diff --git a/README.md b/README.md index 0c06b73f0..dda896cbf 100644 --- a/README.md +++ b/README.md @@ -1,16 +1,17 @@
- - icon + + icon +

NextChat (ChatGPT Next Web)

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] @@ -18,9 +19,9 @@ One-Click to get a well-designed cross-platform ChatGPT web UI, with GPT3, GPT4 [![MacOS][MacOS-image]][download-url] [![Linux][Linux-image]][download-url] -[NextChatAI](https://nextchat.dev/chat?utm_source=readme) / [Web App](https://app.nextchat.dev) / [Desktop App](https://github.com/Yidadaa/ChatGPT-Next-Web/releases) / [Discord](https://discord.gg/YCkeafCafC) / [Enterprise Edition](#enterprise-edition) / [Twitter](https://twitter.com/NextChatDev) +[NextChatAI](https://nextchat.dev/chat?utm_source=readme) / [Web App Demo](https://app.nextchat.dev) / [Desktop App](https://github.com/Yidadaa/ChatGPT-Next-Web/releases) / [Discord](https://discord.gg/YCkeafCafC) / [Enterprise Edition](#enterprise-edition) / [Twitter](https://twitter.com/NextChatDev) -[NextChatAI](https://nextchat.dev/chat) / [网页版](https://app.nextchat.dev) / [客户端](https://github.com/Yidadaa/ChatGPT-Next-Web/releases) / [企业版](#%E4%BC%81%E4%B8%9A%E7%89%88) / [反馈](https://github.com/Yidadaa/ChatGPT-Next-Web/issues) +[NextChatAI](https://nextchat.dev/chat) / [自部署网页版](https://app.nextchat.dev) / [客户端](https://github.com/Yidadaa/ChatGPT-Next-Web/releases) / [企业版](#%E4%BC%81%E4%B8%9A%E7%89%88) / [反馈](https://github.com/Yidadaa/ChatGPT-Next-Web/issues) [saas-url]: https://nextchat.dev/chat?utm_source=readme [saas-image]: https://img.shields.io/badge/NextChat-Saas-green?logo=microsoftedge @@ -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 -[Deploy on Vercel](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) [Deploy on Zeabur](https://zeabur.com/templates/ZBUEFA) [Open in Gitpod](https://gitpod.io/#https://github.com/Yidadaa/ChatGPT-Next-Web) [BT Deply Install](https://www.bt.cn/new/download.html) [Deploy to Alibaba Cloud](https://computenest.aliyun.com/market/service-f1c9b75e59814dc49d52) +[Deploy on Vercel](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) [Deploy on Zeabur](https://zeabur.com/templates/ZBUEFA) [Open in Gitpod](https://gitpod.io/#https://github.com/Yidadaa/ChatGPT-Next-Web) [BT Deply Install](https://www.bt.cn/new/download.html) [](https://monica.im/?utm=nxcrp) @@ -311,6 +312,14 @@ ChatGLM Api Key. ChatGLM Api Url. +### `DEEPSEEK_API_KEY` (optional) + +DeepSeek Api Key. + +### `DEEPSEEK_URL` (optional) + +DeepSeek Api Url. + ### `HIDE_USER_API_KEY` (optional) > Default: Empty @@ -355,6 +364,13 @@ 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： diff --git a/README_CN.md b/README_CN.md index d4da8b9da..aa95d6b5c 100644 --- a/README_CN.md +++ b/README_CN.md @@ -192,6 +192,14 @@ ChatGLM Api Key. ChatGLM Api Url. +### `DEEPSEEK_API_KEY` (可选) + +DeepSeek Api Key. + +### `DEEPSEEK_URL` (可选) + +DeepSeek Api Url. + ### `HIDE_USER_API_KEY` （可选） @@ -235,6 +243,13 @@ 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 diff --git a/README_JA.md b/README_JA.md index 062c11262..29eb0d275 100644 --- a/README_JA.md +++ b/README_JA.md @@ -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。 diff --git a/app/api/[provider]/[...path]/route.ts b/app/api/[provider]/[...path]/route.ts index 3017fd371..3b5833d7e 100644 --- a/app/api/[provider]/[...path]/route.ts +++ b/app/api/[provider]/[...path]/route.ts @@ -10,6 +10,7 @@ import { handle as alibabaHandler } from "../../alibaba"; import { handle as moonshotHandler } from "../../moonshot"; import { handle as stabilityHandler } from "../../stability"; import { handle as iflytekHandler } from "../../iflytek"; +import { handle as deepseekHandler } from "../../deepseek"; import { handle as xaiHandler } from "../../xai"; import { handle as chatglmHandler } from "../../glm"; import { handle as proxyHandler } from "../../proxy"; @@ -40,6 +41,8 @@ async function handle( return stabilityHandler(req, { params }); case ApiPath.Iflytek: return iflytekHandler(req, { params }); + case ApiPath.DeepSeek: + return deepseekHandler(req, { params }); case ApiPath.XAI: return xaiHandler(req, { params }); case ApiPath.ChatGLM: diff --git a/app/api/alibaba.ts b/app/api/alibaba.ts index 894b1ae4c..20f6caefa 100644 --- a/app/api/alibaba.ts +++ b/app/api/alibaba.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; const serverConfig = getServerSideConfig(); @@ -89,7 +89,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.Alibaba as string, diff --git a/app/api/anthropic.ts b/app/api/anthropic.ts index 7a4444371..b96637b2c 100644 --- a/app/api/anthropic.ts +++ b/app/api/anthropic.ts @@ -9,7 +9,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "./auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; import { cloudflareAIGatewayUrl } from "@/app/utils/cloudflare"; const ALLOWD_PATH = new Set([Anthropic.ChatPath, Anthropic.ChatPath1]); @@ -122,7 +122,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.Anthropic as string, diff --git a/app/api/auth.ts b/app/api/auth.ts index 6703b64bd..1760c249c 100644 --- a/app/api/auth.ts +++ b/app/api/auth.ts @@ -92,6 +92,9 @@ export function auth(req: NextRequest, modelProvider: ModelProvider) { systemApiKey = serverConfig.iflytekApiKey + ":" + serverConfig.iflytekApiSecret; break; + case ModelProvider.DeepSeek: + systemApiKey = serverConfig.deepseekApiKey; + break; case ModelProvider.XAI: systemApiKey = serverConfig.xaiApiKey; break; diff --git a/app/api/baidu.ts b/app/api/baidu.ts index 0408b43c5..0f4e05ee8 100644 --- a/app/api/baidu.ts +++ b/app/api/baidu.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; import { getAccessToken } from "@/app/utils/baidu"; const serverConfig = getServerSideConfig(); @@ -104,7 +104,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.Baidu as string, diff --git a/app/api/bytedance.ts b/app/api/bytedance.ts index cb65b1061..51b39ceb7 100644 --- a/app/api/bytedance.ts +++ b/app/api/bytedance.