0 0 48 48" fill="none" xmlns="http://www.w3.org/2000/svg">
+    <path
+        d="M14.5 8C13.8406 8.37652 13.2062 8.79103 12.6 9.24051C11.5625 10.0097 10.6074 10.8814 9.75 11.8402C6.79377 15.1463 5 19.4891 5 24.2455C5 34.6033 13.5066 43 24 43C34.4934 43 43 34.6033 43 24.2455C43 19.4891 41.2062 15.1463 38.25 11.8402C37.3926 10.8814 36.4375 10.0097 35.4 9.24051C34.7938 8.79103 34.1594 8.37652 33.5 8"
+        stroke="#333" stroke-width="4" stroke-linecap="round" stroke-linejoin="round" />
+    <path d="M24 4V24" stroke="#333" stroke-width="4" stroke-linecap="round" stroke-linejoin="round" />
+</svg>

app/icons/voice-off.svg

@@ -0,0 +1,13 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="24" height="24" viewBox="0 0 48 48" fill="none" xmlns="http://www.w3.org/2000/svg">
+    <path
+        d="M31 24V11C31 7.13401 27.866 4 24 4C20.134 4 17 7.13401 17 11V24C17 27.866 20.134 31 24 31C27.866 31 31 27.866 31 24Z"
+        stroke="#d0021b" stroke-width="4" stroke-linejoin="round" />
+    <path
+        d="M9 23C9 31.2843 15.7157 38 24 38C25.7532 38 27.4361 37.6992 29 37.1465M39 23C39 25.1333 38.5547 27.1626 37.7519 29"
+        stroke="#d0021b" stroke-width="4" stroke-linecap="round" stroke-linejoin="round" />
+    <path d="M24 38V44" stroke="#d0021b" stroke-width="4" stroke-linecap="round"
+        stroke-linejoin="round" />
+    <path d="M42 42L6 6" stroke="#d0021b" stroke-width="4" stroke-linecap="round"
+        stroke-linejoin="round" />
+</svg>

app/icons/voice.svg

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="24" height="24" viewBox="0 0 48 48" fill="none" xmlns="http://www.w3.org/2000/svg">
+    <rect x="17" y="4" width="14" height="27" rx="7" fill="none" stroke="#333" stroke-width="4"
+        stroke-linejoin="round" />
+    <path d="M9 23C9 31.2843 15.7157 38 24 38C32.2843 38 39 31.2843 39 23" stroke="#333"
+        stroke-width="4" stroke-linecap="round" stroke-linejoin="round" />
+    <path d="M24 38V44" stroke="#333" stroke-width="4" stroke-linecap="round"
+        stroke-linejoin="round" />
+</svg>