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; const serverConfig = getServerSideConfig(); @@ -88,7 +88,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.ByteDance as string, diff --git a/app/api/common.ts b/app/api/common.ts index 495a12ccd..b7e41fa26 100644 --- a/app/api/common.ts +++ b/app/api/common.ts @@ -2,7 +2,7 @@ import { NextRequest, NextResponse } from "next/server"; import { getServerSideConfig } from "../config/server"; import { OPENAI_BASE_URL, ServiceProvider } from "../constant"; import { cloudflareAIGatewayUrl } from "../utils/cloudflare"; -import { getModelProvider, isModelAvailableInServer } from "../utils/model"; +import { getModelProvider, isModelNotavailableInServer } from "../utils/model"; const serverConfig = getServerSideConfig(); @@ -118,15 +118,14 @@ export async function requestOpenai(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, - ServiceProvider.OpenAI as string, - ) || - isModelAvailableInServer( - serverConfig.customModels, - jsonBody?.model as string, - ServiceProvider.Azure as string, + [ + ServiceProvider.OpenAI, + ServiceProvider.Azure, + jsonBody?.model as string, // support provider-unspecified model + ], ) ) { return NextResponse.json( diff --git a/app/api/config/route.ts b/app/api/config/route.ts index b0d9da031..855a5db01 100644 --- a/app/api/config/route.ts +++ b/app/api/config/route.ts @@ -14,6 +14,7 @@ const DANGER_CONFIG = { disableFastLink: serverConfig.disableFastLink, customModels: serverConfig.customModels, defaultModel: serverConfig.defaultModel, + visionModels: serverConfig.visionModels, }; declare global { diff --git a/app/api/deepseek.ts b/app/api/deepseek.ts new file mode 100644 index 000000000..a9879eced --- /dev/null +++ b/app/api/deepseek.ts @@ -0,0 +1,128 @@ +import { getServerSideConfig } from "@/app/config/server"; +import { + DEEPSEEK_BASE_URL, + ApiPath, + ModelProvider, + ServiceProvider, +} from "@/app/constant"; +import { prettyObject } from "@/app/utils/format"; +import { NextRequest, NextResponse } from "next/server"; +import { auth } from "@/app/api/auth"; +import { isModelNotavailableInServer } from "@/app/utils/model"; + +const serverConfig = getServerSideConfig(); + +export async function handle( + req: NextRequest, + { params }: { params: { path: string[] } }, +) { + console.log("[DeepSeek Route] params ", params); + + if (req.method === "OPTIONS") { + return NextResponse.json({ body: "OK" }, { status: 200 }); + } + + const authResult = auth(req, ModelProvider.DeepSeek); + if (authResult.error) { + return NextResponse.json(authResult, { + status: 401, + }); + } + + try { + const response = await request(req); + return response; + } catch (e) { + console.error("[DeepSeek] ", e); + return NextResponse.json(prettyObject(e)); + } +} + +async function request(req: NextRequest) { + const controller = new AbortController(); + + // alibaba use base url or just remove the path + let path = `${req.nextUrl.pathname}`.replaceAll(ApiPath.DeepSeek, ""); + + let baseUrl = serverConfig.deepseekUrl || DEEPSEEK_BASE_URL; + + if (!baseUrl.startsWith("http")) { + baseUrl = `https://${baseUrl}`; + } + + if (baseUrl.endsWith("/")) { + baseUrl = baseUrl.slice(0, -1); + } + + console.log("[Proxy] ", path); + console.log("[Base Url]", baseUrl); + + const timeoutId = setTimeout( + () => { + controller.abort(); + }, + 10 * 60 * 1000, + ); + + const fetchUrl = `${baseUrl}${path}`; + const fetchOptions: RequestInit = { + headers: { + "Content-Type": "application/json", + Authorization: req.headers.get("Authorization") ?? "", + }, + method: req.method, + body: req.body, + redirect: "manual", + // @ts-ignore + duplex: "half", + signal: controller.signal, + }; + + // #1815 try to refuse some request to some models + if (serverConfig.customModels && req.body) { + try { + const clonedBody = await req.text(); + fetchOptions.body = clonedBody; + + const jsonBody = JSON.parse(clonedBody) as { model?: string }; + + // not undefined and is false + if ( + isModelNotavailableInServer( + serverConfig.customModels, + jsonBody?.model as string, + ServiceProvider.DeepSeek as string, + ) + ) { + return NextResponse.json( + { + error: true, + message: `you are not allowed to use ${jsonBody?.model} model`, + }, + { + status: 403, + }, + ); + } + } catch (e) { + console.error(`[DeepSeek] filter`, e); + } + } + try { + const res = await fetch(fetchUrl, fetchOptions); + + // to prevent browser prompt for credentials + const newHeaders = new Headers(res.headers); + newHeaders.delete("www-authenticate"); + // to disable nginx buffering + newHeaders.set("X-Accel-Buffering", "no"); + + return new Response(res.body, { + status: res.status, + statusText: res.statusText, + headers: newHeaders, + }); + } finally { + clearTimeout(timeoutId); + } +} diff --git a/app/api/glm.ts b/app/api/glm.ts index 3625b9f7b..8431c5db5 100644 --- a/app/api/glm.ts +++ b/app/api/glm.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; const serverConfig = getServerSideConfig(); @@ -89,7 +89,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.ChatGLM as string, diff --git a/app/api/iflytek.ts b/app/api/iflytek.ts index 8b8227dce..6624f74e9 100644 --- a/app/api/iflytek.ts +++ b/app/api/iflytek.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; // iflytek const serverConfig = getServerSideConfig(); @@ -89,7 +89,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.Iflytek as string, diff --git a/app/api/moonshot.ts b/app/api/moonshot.ts index 5bf4807e3..792d14d33 100644 --- a/app/api/moonshot.ts +++ b/app/api/moonshot.