app/lib/audio.ts

@@ -0,0 +1,200 @@
+export class AudioHandler {
+  private context: AudioContext;
+  private mergeNode: ChannelMergerNode;
+  private analyserData: Uint8Array;
+  public analyser: AnalyserNode;
+  private workletNode: AudioWorkletNode | null = null;
+  private stream: MediaStream | null = null;
+  private source: MediaStreamAudioSourceNode | null = null;
+  private recordBuffer: Int16Array[] = [];
+  private readonly sampleRate = 24000;
+
+  private nextPlayTime: number = 0;
+  private isPlaying: boolean = false;
+  private playbackQueue: AudioBufferSourceNode[] = [];
+  private playBuffer: Int16Array[] = [];
+
+  constructor() {
+    this.context = new AudioContext({ sampleRate: this.sampleRate });
+    // using ChannelMergerNode to get merged audio data, and then get analyser data.
+    this.mergeNode = new ChannelMergerNode(this.context, { numberOfInputs: 2 });
+    this.analyser = new AnalyserNode(this.context, { fftSize: 256 });
+    this.analyserData = new Uint8Array(this.analyser.frequencyBinCount);
+    this.mergeNode.connect(this.analyser);
+  }
+
+  getByteFrequencyData() {
+    this.analyser.getByteFrequencyData(this.analyserData);
+    return this.analyserData;
+  }
+
+  async initialize() {
+    await this.context.audioWorklet.addModule("/audio-processor.js");
+  }
+
+  async startRecording(onChunk: (chunk: Uint8Array) => void) {
+    try {
+      if (!this.workletNode) {
+        await this.initialize();
+      }
+
+      this.stream = await navigator.mediaDevices.getUserMedia({
+        audio: {
+          channelCount: 1,
+          sampleRate: this.sampleRate,
+          echoCancellation: true,
+          noiseSuppression: true,
+        },
+      });
+
+      await this.context.resume();
+      this.source = this.context.createMediaStreamSource(this.stream);
+      this.workletNode = new AudioWorkletNode(
+        this.context,
+        "audio-recorder-processor",
+      );
+
+      this.workletNode.port.onmessage = (event) => {
+        if (event.data.eventType === "audio") {
+          const float32Data = event.data.audioData;
+          const int16Data = new Int16Array(float32Data.length);
+
+          for (let i = 0; i < float32Data.length; i++) {
+            const s = Math.max(-1, Math.min(1, float32Data[i]));
+            int16Data[i] = s < 0 ? s * 0x8000 : s * 0x7fff;
+          }
+
+          const uint8Data = new Uint8Array(int16Data.buffer);
+          onChunk(uint8Data);
+          // save recordBuffer
+          // @ts-ignore
+          this.recordBuffer.push.apply(this.recordBuffer, int16Data);
+        }
+      };
+
+      this.source.connect(this.workletNode);
+      this.source.connect(this.mergeNode, 0, 0);
+      this.workletNode.connect(this.context.destination);
+
+      this.workletNode.port.postMessage({ command: "START_RECORDING" });
+    } catch (error) {
+      console.error("Error starting recording:", error);
+      throw error;
+    }
+  }
+
+  stopRecording() {
+    if (!this.workletNode || !this.source || !this.stream) {
+      throw new Error("Recording not started");
+    }
+
+    this.workletNode.port.postMessage({ command: "STOP_RECORDING" });
+
+    this.workletNode.disconnect();
+    this.source.disconnect();
+    this.stream.getTracks().forEach((track) => track.stop());
+  }
+  startStreamingPlayback() {
+    this.isPlaying = true;
+    this.nextPlayTime = this.context.currentTime;
+  }
+
+  stopStreamingPlayback() {
+    this.isPlaying = false;
+    this.playbackQueue.forEach((source) => source.stop());
+    this.playbackQueue = [];
+    this.playBuffer = [];
+  }
+
+  playChunk(chunk: Uint8Array) {
+    if (!this.isPlaying) return;
+
+    const int16Data = new Int16Array(chunk.buffer);
+    // @ts-ignore
+    this.playBuffer.push.apply(this.playBuffer, int16Data); // save playBuffer
+
+    const float32Data = new Float32Array(int16Data.length);
+    for (let i = 0; i < int16Data.length; i++) {
+      float32Data[i] = int16Data[i] / (int16Data[i] < 0 ? 0x8000 : 0x7fff);
+    }
+
+    const audioBuffer = this.context.createBuffer(
+      1,
+      float32Data.length,
+      this.sampleRate,
+    );
+    audioBuffer.getChannelData(0).set(float32Data);
+
+    const source = this.context.createBufferSource();
+    source.buffer = audioBuffer;
+    source.connect(this.context.destination);
+    source.connect(this.mergeNode, 0, 1);
+
+    const chunkDuration = audioBuffer.length / this.sampleRate;
+
+    source.start(this.nextPlayTime);
+
+    this.playbackQueue.push(source);
+    source.onended = () => {
+      const index = this.playbackQueue.indexOf(source);
+      if (index > -1) {
+        this.playbackQueue.splice(index, 1);
+      }
+    };
+
+    this.nextPlayTime += chunkDuration;
+
+    if (this.nextPlayTime < this.context.currentTime) {
+      this.nextPlayTime = this.context.currentTime;
+    }
+  }
+  _saveData(data: Int16Array, bytesPerSample = 16): Blob {
+    const headerLength = 44;
+    const numberOfChannels = 1;
+    const byteLength = data.buffer.byteLength;
+    const header = new Uint8Array(headerLength);
+    const view = new DataView(header.buffer);
+    view.setUint32(0, 1380533830, false); // RIFF identifier 'RIFF'
+    view.setUint32(4, 36 + byteLength, true); // file length minus RIFF identifier length and file description length
+    view.setUint32(8, 1463899717, false); // RIFF type 'WAVE'
+    view.setUint32(12, 1718449184, false); // format chunk identifier 'fmt '
+    view.setUint32(16, 16, true); // format chunk length
+    view.setUint16(20, 1, true); // sample format (raw)
+    view.setUint16(22, numberOfChannels, true); // channel count
+    view.setUint32(24, this.sampleRate, true); // sample rate
+    view.setUint32(28, this.sampleRate * 4, true); // byte rate (sample rate * block align)
+    view.setUint16(32, numberOfChannels * 2, true); // block align (channel count * bytes per sample)
+    view.setUint16(34, bytesPerSample, true); // bits per sample
+    view.setUint32(36, 1684108385, false); // data chunk identifier 'data'
+    view.setUint32(40, byteLength, true); // data chunk length
+
+    // using data.buffer, so no need to setUint16 to view.
+    return new Blob([view, data.buffer], { type: "audio/mpeg" });
+  }
+  savePlayFile() {
+    // @ts-ignore
+    return this._saveData(new Int16Array(this.playBuffer));
+  }
+  saveRecordFile(
+    audioStartMillis: number | undefined,
+    audioEndMillis: number | undefined,
+  ) {
+    const startIndex = audioStartMillis
+      ? Math.floor((audioStartMillis * this.sampleRate) / 1000)
+      : 0;
+    const endIndex = audioEndMillis
+      ? Math.floor((audioEndMillis * this.sampleRate) / 1000)
+      : this.recordBuffer.length;
+    return this._saveData(
+      // @ts-ignore
+      new Int16Array(this.recordBuffer.slice(startIndex, endIndex)),
+    );
+  }
+  async close() {
+    this.recordBuffer = [];
+    this.workletNode?.disconnect();
+    this.source?.disconnect();
+    this.stream?.getTracks().forEach((track) => track.stop());
+    await this.context.close();
+  }
+}