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; const serverConfig = getServerSideConfig(); @@ -88,7 +88,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.Moonshot as string, diff --git a/app/api/xai.ts b/app/api/xai.ts index a4ee8b397..4aad5e5fb 100644 --- a/app/api/xai.ts +++ b/app/api/xai.ts @@ -8,7 +8,7 @@ import { import { prettyObject } from "@/app/utils/format"; import { NextRequest, NextResponse } from "next/server"; import { auth } from "@/app/api/auth"; -import { isModelAvailableInServer } from "@/app/utils/model"; +import { isModelNotavailableInServer } from "@/app/utils/model"; const serverConfig = getServerSideConfig(); @@ -88,7 +88,7 @@ async function request(req: NextRequest) { // not undefined and is false if ( - isModelAvailableInServer( + isModelNotavailableInServer( serverConfig.customModels, jsonBody?.model as string, ServiceProvider.XAI as string, diff --git a/app/client/api.ts b/app/client/api.ts index 1da81e964..8f263763b 100644 --- a/app/client/api.ts +++ b/app/client/api.ts @@ -20,6 +20,7 @@ import { QwenApi } from "./platforms/alibaba"; import { HunyuanApi } from "./platforms/tencent"; import { MoonshotApi } from "./platforms/moonshot"; import { SparkApi } from "./platforms/iflytek"; +import { DeepSeekApi } from "./platforms/deepseek"; import { XAIApi } from "./platforms/xai"; import { ChatGLMApi } from "./platforms/glm"; @@ -154,6 +155,9 @@ export class ClientApi { case ModelProvider.Iflytek: this.llm = new SparkApi(); break; + case ModelProvider.DeepSeek: + this.llm = new DeepSeekApi(); + break; case ModelProvider.XAI: this.llm = new XAIApi(); break; @@ -247,6 +251,7 @@ export function getHeaders(ignoreHeaders: boolean = false) { const isAlibaba = modelConfig.providerName === ServiceProvider.Alibaba; const isMoonshot = modelConfig.providerName === ServiceProvider.Moonshot; const isIflytek = modelConfig.providerName === ServiceProvider.Iflytek; + const isDeepSeek = modelConfig.providerName === ServiceProvider.DeepSeek; const isXAI = modelConfig.providerName === ServiceProvider.XAI; const isChatGLM = modelConfig.providerName === ServiceProvider.ChatGLM; const isEnabledAccessControl = accessStore.enabledAccessControl(); @@ -264,6 +269,8 @@ export function getHeaders(ignoreHeaders: boolean = false) { ? accessStore.moonshotApiKey : isXAI ? accessStore.xaiApiKey + : isDeepSeek + ? accessStore.deepseekApiKey : isChatGLM ? accessStore.chatglmApiKey : isIflytek @@ -280,6 +287,7 @@ export function getHeaders(ignoreHeaders: boolean = false) { isAlibaba, isMoonshot, isIflytek, + isDeepSeek, isXAI, isChatGLM, apiKey, @@ -302,6 +310,13 @@ export function getHeaders(ignoreHeaders: boolean = false) { isAzure, isAnthropic, isBaidu, + isByteDance, + isAlibaba, + isMoonshot, + isIflytek, + isDeepSeek, + isXAI, + isChatGLM, apiKey, isEnabledAccessControl, } = getConfig(); @@ -344,6 +359,8 @@ export function getClientApi(provider: ServiceProvider): ClientApi { return new ClientApi(ModelProvider.Moonshot); case ServiceProvider.Iflytek: return new ClientApi(ModelProvider.Iflytek); + case ServiceProvider.DeepSeek: + return new ClientApi(ModelProvider.DeepSeek); case ServiceProvider.XAI: return new ClientApi(ModelProvider.XAI); case ServiceProvider.ChatGLM: diff --git a/app/client/platforms/deepseek.ts b/app/client/platforms/deepseek.ts new file mode 100644 index 000000000..e2ae645c6 --- /dev/null +++ b/app/client/platforms/deepseek.ts @@ -0,0 +1,200 @@ +"use client"; +// azure and openai, using same models. so using same LLMApi. +import { + ApiPath, + DEEPSEEK_BASE_URL, + DeepSeek, + REQUEST_TIMEOUT_MS, +} from "@/app/constant"; +import { + useAccessStore, + useAppConfig, + useChatStore, + ChatMessageTool, + usePluginStore, +} from "@/app/store"; +import { stream } from "@/app/utils/chat"; +import { + ChatOptions, + getHeaders, + LLMApi, + LLMModel, + SpeechOptions, +} from "../api"; +import { getClientConfig } from "@/app/config/client"; +import { getMessageTextContent } from "@/app/utils"; +import { RequestPayload } from "./openai"; +import { fetch } from "@/app/utils/stream"; + +export class DeepSeekApi implements LLMApi { + private disableListModels = true; + + path(path: string): string { + const accessStore = useAccessStore.getState(); + + let baseUrl = ""; + + if (accessStore.useCustomConfig) { + baseUrl = accessStore.deepseekUrl; + } + + if (baseUrl.length === 0) { + const isApp = !!getClientConfig()?.isApp; + const apiPath = ApiPath.DeepSeek; + baseUrl = isApp ? DEEPSEEK_BASE_URL : apiPath; + } + + if (baseUrl.endsWith("/")) { + baseUrl = baseUrl.slice(0, baseUrl.length - 1); + } + if (!baseUrl.startsWith("http") && !baseUrl.startsWith(ApiPath.DeepSeek)) { + baseUrl = "https://" + baseUrl; + } + + console.log("[Proxy Endpoint] ", baseUrl, path); + + return [baseUrl, path].join("/"); + } + + extractMessage(res: any) { + return res.choices?.at(0)?.message?.content ?? ""; + } + + speech(options: SpeechOptions): Promise { + throw new Error("Method not implemented."); + } + + async chat(options: ChatOptions) { + const messages: ChatOptions["messages"] = []; + for (const v of options.messages) { + const content = getMessageTextContent(v); + messages.push({ role: v.role, content }); + } + + const modelConfig = { + ...useAppConfig.getState().modelConfig, + ...useChatStore.getState().currentSession().mask.modelConfig, + ...{ + model: options.config.model, + providerName: options.config.providerName, + }, + }; + + const requestPayload: RequestPayload = { + 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, + // max_tokens: Math.max(modelConfig.max_tokens, 1024), + // Please do not ask me why not send max_tokens, no reason, this param is just shit, I dont want to explain anymore. + }; + + console.log("[Request] openai payload: ", requestPayload); + + const shouldStream = !!options.config.stream; + const controller = new AbortController(); + options.onController?.