app/locales/cn.ts

@@ -562,6 +562,39 @@ const cn = {
         SubTitle: "生成语音的速度",
       },
     },
+    Realtime: {
+      Enable: {
+        Title: "实时聊天",
+        SubTitle: "开启实时聊天功能",
+      },
+      Provider: {
+        Title: "模型服务商",
+        SubTitle: "切换不同的服务商",
+      },
+      Model: {
+        Title: "模型",
+        SubTitle: "选择一个模型",
+      },
+      ApiKey: {
+        Title: "API Key",
+        SubTitle: "API Key",
+        Placeholder: "API Key",
+      },
+      Azure: {
+        Endpoint: {
+          Title: "接口地址",
+          SubTitle: "接口地址",
+        },
+        Deployment: {
+          Title: "部署名称",
+          SubTitle: "部署名称",
+        },
+      },
+      Temperature: {
+        Title: "随机性 (temperature)",
+        SubTitle: "值越大，回复越随机",
+      },
+    },
   },
   Store: {
     DefaultTopic: "新的聊天",

app/locales/en.ts

@@ -570,6 +570,39 @@ const en: LocaleType = {
       },
       Engine: "TTS Engine",
     },
+    Realtime: {
+      Enable: {
+        Title: "Realtime Chat",
+        SubTitle: "Enable realtime chat feature",
+      },
+      Provider: {
+        Title: "Model Provider",
+        SubTitle: "Switch between different providers",
+      },
+      Model: {
+        Title: "Model",
+        SubTitle: "Select a model",
+      },
+      ApiKey: {
+        Title: "API Key",
+        SubTitle: "API Key",
+        Placeholder: "API Key",
+      },
+      Azure: {
+        Endpoint: {
+          Title: "Endpoint",
+          SubTitle: "Endpoint",
+        },
+        Deployment: {
+          Title: "Deployment Name",
+          SubTitle: "Deployment Name",
+        },
+      },
+      Temperature: {
+        Title: "Randomness (temperature)",
+        SubTitle: "Higher values result in more random responses",
+      },
+    },
   },
   Store: {
     DefaultTopic: "New Conversation",

app/masks/cn.ts

@@ -3,7 +3,7 @@ import { BuiltinMask } from "./typing";
 export const CN_MASKS: BuiltinMask[] = [
   {
     avatar: "1f5bc-fe0f",
-    name: "以文搜图",
+    name: "AI文生图",
     context: [
       {
         id: "text-to-pic-0",
@@ -28,7 +28,7 @@ export const CN_MASKS: BuiltinMask[] = [
         id: "text-to-pic-3",
         role: "system",
         content:
-          "助手善于判断用户意图，当确定需要提供图片时，助手会变得沉默寡言，只使用以下格式输出markdown图片：![描述](https://image.pollinations.ai/prompt/描述)，因为这个语法可以自动按照提示生成并渲染图片。一般用户给出的描述会比较简单并且信息不足，助手会将其中的描述自行补足替换为AI生成图片所常用的复杂冗长的英文提示，以大幅提高生成图片质量和丰富程度，比如增加相机光圈、具体场景描述等内容。助手会避免用代码块或原始块包围markdown标记，因为那样只会渲染出代码块或原始块而不是图片。",
+          "助手善于判断用户意图，当确定需要提供图片时，助手会变得沉默寡言，只使用以下格式输出markdown图片：![description](https://image.pollinations.ai/prompt/description?nologo=true)，因为这个语法可以自动按照提示生成并渲染图片。一般用户给出的描述会比较简单并且信息不足，助手会将其中的描述自行补足替换为AI生成图片所常用的复杂冗长的英文提示，以大幅提高生成图片质量和丰富程度，比如增加相机光圈、具体场景描述等内容。助手会避免用代码块或原始块包围markdown标记，因为那样只会渲染出代码块或原始块而不是图片。url中的空格等符号需要转义。",
         date: "",
       },
     ],

app/store/access.ts

@@ -21,6 +21,7 @@ import { getClientConfig } from "../config/client";
 import { createPersistStore } from "../utils/store";
 import { ensure } from "../utils/clone";
 import { DEFAULT_CONFIG } from "./config";
+import { getModelProvider } from "../utils/model";
 
 let fetchState = 0; // 0 not fetch, 1 fetching, 2 done
 
@@ -226,9 +227,9 @@ export const useAccessStore = createPersistStore(
         .then((res) => {
           const defaultModel = res.defaultModel ?? "";
           if (defaultModel !== "") {
-            const [model, providerName] = defaultModel.split("@");
+            const [model, providerName] = getModelProvider(defaultModel);
             DEFAULT_CONFIG.modelConfig.model = model;
-            DEFAULT_CONFIG.modelConfig.providerName = providerName;
+            DEFAULT_CONFIG.modelConfig.providerName = providerName as any;
           }
 
           return res;

app/store/chat.ts

@@ -52,6 +52,7 @@ export type ChatMessage = RequestMessage & {
   id: string;
   model?: ModelType;
   tools?: ChatMessageTool[];
+  audio_url?: string;
 };
 
 export function createMessage(override: Partial<ChatMessage>): ChatMessage {
@@ -357,7 +358,7 @@ export const useChatStore = createPersistStore(
           session.messages = session.messages.concat();
           session.lastUpdate = Date.now();
         });
-        get().updateStat(message);
+        get().updateStat(message, targetSession);
         get().summarizeSession(false, targetSession);
       },
 
@@ -396,10 +397,10 @@ export const useChatStore = createPersistStore(
         // get recent messages
         const recentMessages = get().getMessagesWithMemory();
         const sendMessages = recentMessages.concat(userMessage);
-        const messageIndex = get().currentSession().messages.length + 1;
+        const messageIndex = session.messages.length + 1;
 
         // save user's and bot's message
-        get().updateCurrentSession((session) => {
+        get().updateTargetSession(session, (session) => {
           const savedUserMessage = {
             ...userMessage,
             content: mContent,
@@ -420,7 +421,7 @@ export const useChatStore = createPersistStore(
             if (message) {
               botMessage.content = message;
             }
-            get().updateCurrentSession((session) => {
+            get().updateTargetSession(session, (session) => {
               session.messages = session.messages.concat();
             });
           },
@@ -428,13 +429,14 @@ export const useChatStore = createPersistStore(
             botMessage.streaming = false;
             if (message) {
               botMessage.content = message;
+              botMessage.date = new Date().toLocaleString();
               get().onNewMessage(botMessage, session);
             }
             ChatControllerPool.remove(session.id, botMessage.id);
           },
           onBeforeTool(tool: ChatMessageTool) {
             (botMessage.tools = botMessage?.tools || []).push(tool);
-            get().updateCurrentSession((session) => {
+            get().updateTargetSession(session, (session) => {
               session.messages = session.messages.concat();
             });
           },
@@ -444,7 +446,7 @@ export const useChatStore = createPersistStore(
                 tools[i] = { ...tool };
               }
             });
-            get().updateCurrentSession((session) => {
+            get().updateTargetSession(session, (session) => {
               session.messages = session.messages.concat();
             });
           },
@@ -459,7 +461,7 @@ export const useChatStore = createPersistStore(
             botMessage.streaming = false;
             userMessage.isError = !isAborted;
             botMessage.isError = !isAborted;
-            get().updateCurrentSession((session) => {
+            get().updateTargetSession(session, (session) => {
               session.messages = session.messages.concat();
             });
             ChatControllerPool.remove(
@@ -591,8 +593,8 @@ export const useChatStore = createPersistStore(
         set(() => ({ sessions }));
       },
 
-      resetSession() {
+      resetSession(session: ChatSession) {
-        get().updateCurrentSession((session) => {
+        get().updateTargetSession(session, (session) => {
           session.messages = [];
           session.memoryPrompt = "";
         });
@@ -736,19 +738,12 @@ export const useChatStore = createPersistStore(
         }
       },
 