(controller); + + try { + const chatPath = this.path(DeepSeek.ChatPath); + const chatPayload = { + method: "POST", + body: JSON.stringify(requestPayload), + signal: controller.signal, + headers: getHeaders(), + }; + + // make a fetch request + const requestTimeoutId = setTimeout( + () => controller.abort(), + REQUEST_TIMEOUT_MS, + ); + + if (shouldStream) { + const [tools, funcs] = usePluginStore + .getState() + .getAsTools( + useChatStore.getState().currentSession().mask?.plugin || [], + ); + return stream( + chatPath, + requestPayload, + getHeaders(), + tools as any, + funcs, + controller, + // parseSSE + (text: string, runTools: ChatMessageTool[]) => { + // console.log("parseSSE", text, runTools); + const json = JSON.parse(text); + const choices = json.choices as Array<{ + delta: { + content: string; + tool_calls: ChatMessageTool[]; + }; + }>; + const tool_calls = choices[0]?.delta?.tool_calls; + if (tool_calls?.length > 0) { + const index = tool_calls[0]?.index; + const id = tool_calls[0]?.id; + const args = tool_calls[0]?.function?.arguments; + if (id) { + runTools.push({ + id, + type: tool_calls[0]?.type, + function: { + name: tool_calls[0]?.function?.name as string, + arguments: args, + }, + }); + } else { + // @ts-ignore + runTools[index]["function"]["arguments"] += args; + } + } + return choices[0]?.delta?.content; + }, + // processToolMessage, include tool_calls message and tool call results + ( + requestPayload: RequestPayload, + toolCallMessage: any, + toolCallResult: any[], + ) => { + // @ts-ignore + requestPayload?.messages?.splice( + // @ts-ignore + requestPayload?.messages?.length, + 0, + toolCallMessage, + ...toolCallResult, + ); + }, + options, + ); + } else { + const res = await fetch(chatPath, chatPayload); + clearTimeout(requestTimeoutId); + + const resJson = await res.json(); + const message = this.extractMessage(resJson); + options.onFinish(message, res); + } + } catch (e) { + console.log("[Request] failed to make a chat request", e); + options.onError?.(e as Error); + } + } + async usage() { + return { + used: 0, + total: 0, + }; + } + + async models(): Promise { + return []; + } +} diff --git a/app/client/platforms/glm.ts b/app/client/platforms/glm.ts index a7965947f..a8d1869e3 100644 --- a/app/client/platforms/glm.ts +++ b/app/client/platforms/glm.ts @@ -21,16 +21,108 @@ import { SpeechOptions, } from "../api"; import { getClientConfig } from "@/app/config/client"; -import { getMessageTextContent } from "@/app/utils"; +import { getMessageTextContent, isVisionModel } from "@/app/utils"; import { RequestPayload } from "./openai"; import { fetch } from "@/app/utils/stream"; +import { preProcessImageContent } from "@/app/utils/chat"; + +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 ? `` : ""; + } + default: + return this.extractMessage(json); + } + } + path(path: string): string { const accessStore = useAccessStore.getState(); - let baseUrl = ""; if (accessStore.useCustomConfig) { @@ -51,7 +143,6 @@ export class ChatGLMApi implements LLMApi { } console.log("[Proxy Endpoint] ", baseUrl, path); - return [baseUrl, path].join("/"); } @@ -64,9 +155,12 @@ export class ChatGLMApi implements LLMApi { } async chat(options: ChatOptions) { + const visionModel = isVisionModel(options.config.model); const messages: ChatOptions["messages"] = []; for (const v of options.messages) { - const content = getMessageTextContent(v); + const content = visionModel + ? await preProcessImageContent(v.content) + : getMessageTextContent(v); messages.push({ role: v.role, content }); } @@ -78,25 +172,16 @@ export class ChatGLMApi implements LLMApi { providerName: options.config.providerName, }, }; + const modelType = this.getModelType(modelConfig.model); + const requestPayload = this.createPayload(messages, modelConfig, options); + const path = this.path(this.getModelPath(modelType)); - const requestPayload: RequestPayload = { - 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, - }; + console.log(`[Request] glm ${modelType} payload: `, requestPayload); - console.log("[Request] glm 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 +189,23 @@ 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 +213,7 @@ export class ChatGLMApi implements LLMApi { useChatStore.getState().currentSession().mask?.plugin || [], ); return stream( - chatPath, + path, requestPayload, getHeaders(), tools as any, @@ -125,7 +221,6 @@ 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 +249,7 @@ 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 +267,7 @@ 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 +279,7 @@ export class ChatGLMApi implements LLMApi { options.onError?.(e as Error); } } + async usage() { return { used: 0, diff --git a/app/client/platforms/google.ts b/app/client/platforms/google.ts index a7bce4fc2..5ca8e1071 100644 --- a/app/client/platforms/google.ts +++ b/app/client/platforms/google.ts @@ -60,9 +60,18 @@ export class GeminiProApi implements LLMApi { extractMessage(res: any) { console.log("[Response] gemini-pro response: ", res); + const getTextFromParts = (parts: any[]) => { + if (!Array.isArray(parts)) return ""; + + return parts + .map((part) => part?.text || "") + .filter((text) => text.trim() !