-      updateStat(message: ChatMessage) {
+      updateStat(message: ChatMessage, session: ChatSession) {
-        get().updateCurrentSession((session) => {
+        get().updateTargetSession(session, (session) => {
           session.stat.charCount += message.content.length;
           // TODO: should update chat count and word count
         });
       },
-
-      updateCurrentSession(updater: (session: ChatSession) => void) {
-        const sessions = get().sessions;
-        const index = get().currentSessionIndex;
-        updater(sessions[index]);
-        set(() => ({ sessions }));
-      },
       updateTargetSession(
         targetSession: ChatSession,
         updater: (session: ChatSession) => void,

app/store/config.ts

@@ -1,5 +1,5 @@
 import { LLMModel } from "../client/api";
-import { DalleSize, DalleQuality, DalleStyle } from "../typing";
+import { DalleQuality, DalleStyle, ModelSize } from "../typing";
 import { getClientConfig } from "../config/client";
 import {
   DEFAULT_INPUT_TEMPLATE,
@@ -15,6 +15,7 @@ import {
   ServiceProvider,
 } from "../constant";
 import { createPersistStore } from "../utils/store";
+import type { Voice } from "rt-client";
 
 export type ModelType = (typeof DEFAULT_MODELS)[number]["name"];
 export type TTSModelType = (typeof DEFAULT_TTS_MODELS)[number];
@@ -77,7 +78,7 @@ export const DEFAULT_CONFIG = {
     compressProviderName: "",
     enableInjectSystemPrompts: true,
     template: config?.template ?? DEFAULT_INPUT_TEMPLATE,
-    size: "1024x1024" as DalleSize,
+    size: "1024x1024" as ModelSize,
     quality: "standard" as DalleQuality,
     style: "vivid" as DalleStyle,
   },
@@ -90,12 +91,26 @@ export const DEFAULT_CONFIG = {
     voice: DEFAULT_TTS_VOICE,
     speed: 1.0,
   },
+
+  realtimeConfig: {
+    enable: false,
+    provider: "OpenAI" as ServiceProvider,
+    model: "gpt-4o-realtime-preview-2024-10-01",
+    apiKey: "",
+    azure: {
+      endpoint: "",
+      deployment: "",
+    },
+    temperature: 0.9,
+    voice: "alloy" as Voice,
+  },
 };
 
 export type ChatConfig = typeof DEFAULT_CONFIG;
 
 export type ModelConfig = ChatConfig["modelConfig"];
 export type TTSConfig = ChatConfig["ttsConfig"];
+export type RealtimeConfig = ChatConfig["realtimeConfig"];
 
 export function limitNumber(
   x: number,

app/typing.ts

@@ -11,3 +11,14 @@ export interface RequestMessage {
 export type DalleSize = "1024x1024" | "1792x1024" | "1024x1792";
 export type DalleQuality = "standard" | "hd";
 export type DalleStyle = "vivid" | "natural";
+
+export type ModelSize =
+  | "1024x1024"
+  | "1792x1024"
+  | "1024x1792"
+  | "768x1344"
+  | "864x1152"
+  | "1344x768"
+  | "1152x864"
+  | "1440x720"
+  | "720x1440";

app/utils.ts

@@ -5,6 +5,9 @@ import { RequestMessage } from "./client/api";
 import { ServiceProvider } from "./constant";
 // import { fetch as tauriFetch, ResponseType } from "@tauri-apps/api/http";
 import { fetch as tauriStreamFetch } from "./utils/stream";
+import { VISION_MODEL_REGEXES, EXCLUDE_VISION_MODEL_REGEXES } from "./constant";
+import { getClientConfig } from "./config/client";
+import { ModelSize } from "./typing";
 
 export function trimTopic(topic: string) {
   // Fix an issue where double quotes still show in the Indonesian language
@@ -252,23 +255,16 @@ export function getMessageImages(message: RequestMessage): string[] {
 }
 
 export function isVisionModel(model: string) {
-  // Note: This is a better way using the TypeScript feature instead of `&&` or `||` (ts v5.5.0-dev.20240314 I've been using)
+  const clientConfig = getClientConfig();
-
+  const envVisionModels = clientConfig?.visionModels
-  const visionKeywords = [
+    ?.split(",")
-    "vision",
+    .map((m) => m.trim());
-    "claude-3",
+  if (envVisionModels?.includes(model)) {
-    "gemini-1.5-pro",
+    return true;
-    "gemini-1.5-flash",
+  }
-    "gpt-4o",
-    "gpt-4o-mini",
-  ];
-  const isGpt4Turbo =
-    model.includes("gpt-4-turbo") && !model.includes("preview");
-
   return (
-    visionKeywords.some((keyword) => model.includes(keyword)) ||
+    !EXCLUDE_VISION_MODEL_REGEXES.some((regex) => regex.test(model)) &&
-    isGpt4Turbo ||
+    VISION_MODEL_REGEXES.some((regex) => regex.test(model))
-    isDalle3(model)
   );
 }
 
@@ -276,6 +272,28 @@ export function isDalle3(model: string) {
   return "dall-e-3" === model;
 }
 
+export function getModelSizes(model: string): ModelSize[] {
+  if (isDalle3(model)) {
+    return ["1024x1024", "1792x1024", "1024x1792"];
+  }
+  if (model.toLowerCase().includes("cogview")) {
+    return [
+      "1024x1024",
+      "768x1344",
+      "864x1152",
+      "1344x768",
+      "1152x864",
+      "1440x720",
+      "720x1440",
+    ];
+  }
+  return [];
+}
+
+export function supportsCustomSize(model: string): boolean {
+  return getModelSizes(model).length > 0;
+}
+
 export function showPlugins(provider: ServiceProvider, model: string) {
   if (
     provider == ServiceProvider.OpenAI ||

app/utils/chat.ts

@@ -138,7 +138,7 @@ export function uploadImage(file: Blob): Promise<string> {
   })
     .then((res) => res.json())
     .then((res) => {
-      console.log("res", res);
+      // console.log("res", res);
       if (res?.code == 0 && res?.data) {
         return res?.data;
       }

app/utils/model.ts

@@ -37,6 +37,17 @@ const sortModelTable = (models: ReturnType<typeof collectModels>) =>
     }
   });
 
+/**
+ * get model name and provider from a formatted string,
+ * e.g. `gpt-4@OpenAi` or `claude-3-5-sonnet@20240620@Google`
+ * @param modelWithProvider model name with provider separated by last `@` char,
+ * @returns [model, provider] tuple, if no `@` char found, provider is undefined
+ */
+export function getModelProvider(modelWithProvider: string): [string, string?] {
+  const [model, provider] = modelWithProvider.split(/@(?!.*@)/);
+  return [model, provider];
+}
+
 export function collectModelTable(
   models: readonly LLMModel[],
   customModels: string,
@@ -79,10 +90,10 @@ export function collectModelTable(
         );
       } else {
         // 1. find model by name, and set available value
-        const [customModelName, customProviderName] = name.split("@");
+        const [customModelName, customProviderName] = getModelProvider(name);
         let count = 0;
         for (const fullName in modelTable) {
-          const [modelName, providerName] = fullName.split("@");
+          const [modelName, providerName] = getModelProvider(fullName);
           if (
             customModelName == modelName &&
             (customProviderName === undefined ||
@@ -102,7 +113,7 @@ export function collectModelTable(
         }
         // 2. if model not exists, create new model with available value
         if (count === 0) {
-          let [customModelName, customProviderName] = name.split("@");
+          let [customModelName, customProviderName] = getModelProvider(name);
           const provider = customProvider(
             customProviderName || customModelName,
           );
@@ -139,7 +150,7 @@ export function collectModelTableWithDefaultModel(
       for (const key of Object.keys(modelTable)) {
         if (
           modelTable[key].available &&
-          key.split("@").shift() == defaultModel
+          getModelProvider(key)[0] == defaultModel
         ) {
           modelTable[key].isDefault = true;
           break;

package.json

@@ -27,6 +27,7 @@
     "@vercel/analytics": "^0.1.11",
     "@vercel/speed-insights": "^1.0.2",
     "axios": "^1.7.5",
+    "clsx": "^2.1.1",
     "emoji-picker-react": "^4.9.2",
     "fuse.js": "^7.0.0",
     "heic2any": "^0.0.4",
@@ -51,14 +52,15 @@
     "sass": "^1.59.2",
     "spark-md5": "^3.0.2",
     "use-debounce": "^9.0.4",
-    "zustand": "^4.3.8"
+    "zustand": "^4.3.8",
+    "rt-client": "https://github.com/Azure-Samples/aoai-realtime-audio-sdk/releases/download/js/v0.5.0/rt-client-0.5.0.tgz"
   },
   "devDependencies": {
     "@tauri-apps/api": "^1.6.0",
     "@tauri-apps/cli": "1.5.11",
     "@testing-library/dom": "^10.4.0",
-    "@testing-library/jest-dom": "^6.6.2",
+    "@testing-library/jest-dom": "^6.6.3",
-    "@testing-library/react": "^16.0.1",
+    "@testing-library/react": "^16.1.0",
     "@types/jest": "^29.5.14",
     "@types/js-yaml": "4.0.9",
     "@types/lodash-es": "^4.17.12",

public/audio-processor.js

@@ -0,0 +1,48 @@
+// @ts-nocheck
+class AudioRecorderProcessor extends AudioWorkletProcessor {
+  constructor() {
+    super();
+    this.isRecording = false;
+    this.bufferSize = 2400; // 100ms at 24kHz
+    this.currentBuffer = [];
+
+    this.port.onmessage = (event) => {
+      if (event.data.command === "START_RECORDING") {
+        this.isRecording = true;
+      } else if (event.data.command === "STOP_RECORDING") {
+        this.isRecording = false;
+
+        if (this.currentBuffer.length > 0) {
+          this.sendBuffer();
+        }
+      }
+    };
+  }
+
+  sendBuffer() {
+    if (this.currentBuffer.length > 0) {
+      const audioData = new Float32Array(this.currentBuffer);
+      this.port.postMessage({
+        eventType: "audio",
+        audioData: audioData,
+      });
+      this.currentBuffer = [];
+    }
+  }
+
+  process(inputs) {
+    const input = inputs[0];
+    if (input.length > 0 && this.isRecording) {
+      const audioData = input[0];
+
+      this.currentBuffer.push(...audioData);
+
+      if (this.currentBuffer.length >= this.bufferSize) {
+        this.sendBuffer();
+      }
+    }
+    return true;
+  }
+}
+
+registerProcessor("audio-recorder-processor", AudioRecorderProcessor);