== "") + .join("\n\n"); + }; + return ( - res?.candidates?.at(0)?.content?.parts.at(0)?.text || - res?.at(0)?.candidates?.at(0)?.content?.parts.at(0)?.text || + getTextFromParts(res?.candidates?.at(0)?.content?.parts) || + getTextFromParts(res?.at(0)?.candidates?.at(0)?.content?.parts) || res?.error?.message || "" ); @@ -223,7 +232,10 @@ export class GeminiProApi implements LLMApi { }, }); } - return chunkJson?.candidates?.at(0)?.content.parts.at(0)?.text; + return chunkJson?.candidates + ?.at(0) + ?.content.parts?.map((part: { text: string }) => part.text) + .join("\n\n"); }, // processToolMessage, include tool_calls message and tool call results ( diff --git a/app/client/platforms/openai.ts b/app/client/platforms/openai.ts index 15cfb7ca6..5a110b84b 100644 --- a/app/client/platforms/openai.ts +++ b/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, @@ -73,7 +73,7 @@ export interface DalleRequestPayload { prompt: string; response_format: "url" | "b64_json"; n: number; - size: DalleSize; + size: ModelSize; quality: DalleQuality; style: DalleStyle; } diff --git a/app/components/chat.tsx b/app/components/chat.tsx index 51fe74fe7..9990a359e 100644 --- a/app/components/chat.tsx +++ b/app/components/chat.tsx @@ -72,6 +72,8 @@ import { isDalle3, showPlugins, safeLocalStorage, + getModelSizes, + supportsCustomSize, } from "../utils"; import { uploadImage as uploadImageRemote } from "@/app/utils/chat"; @@ -79,7 +81,7 @@ import { uploadImage as uploadImageRemote } from "@/app/utils/chat"; import dynamic from "next/dynamic"; import { ChatControllerPool } from "../client/controller"; -import { DalleSize, DalleQuality, DalleStyle } from "../typing"; +import { DalleQuality, DalleStyle, ModelSize } from "../typing"; import { Prompt, usePromptStore } from "../store/prompt"; import Locale from "../locales"; @@ -519,10 +521,11 @@ export function ChatActions(props: { const [showSizeSelector, setShowSizeSelector] = useState(false); const [showQualitySelector, setShowQualitySelector] = useState(false); const [showStyleSelector, setShowStyleSelector] = useState(false); - const dalle3Sizes: DalleSize[] = ["1024x1024", "1792x1024", "1024x1792"]; + const modelSizes = getModelSizes(currentModel); const dalle3Qualitys: DalleQuality[] = ["standard", "hd"]; const dalle3Styles: DalleStyle[] = ["vivid", "natural"]; - const currentSize = session.mask.modelConfig?.size ?? "1024x1024"; + const currentSize = + session.mask.modelConfig?.size ?? ("1024x1024" as ModelSize); const currentQuality = session.mask.modelConfig?.quality ?? "standard"; const currentStyle = session.mask.modelConfig?.style ?? "vivid"; @@ -673,7 +676,7 @@ export function ChatActions(props: { /> )} - {isDalle3(currentModel) && ( + {supportsCustomSize(currentModel) && ( setShowSizeSelector(true)} text={currentSize} @@ -684,7 +687,7 @@ export function ChatActions(props: { {showSizeSelector && ( ({ + items={modelSizes.map((m) => ({ title: m, value: m, }))} @@ -897,6 +900,12 @@ export function ShortcutKeyModal(props: { onClose: () => void }) { title: Locale.Chat.ShortcutKey.showShortcutKey, keys: isMac ? ["⌘", "/"] : ["Ctrl", "/"], }, + { + title: Locale.Chat.ShortcutKey.clearContext, + keys: isMac + ? ["⌘", "Shift", "backspace"] + : ["Ctrl", "Shift", "backspace"], + }, ]; return (
@@ -1549,7 +1558,7 @@ function _Chat() { const [showShortcutKeyModal, setShowShortcutKeyModal] = useState(false); useEffect(() => { - const handleKeyDown = (event: any) => { + const handleKeyDown = (event: KeyboardEvent) => { // 打开新聊天 command + shift + o if ( (event.metaKey || event.ctrlKey) && @@ -1600,14 +1609,30 @@ function _Chat() { event.preventDefault(); setShowShortcutKeyModal(true); } + // 清除上下文 command + shift + backspace + else if ( + (event.metaKey || event.ctrlKey) && + event.shiftKey && + event.key.toLowerCase() === "backspace" + ) { + event.preventDefault(); + chatStore.updateTargetSession(session, (session) => { + if (session.clearContextIndex === session.messages.length) { + session.clearContextIndex = undefined; + } else { + session.clearContextIndex = session.messages.length; + session.memoryPrompt = ""; // will clear memory + } + }); + } }; - window.addEventListener("keydown", handleKeyDown); + document.addEventListener("keydown", handleKeyDown); return () => { - window.removeEventListener("keydown", handleKeyDown); + document.removeEventListener("keydown", handleKeyDown); }; - }, [messages, chatStore, navigate]); + }, [messages, chatStore, navigate, session]); const [showChatSidePanel, setShowChatSidePanel] = useState(false); diff --git a/app/components/settings.tsx b/app/components/settings.tsx index a74ff17b1..3b990ed2c 100644 --- a/app/components/settings.tsx +++ b/app/components/settings.tsx @@ -73,6 +73,7 @@ import { Iflytek, SAAS_CHAT_URL, ChatGLM, + DeepSeek, } from "../constant"; import { Prompt, SearchService, usePromptStore } from "../store/prompt"; import { ErrorBoundary } from "./error"; @@ -1197,6 +1198,47 @@ export function Settings() { ); + const deepseekConfigComponent = accessStore.provider === + ServiceProvider.DeepSeek && ( + <> + + + accessStore.update( + (access) => (access.deepseekUrl = e.currentTarget.value), + ) + } + > + + + { + accessStore.update( + (access) => (access.deepseekApiKey = e.currentTarget.value), + ); + }} + /> + + + ); + const XAIConfigComponent = accessStore.provider === ServiceProvider.XAI && ( <> (await import("./chat-list")).ChatList, { loading: () => null, }); @@ -219,7 +224,7 @@ export function SideBarTail(props: { export function SideBar(props: { className?: string }) { useHotKey(); const { onDragStart, shouldNarrow } = useDragSideBar(); - const [showPluginSelector, setShowPluginSelector] = useState(false); + const [showDiscoverySelector, setshowDiscoverySelector] = useState(false); const navigate = useNavigate(); const config = useAppConfig(); const chatStore = useChatStore(); @@ -254,21 +259,21 @@ export function SideBar(props: { className?: string }) { icon={} text={shouldNarrow ? undefined : Locale.Discovery.Name} className={styles["sidebar-bar-button"]} - onClick={() => setShowPluginSelector(true)} + onClick={() => setshowDiscoverySelector(true)} shadow />
- {showPluginSelector && ( + {showDiscoverySelector && ( { + ...DISCOVERY.map((item) => { return { title: item.name, value: item.path, }; }), ]} - onClose={() => setShowPluginSelector(false)} + onClose={() => setshowDiscoverySelector(false)} onSelection={(s) => { navigate(s[0], { state: { fromHome: true } }); }} diff --git a/app/config/server.ts b/app/config/server.ts index 9d6b3c2b8..73faa8815 100644 --- a/app/config/server.ts +++ b/app/config/server.ts @@ -1,5 +1,6 @@ import md5 from "spark-md5"; import { DEFAULT_MODELS, DEFAULT_GA_ID } from "../constant"; +import { isGPT4Model } from "../utils/model"; declare global { namespace NodeJS { @@ -22,6 +23,7 @@ declare global { DISABLE_FAST_LINK?: string; // disallow parse settings from url or not CUSTOM_MODELS?: string; // to control custom models DEFAULT_MODEL?: string; // to control default model in every new chat window + VISION_MODELS?: string; // to control vision models // stability only STABILITY_URL?: string; @@ -71,6 +73,9 @@ declare global { IFLYTEK_API_KEY?: string; IFLYTEK_API_SECRET?: string; + DEEPSEEK_URL?: string; + DEEPSEEK_API_KEY?: string; + // xai only XAI_URL?: string; XAI_API_KEY?: string; @@ -124,23 +129,16 @@ export const getServerSideConfig = () => { const disableGPT4 = !!process.env.DISABLE_GPT4; let customModels = process.env.CUSTOM_MODELS ?? ""; let defaultModel = process.env.DEFAULT_MODEL ?? ""; + let visionModels = process.env.VISION_MODELS ?? ""; if (disableGPT4) { if (customModels) customModels += ","; - customModels += DEFAULT_MODELS.filter( - (m) => - (m.name.startsWith("gpt-4") || m.name.startsWith("chatgpt-4o") || m.name.startsWith("o1")) && - !m.name.startsWith("gpt-4o-mini"), - ) + customModels += DEFAULT_MODELS.filter((m) => isGPT4Model(m.name)) .map((m) => "-" + m.name) .join(","); - if ( - (defaultModel.startsWith("gpt-4") || - defaultModel.startsWith("chatgpt-4o") || - defaultModel.startsWith("o1")) && - !defaultModel.startsWith("gpt-4o-mini") - ) + if (defaultModel && isGPT4Model(defaultModel)) { defaultModel = ""; + } } const isStability = !!process.env.STABILITY_API_KEY; @@ -155,6 +153,7 @@ export const getServerSideConfig = () => { const isAlibaba = !!process.env.ALIBABA_API_KEY; const isMoonshot = !!process.env.MOONSHOT_API_KEY; const isIflytek = !!process.env.IFLYTEK_API_KEY; + const isDeepSeek = !!process.env.DEEPSEEK_API_KEY; const isXAI = !!process.env.XAI_API_KEY; const isChatGLM = !!process.env.CHATGLM_API_KEY; // const apiKeyEnvVar = process.env.OPENAI_API_KEY ?? ""; @@ -219,6 +218,10 @@ export const getServerSideConfig = () => { iflytekApiKey: process.env.IFLYTEK_API_KEY, iflytekApiSecret: process.env.IFLYTEK_API_SECRET, + isDeepSeek, + deepseekUrl: process.env.DEEPSEEK_URL, + deepseekApiKey: getApiKey(process.env.DEEPSEEK_API_KEY), + isXAI, xaiUrl: process.env.XAI_URL, xaiApiKey: getApiKey(process.env.XAI_API_KEY), @@ -248,6 +251,7 @@ export const getServerSideConfig = () => { disableFastLink: !!process.env.DISABLE_FAST_LINK, customModels, defaultModel, + visionModels, allowedWebDavEndpoints, }; }; diff --git a/app/constant.ts b/app/constant.ts index 264fe0013..f34d407c1 100644 --- a/app/constant.ts +++ b/app/constant.ts @@ -28,6 +28,8 @@ export const TENCENT_BASE_URL = "https://hunyuan.tencentcloudapi.com"; export const MOONSHOT_BASE_URL = "https://api.moonshot.cn"; export const IFLYTEK_BASE_URL = "https://spark-api-open.xf-yun.com"; +export const DEEPSEEK_BASE_URL = "https://api.deepseek.com"; + export const XAI_BASE_URL = "https://api.x.ai"; export const CHATGLM_BASE_URL = "https://open.bigmodel.cn"; @@ -65,6 +67,7 @@ export enum ApiPath { Artifacts = "/api/artifacts", XAI = "/api/xai", ChatGLM = "/api/chatglm", + DeepSeek = "/api/deepseek", } export enum SlotID { @@ -119,6 +122,7 @@ export enum ServiceProvider { Iflytek = "Iflytek", XAI = "XAI", ChatGLM = "ChatGLM", + DeepSeek = "DeepSeek", } // Google API safety settings, see https://ai.google.dev/gemini-api/docs/safety-settings @@ -143,6 +147,7 @@ export enum ModelProvider { Iflytek = "Iflytek", XAI = "XAI", ChatGLM = "ChatGLM", + DeepSeek = "DeepSeek", } export const Stability = { @@ -225,6 +230,11 @@ export const Iflytek = { ChatPath: "v1/chat/completions", }; +export const DeepSeek = { + ExampleEndpoint: DEEPSEEK_BASE_URL, + ChatPath: "chat/completions", +}; + export const XAI = { ExampleEndpoint: XAI_BASE_URL, ChatPath: "v1/chat/completions", @@ -233,6 +243,8 @@ export const XAI = { export const ChatGLM = { ExampleEndpoint: CHATGLM_BASE_URL, 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 @@ -275,6 +287,8 @@ export const KnowledgeCutOffDate: Record = { // it's now easier to add "KnowledgeCutOffDate" instead of stupid hardcoding it, as was done previously. "gemini-pro": "2023-12", "gemini-pro-vision": "2023-12", + "deepseek-chat": "2024-07", + "deepseek-coder": "2024-07", }; export const DEFAULT_TTS_ENGINE = "Edge-TTS"; @@ -291,6 +305,23 @@ 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" + /glm-4v/, +]; + +export const EXCLUDE_VISION_MODEL_REGEXES = [/claude-3-5-haiku-20241022/]; + const openaiModels = [ "gpt-3.5-turbo", "gpt-3.5-turbo-1106", @@ -317,13 +348,23 @@ 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-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", - "learnlm-1.5-pro-experimental", - "gemini-pro-vision", + "gemini-exp-1206", + "gemini-2.0-flash-exp", + "gemini-2.0-flash-thinking-exp-1219", ]; const anthropicModels = [ @@ -394,6 +435,8 @@ const iflytekModels = [ "4.0Ultra", ]; +const deepseekModels = ["deepseek-chat", "deepseek-coder"]; + const xAIModes = ["grok-beta"]; const chatglmModels = [ @@ -405,6 +448,15 @@ 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 @@ -541,6 +593,17 @@ export const DEFAULT_MODELS = [ sorted: 12, }, })), + ...deepseekModels.map((name) => ({ + name, + available: true, + sorted: seq++, + provider: { + id: "deepseek", + providerName: "DeepSeek", + providerType: "deepseek", + sorted: 13, + }, + })), ] as const; export const CHAT_PAGE_SIZE = 15; @@ -560,11 +623,6 @@ export const internalAllowedWebDavEndpoints = [ ]; export const DEFAULT_GA_ID = "G-89WN60ZK2E"; -export const PLUGINS = [ - { name: "Plugins", path: Path.Plugins }, - { name: "Stable Diffusion", path: Path.Sd }, - { name: "Search Chat", path: Path.SearchChat }, -]; export const SAAS_CHAT_URL = "https://kaiyanmedical.com"; export const SAAS_CHAT_UTM_URL = "https://kaiyanmedical.com/chat?utm=github"; diff --git a/app/locales/cn.ts b/app/locales/cn.ts index 47be019a8..25f49be7d 100644 --- a/app/locales/cn.ts +++ b/app/locales/cn.ts @@ -106,6 +106,7 @@ const cn = { copyLastMessage: "复制最后一个回复", copyLastCode: "复制最后一个代码块", showShortcutKey: "显示快捷方式", + clearContext: "清除上下文", }, }, Export: { @@ -176,7 +177,7 @@ const cn = { }, }, Lang: { - Name: "Language", // ATTENTION: if you wanna add a new translation, please do not translate this value, leave it as `Language` + Name: "Language", // 注意：如果要添加新的翻译，请不要翻译此值，将它保留为 `Language` All: "所有语言", }, Avatar: "头像", @@ -462,6 +463,17 @@ const cn = { SubTitle: "样例：", }, }, + DeepSeek: { + ApiKey: { + Title: "接口密钥", + SubTitle: "使用自定义DeepSeek API Key", + Placeholder: "DeepSeek API Key", + }, + Endpoint: { + Title: "接口地址", + SubTitle: "样例：", + }, + }, XAI: { ApiKey: { Title: "接口密钥", @@ -630,7 +642,7 @@ const cn = { Sysmessage: "你是一个助手", }, SearchChat: { - Name: "搜索", + Name: "搜索聊天记录", Page: { Title: "搜索聊天记录", Search: "输入搜索关键词", diff --git a/app/locales/en.ts b/app/locales/en.ts index d3b684ae4..9b09101d4 100644 --- a/app/locales/en.ts +++ b/app/locales/en.ts @@ -103,6 +103,7 @@ const en: LocaleType = { copyLastMessage: "Copy Last Reply", copyLastCode: "Copy Last Code Block", showShortcutKey: "Show Shortcuts", + clearContext: "Clear Context", }, }, Export: { @@ -442,6 +443,17 @@ const en: LocaleType = { SubTitle: "Example: ", }, }, + DeepSeek: { + ApiKey: { + Title: "DeepSeek API Key", + SubTitle: "Use a custom DeepSeek API Key", + Placeholder: "DeepSeek API Key", + }, + Endpoint: { + Title: "Endpoint Address", + SubTitle: "Example: ", + }, + }, XAI: { ApiKey: { Title: "XAI API Key", diff --git a/app/locales/tw.ts b/app/locales/tw.ts index c800ad15d..83dd547b8 100644 --- a/app/locales/tw.ts +++ b/app/locales/tw.ts @@ -100,6 +100,7 @@ const tw = { copyLastMessage: "複製最後一個回覆", copyLastCode: "複製最後一個程式碼區塊", showShortcutKey: "顯示快捷方式", + clearContext: "清除上下文", }, }, Export: { @@ -485,7 +486,7 @@ const tw = { }, }, SearchChat: { - Name: "搜尋", + Name: "搜尋聊天記錄", Page: { Title: "搜尋聊天記錄", Search: "輸入搜尋關鍵詞", diff --git a/app/masks/cn.ts b/app/masks/cn.ts index ed507d734..64842f6e8 100644 --- a/app/masks/cn.ts +++ b/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: "", }, ], diff --git a/app/store/access.ts b/app/store/access.ts index 4796b2fe8..1fed5dfed 100644 --- a/app/store/access.ts +++ b/app/store/access.ts @@ -13,6 +13,7 @@ import { MOONSHOT_BASE_URL, STABILITY_BASE_URL, IFLYTEK_BASE_URL, + DEEPSEEK_BASE_URL, XAI_BASE_URL, CHATGLM_BASE_URL, } from "../constant"; @@ -47,6 +48,8 @@ const DEFAULT_STABILITY_URL = isApp ? STABILITY_BASE_URL : ApiPath.Stability; const DEFAULT_IFLYTEK_URL = isApp ? IFLYTEK_BASE_URL : ApiPath.Iflytek; +const DEFAULT_DEEPSEEK_URL = isApp ? DEEPSEEK_BASE_URL : ApiPath.DeepSeek; + const DEFAULT_XAI_URL = isApp ? XAI_BASE_URL : ApiPath.XAI; const DEFAULT_CHATGLM_URL = isApp ? CHATGLM_BASE_URL : ApiPath.ChatGLM; @@ -108,6 +111,10 @@ const DEFAULT_ACCESS_STATE = { iflytekApiKey: "", iflytekApiSecret: "", + // deepseek + deepseekUrl: DEFAULT_DEEPSEEK_URL, + deepseekApiKey: "", + // xai xaiUrl: DEFAULT_XAI_URL, xaiApiKey: "", @@ -124,6 +131,7 @@ const DEFAULT_ACCESS_STATE = { disableFastLink: false, customModels: "", defaultModel: "", + visionModels: "", // tts config edgeTTSVoiceName: "zh-CN-YunxiNeural", @@ -138,7 +146,10 @@ export const useAccessStore = createPersistStore( return get().needCode; }, - + getVisionModels() { + this.fetch(); + return get().visionModels; + }, edgeVoiceName() { this.fetch(); @@ -183,6 +194,9 @@ export const useAccessStore = createPersistStore( isValidIflytek() { return ensure(get(), ["iflytekApiKey"]); }, + isValidDeepSeek() { + return ensure(get(), ["deepseekApiKey"]); + }, isValidXAI() { return ensure(get(), ["xaiApiKey"]); @@ -207,6 +221,7 @@ export const useAccessStore = createPersistStore( this.isValidTencent() || this.isValidMoonshot() || this.isValidIflytek() || + this.isValidDeepSeek() || this.isValidXAI() || this.isValidChatGLM() || !this.enabledAccessControl() || diff --git a/app/store/config.ts b/app/store/config.ts index 4ed335272..9b08fb734 100644 --- a/app/store/config.ts +++ b/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, @@ -78,7 +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, }, diff --git a/app/typing.ts b/app/typing.ts index 0336be75d..ecb327936 100644 --- a/app/typing.ts +++ b/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"; diff --git a/app/utils.ts b/app/utils.ts index b62bc126d..4f5b7b0b7 100644 --- a/app/utils.ts +++ b/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 { useAccessStore } from "./store"; +import { ModelSize } from "./typing"; export function trimTopic(topic: string) { // Fix an issue where double quotes still show in the Indonesian language @@ -252,27 +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 excludeKeywords = ["claude-3-5-haiku-20241022"]; - const visionKeywords = [ - "vision", - "gpt-4o", - "claude-3", - "gemini-1.5", - "gemini-exp", - "learnlm", - "qwen-vl", - "qwen2-vl", - ]; - const isGpt4Turbo = - model.includes("gpt-4-turbo") && !model.includes("preview"); - + const visionModels = useAccessStore.getState().visionModels; + const envVisionModels = visionModels + ?.split(",") + .map((m) => m.trim()); + if (envVisionModels?.includes(model)) { + return true; + } return ( - !excludeKeywords.some((keyword) => model.includes(keyword)) && - (visionKeywords.some((keyword) => model.includes(keyword)) || - isGpt4Turbo || - isDalle3(model)) + !EXCLUDE_VISION_MODEL_REGEXES.some((regex) => regex.test(model)) && + VISION_MODEL_REGEXES.some((regex) => regex.test(model)) ); } @@ -280,6 +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 || diff --git a/app/utils/model.ts b/app/utils/model.ts index a1b7df1b6..a1a38a2f8 100644 --- a/app/utils/model.ts +++ b/app/utils/model.ts @@ -202,3 +202,52 @@ export function isModelAvailableInServer( const modelTable = collectModelTable(DEFAULT_MODELS, customModels); return modelTable[fullName]?.available === false; } + +/** + * Check if the model name is a GPT-4 related model + * + * @param modelName The name of the model to check + * @returns True if the model is a GPT-4 related model (excluding gpt-4o-mini) + */ +export function isGPT4Model(modelName: string): boolean { + return ( + (modelName.startsWith("gpt-4") || + modelName.startsWith("chatgpt-4o") || + modelName.startsWith("o1")) && + !modelName.startsWith("gpt-4o-mini") + ); +} + +/** + * Checks if a model is not available on any of the specified providers in the server. + * + * @param {string} customModels - A string of custom models, comma-separated. + * @param {string} modelName - The name of the model to check. + * @param {string|string[]} providerNames - A string or array of provider names to check against. + * + * @returns {boolean} True if the model is not available on any of the specified providers, false otherwise. + */ +export function isModelNotavailableInServer( + customModels: string, + modelName: string, + providerNames: string | string[], +): boolean { + // Check DISABLE_GPT4 environment variable + if ( + process.env.DISABLE_GPT4 === "1" && + isGPT4Model(modelName.toLowerCase()) + ) { + return true; + } + + const modelTable = collectModelTable(DEFAULT_MODELS, customModels); + + const providerNamesArray = Array.isArray(providerNames) + ? providerNames + : [providerNames]; + for (const providerName of providerNamesArray) { + const fullName = `${modelName}@${providerName.toLowerCase()}`; + if (modelTable?.[fullName]?.available === true) return false; + } + return true; +} diff --git a/package.json b/package.json index 39c04368a..12a26cdd3 100644 --- a/package.json +++ b/package.json @@ -62,7 +62,7 @@ "@tauri-apps/cli": "1.5.11", "@testing-library/dom": "^10.4.0", "@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", diff --git a/test/model-available.test.ts b/test/model-available.test.ts new file mode 100644 index 000000000..5c9fa9977 --- /dev/null +++ b/test/model-available.test.ts @@ -0,0 +1,80 @@ +import { isModelNotavailableInServer } from "../app/utils/model"; + +describe("isModelNotavailableInServer", () => { + test("test model will return false, which means the model is available", () => { + const customModels = ""; + const modelName = "gpt-4"; + const providerNames = "OpenAI"; + const result = isModelNotavailableInServer( + customModels, + modelName, + providerNames, + ); + expect(result).toBe(false); + }); + + test("test model will return true when model is not available in custom models", () => { + const customModels = "-all,gpt-4o-mini"; + const modelName = "gpt-4"; + const providerNames = "OpenAI"; + const result = isModelNotavailableInServer( + customModels, + modelName, + providerNames, + ); + expect(result).toBe(true); + }); + + test("should respect DISABLE_GPT4 setting", () => { + process.env.DISABLE_GPT4 = "1"; + const result = isModelNotavailableInServer("", "gpt-4", "OpenAI"); + expect(result).toBe(true); + }); + + test("should handle empty provider names", () => { + const result = isModelNotavailableInServer("-all,gpt-4", "gpt-4", ""); + expect(result).toBe(true); + }); + + test("should be case insensitive for model names", () => { + const result = isModelNotavailableInServer("-all,GPT-4", "gpt-4", "OpenAI"); + expect(result).toBe(true); + }); + + test("support passing multiple providers, model unavailable on one of the providers will return true", () => { + const customModels = "-all,gpt-4@google"; + const modelName = "gpt-4"; + const providerNames = ["OpenAI", "Azure"]; + const result = isModelNotavailableInServer( + customModels, + modelName, + providerNames, + ); + expect(result).toBe(true); + }); + + // FIXME: 这个测试用例有问题，需要修复 + // test("support passing multiple providers, model available on one of the providers will return false", () => { + // const customModels = "-all,gpt-4@google"; + // const modelName = "gpt-4"; + // const providerNames = ["OpenAI", "Google"]; + // const result = isModelNotavailableInServer( + // customModels, + // modelName, + // providerNames, + // ); + // expect(result).toBe(false); + // }); + + test("test custom model without setting provider", () => { + const customModels = "-all,mistral-large"; + const modelName = "mistral-large"; + const providerNames = modelName; + const result = isModelNotavailableInServer( + customModels, + modelName, + providerNames, + ); + expect(result).toBe(false); + }); +}); diff --git a/test/vision-model-checker.test.ts b/test/vision-model-checker.test.ts new file mode 100644 index 000000000..734e992d8 --- /dev/null +++ b/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); + }); +}); \ No newline at end of file diff --git a/yarn.lock b/yarn.lock index b8ecc3f07..d49b9f222 100644 --- a/yarn.lock +++ b/yarn.lock @@ -2807,10 +2807,10 @@ lodash "^4.17.21" redent "^3.0.0" -"@testing-library/react@^16.0.1": - version "16.0.1" - resolved "https://registry.yarnpkg.com/@testing-library/react/-/react-16.0.1.tgz#29c0ee878d672703f5e7579f239005e4e0faa875" - integrity sha512-dSmwJVtJXmku+iocRhWOUFbrERC76TX2Mnf0ATODz8brzAZrMBbzLwQixlBSanZxR6LddK3eiwpSFZgDET1URg== +"@testing-library/react@^16.1.0": + version "16.1.0" + resolved "https://registry.yarnpkg.com/@testing-library/react/-/react-16.1.0.tgz#aa0c61398bac82eaf89776967e97de41ac742d71" + integrity sha512-Q2ToPvg0KsVL0ohND9A3zLJWcOXXcO8IDu3fj11KhNt0UlCWyFyvnCIBkd12tidB2lkiVRG8VFqdhcqhqnAQtg== dependencies: "@babel/runtime" "^7.12.5"