src-tauri/tauri.conf.json

@@ -9,7 +9,7 @@
   },
   "package": {
     "productName": "NextChat",
-    "version": "2.15.7"
+    "version": "2.15.8"
   },
   "tauri": {
     "allowlist": {

test/model-provider.test.ts

@@ -0,0 +1,31 @@
+import { getModelProvider } from "../app/utils/model";
+
+describe("getModelProvider", () => {
+  test("should return model and provider when input contains '@'", () => {
+    const input = "model@provider";
+    const [model, provider] = getModelProvider(input);
+    expect(model).toBe("model");
+    expect(provider).toBe("provider");
+  });
+
+  test("should return model and undefined provider when input does not contain '@'", () => {
+    const input = "model";
+    const [model, provider] = getModelProvider(input);
+    expect(model).toBe("model");
+    expect(provider).toBeUndefined();
+  });
+
+  test("should handle multiple '@' characters correctly", () => {
+    const input = "model@provider@extra";
+    const [model, provider] = getModelProvider(input);
+    expect(model).toBe("model@provider");
+    expect(provider).toBe("extra");
+  });
+
+  test("should return empty strings when input is empty", () => {
+    const input = "";
+    const [model, provider] = getModelProvider(input);
+    expect(model).toBe("");
+    expect(provider).toBeUndefined();
+  });
+});

test/vision-model-checker.test.ts

@@ -0,0 +1,67 @@
+import { isVisionModel } from "../app/utils";
+
+describe("isVisionModel", () => {
+  const originalEnv = process.env;
+
+  beforeEach(() => {
+    jest.resetModules();
+    process.env = { ...originalEnv };
+  });
+
+  afterEach(() => {
+    process.env = originalEnv;
+  });
+
+  test("should identify vision models using regex patterns", () => {
+    const visionModels = [
+      "gpt-4-vision",
+      "claude-3-opus",
+      "gemini-1.5-pro",
+      "gemini-2.0",
+      "gemini-exp-vision",
+      "learnlm-vision",
+      "qwen-vl-max",
+      "qwen2-vl-max",
+      "gpt-4-turbo",
+      "dall-e-3",
+    ];
+
+    visionModels.forEach((model) => {
+      expect(isVisionModel(model)).toBe(true);
+    });
+  });
+
+  test("should exclude specific models", () => {
+    expect(isVisionModel("claude-3-5-haiku-20241022")).toBe(false);
+  });
+
+  test("should not identify non-vision models", () => {
+    const nonVisionModels = [
+      "gpt-3.5-turbo",
+      "gpt-4-turbo-preview",
+      "claude-2",
+      "regular-model",
+    ];
+
+    nonVisionModels.forEach((model) => {
+      expect(isVisionModel(model)).toBe(false);
+    });
+  });
+
+  test("should identify models from VISION_MODELS env var", () => {
+    process.env.VISION_MODELS = "custom-vision-model,another-vision-model";
+    
+    expect(isVisionModel("custom-vision-model")).toBe(true);
+    expect(isVisionModel("another-vision-model")).toBe(true);
+    expect(isVisionModel("unrelated-model")).toBe(false);
+  });
+
+  test("should handle empty or missing VISION_MODELS", () => {
+    process.env.VISION_MODELS = "";
+    expect(isVisionModel("unrelated-model")).toBe(false);
+
+    delete process.env.VISION_MODELS;
+    expect(isVisionModel("unrelated-model")).toBe(false);
+    expect(isVisionModel("gpt-4-vision")).toBe(true);
+  });
+});

yarn.lock

@@ -2114,10 +2114,10 @@
     lz-string "^1.5.0"
     pretty-format "^27.0.2"
 
-"@testing-library/jest-dom@^6.6.2":
+"@testing-library/jest-dom@^6.6.3":
-  version "6.6.2"
+  version "6.6.3"
-  resolved "https://registry.yarnpkg.com/@testing-library/jest-dom/-/jest-dom-6.6.2.tgz#8186aa9a07263adef9cc5a59a4772db8c31f4a5b"
+  resolved "https://registry.yarnpkg.com/@testing-library/jest-dom/-/jest-dom-6.6.3.tgz#26ba906cf928c0f8172e182c6fe214eb4f9f2bd2"
-  integrity sha512-P6GJD4yqc9jZLbe98j/EkyQDTPgqftohZF5FBkHY5BUERZmcf4HeO2k0XaefEg329ux2p21i1A1DmyQ1kKw2Jw==
+  integrity sha512-IteBhl4XqYNkM54f4ejhLRJiZNqcSCoXUOG2CPK7qbD322KjQozM4kHQOfkG2oln9b9HTYqs+Sae8vBATubxxA==
   dependencies:
     "@adobe/css-tools" "^4.4.0"
     aria-query "^5.0.0"
@@ -2127,10 +2127,10 @@
     lodash "^4.17.21"
     redent "^3.0.0"
 
-"@testing-library/react@^16.0.1":
+"@testing-library/react@^16.1.0":
-  version "16.0.1"
+  version "16.1.0"
-  resolved "https://registry.yarnpkg.com/@testing-library/react/-/react-16.0.1.tgz#29c0ee878d672703f5e7579f239005e4e0faa875"
+  resolved "https://registry.yarnpkg.com/@testing-library/react/-/react-16.1.0.tgz#aa0c61398bac82eaf89776967e97de41ac742d71"
-  integrity sha512-dSmwJVtJXmku+iocRhWOUFbrERC76TX2Mnf0ATODz8brzAZrMBbzLwQixlBSanZxR6LddK3eiwpSFZgDET1URg==
+  integrity sha512-Q2ToPvg0KsVL0ohND9A3zLJWcOXXcO8IDu3fj11KhNt0UlCWyFyvnCIBkd12tidB2lkiVRG8VFqdhcqhqnAQtg==
   dependencies:
     "@babel/runtime" "^7.12.5"
 
@@ -3189,6 +3189,11 @@ cliui@^8.0.1:
     strip-ansi "^6.0.1"
     wrap-ansi "^7.0.0"
 
+clsx@^2.1.1:
+  version "2.1.1"
+  resolved "https://registry.yarnpkg.com/clsx/-/clsx-2.1.1.tgz#eed397c9fd8bd882bfb18deab7102049a2f32999"
+  integrity sha512-eYm0QWBtUrBWZWG0d386OGAw16Z995PiOVo2B7bjWSbHedGl5e0ZWaq65kOGgUSNesEIDkB9ISbTg/JK9dhCZA==
+
 co@^4.6.0:
   version "4.6.0"
   resolved "https://registry.npmmirror.com/co/-/co-4.6.0.tgz#6ea6bdf3d853ae54ccb8e47bfa0bf3f9031fb184"
@@ -7450,6 +7455,12 @@ robust-predicates@^3.0.0:
   resolved "https://registry.npmmirror.com/robust-predicates/-/robust-predicates-3.0.1.tgz#ecde075044f7f30118682bd9fb3f123109577f9a"
   integrity sha512-ndEIpszUHiG4HtDsQLeIuMvRsDnn8c8rYStabochtUeCvfuvNptb5TUbVD68LRAILPX7p9nqQGh4xJgn3EHS/g==
 
+"rt-client@https://github.com/Azure-Samples/aoai-realtime-audio-sdk/releases/download/js/v0.5.0/rt-client-0.5.0.tgz":
+  version "0.5.0"
+  resolved "https://github.com/Azure-Samples/aoai-realtime-audio-sdk/releases/download/js/v0.5.0/rt-client-0.5.0.tgz#abf2e9a850201e3571b8d36830f77bc52af3de9b"
+  dependencies:
+    ws "^8.18.0"
+
 run-parallel@^1.1.9:
   version "1.2.0"
   resolved "https://registry.yarnpkg.com/run-parallel/-/run-parallel-1.2.0.tgz#66d1368da7bdf921eb9d95bd1a9229e7f21a43ee"
@@ -8493,9 +8504,9 @@ write-file-atomic@^4.0.2:
     imurmurhash "^0.1.4"
     signal-exit "^3.0.7"
 
-ws@^8.11.0:
+ws@^8.11.0, ws@^8.18.0:
   version "8.18.0"
-  resolved "https://registry.npmmirror.com/ws/-/ws-8.18.0.tgz#0d7505a6eafe2b0e712d232b42279f53bc289bbc"
+  resolved "https://registry.yarnpkg.com/ws/-/ws-8.18.0.tgz#0d7505a6eafe2b0e712d232b42279f53bc289bbc"
   integrity sha512-8VbfWfHLbbwu3+N6OKsOMpBdT4kXPDDB9cJk2bJ6mh9ucxdlnNvH1e+roYkKmN9Nxw2yjz7VzeO9oOz2zJ04Pw==
 
 xml-name-validator@^4.0.0: