Merge ecc65fa775 into c30ddfbb07

Fix: Encourage markdown inclusion in model responses for o1/o3

README.md

@@ -22,7 +22,7 @@ English / [简体中文](./README_CN.md)
 [![MacOS][MacOS-image]][download-url]
 [![Linux][Linux-image]][download-url]
 
-[NextChatAI](https://nextchat.club?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.club?utm_source=readme) / [iOS APP](https://apps.apple.com/us/app/nextchat-ai/id6743085599) / [Web App Demo](https://app.nextchat.dev) / [Desktop App](https://github.com/Yidadaa/ChatGPT-Next-Web/releases) / [Enterprise Edition](#enterprise-edition) 
 
 
 [saas-url]: https://nextchat.club?utm_source=readme
@@ -40,13 +40,14 @@ English / [简体中文](./README_CN.md)
 
 </div>
 
-## 🥳 Cheer for DeepSeek, China's AI star!
- > Purpose-Built UI for DeepSeek Reasoner Model
+## 🥳 Cheer for NextChat iOS Version Online!
+> [👉 Click Here to Install Now](https://apps.apple.com/us/app/nextchat-ai/id6743085599)
+
+> [❤️ Source Code Coming Soon](https://github.com/ChatGPTNextWeb/NextChat-iOS)
+
+![Github iOS Image](https://github.com/user-attachments/assets/e0aa334f-4c13-4dc9-8310-e3b09fa4b9f3)
+
  
-<img src="https://github.com/user-attachments/assets/f3952210-3af1-4dc0-9b81-40eaa4847d9a"/>
-
-
-
 ## 🫣 NextChat Support MCP  ! 
 > Before build, please set env ENABLE_MCP=true
 

app/api/common.ts

@@ -1,6 +1,6 @@
 import { NextRequest, NextResponse } from "next/server";
 import { getServerSideConfig } from "../config/server";
-import { OPENAI_BASE_URL, ServiceProvider } from "../constant";
+import { OPENAI_BASE_URL, ServiceProvider, OpenaiPath } from "../constant";
 import { cloudflareAIGatewayUrl } from "../utils/cloudflare";
 import { getModelProvider, isModelNotavailableInServer } from "../utils/model";
 
@@ -26,8 +26,8 @@ export async function requestOpenai(req: NextRequest) {
     authValue = req.headers.get("Authorization") ?? "";
     authHeaderName = "Authorization";
   }
-
   let path = `${req.nextUrl.pathname}`.replaceAll("/api/openai/", "");
+  let isChatRequest = path.includes(OpenaiPath.ChatPath);
 
   let baseUrl =
     (isAzure ? serverConfig.azureUrl : serverConfig.baseUrl) || OPENAI_BASE_URL;
@@ -117,14 +117,14 @@ export async function requestOpenai(req: NextRequest) {
       const jsonBody = JSON.parse(clonedBody) as { model?: string };
 
       // not undefined and is false
-      if (
+      if ( isChatRequest &&
         isModelNotavailableInServer(
           serverConfig.customModels,
           jsonBody?.model as string,
           [
             ServiceProvider.OpenAI,
             ServiceProvider.Azure,
-            jsonBody?.model as string, // support provider-unspecified model
+            "custom" as string, // support provider-unspecified model
           ],
         )
       ) {

app/client/platforms/openai.ts

@@ -56,7 +56,7 @@ export interface OpenAIListModelResponse {
 
 export interface RequestPayload {
   messages: {
-    role: "system" | "user" | "assistant";
+    role: "developer" | "system" | "user" | "assistant";
     content: string | MultimodalContent[];
   }[];
   stream?: boolean;
@@ -198,7 +198,8 @@ export class ChatGPTApi implements LLMApi {
     const isDalle3 = _isDalle3(options.config.model);
     const isO1OrO3 =
       options.config.model.startsWith("o1") ||
-      options.config.model.startsWith("o3");
+      options.config.model.startsWith("o3") ||
+      options.config.model.startsWith("o4-mini");
     if (isDalle3) {
       const prompt = getMessageTextContent(
         options.messages.slice(-1)?.pop() as any,
@@ -237,13 +238,21 @@ 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 (isO1OrO3) {
+        // by default the o1/o3 models will not attempt to produce output that includes markdown formatting
+        // manually add "Formatting re-enabled" developer message to encourage markdown inclusion in model responses
+        // (https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/reasoning?tabs=python-secure#markdown-output)
+        requestPayload["messages"].unshift({
+          role: "developer",
+          content: "Formatting re-enabled",
+        });
+
+        // o1/o3 uses max_completion_tokens to control the number of tokens (https://platform.openai.com/docs/guides/reasoning#controlling-costs)
         requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
       }
 
       // add max_tokens to vision model
-      if (visionModel) {
+      if (visionModel && !isO1OrO3) {
         requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
       }
     }

app/constant.ts

@@ -417,6 +417,14 @@ export const KnowledgeCutOffDate: Record<string, string> = {
   "gpt-4-turbo": "2023-12",
   "gpt-4-turbo-2024-04-09": "2023-12",
   "gpt-4-turbo-preview": "2023-12",
+  "gpt-4.1": "2024-06",
+  "gpt-4.1-2025-04-14": "2024-06",
+  "gpt-4.1-mini": "2024-06",
+  "gpt-4.1-mini-2025-04-14": "2024-06",
+  "gpt-4.1-nano": "2024-06",
+  "gpt-4.1-nano-2025-04-14": "2024-06",
+  "gpt-4.5-preview": "2023-10",
+  "gpt-4.5-preview-2025-02-27": "2023-10",
   "gpt-4o": "2023-10",
   "gpt-4o-2024-05-13": "2023-10",
   "gpt-4o-2024-08-06": "2023-10",
@@ -458,6 +466,7 @@ export const DEFAULT_TTS_VOICES = [
 export const VISION_MODEL_REGEXES = [
   /vision/,
   /gpt-4o/,
+  /gpt-4\.1/,
   /claude-3/,
   /gemini-1\.5/,
   /gemini-exp/,
@@ -469,6 +478,8 @@ export const VISION_MODEL_REGEXES = [
   /^dall-e-3$/, // Matches exactly "dall-e-3"
   /glm-4v/,
   /vl/i,
+  /o3/,
+  /o4-mini/,
 ];
 
 export const EXCLUDE_VISION_MODEL_REGEXES = [/claude-3-5-haiku-20241022/];
@@ -485,6 +496,14 @@ const openaiModels = [
   "gpt-4-32k-0613",
   "gpt-4-turbo",
   "gpt-4-turbo-preview",
+  "gpt-4.1",
+  "gpt-4.1-2025-04-14",
+  "gpt-4.1-mini",
+  "gpt-4.1-mini-2025-04-14",
+  "gpt-4.1-nano",
+  "gpt-4.1-nano-2025-04-14",
+  "gpt-4.5-preview",
+  "gpt-4.5-preview-2025-02-27",
   "gpt-4o",
   "gpt-4o-2024-05-13",
   "gpt-4o-2024-08-06",
@@ -499,6 +518,8 @@ const openaiModels = [
   "o1-mini",
   "o1-preview",
   "o3-mini",
+  "o3",
+  "o4-mini",
 ];
 
 const googleModels = [
@@ -525,6 +546,7 @@ const googleModels = [
   "gemini-2.0-flash-thinking-exp-01-21",
   "gemini-2.0-pro-exp",
   "gemini-2.0-pro-exp-02-05",
+  "gemini-2.5-pro-preview-06-05",
 ];
 
 const anthropicModels = [
@@ -611,6 +633,18 @@ const xAIModes = [
   "grok-2-vision-1212",
   "grok-2-vision",
   "grok-2-vision-latest",
+  "grok-3-mini-fast-beta",
+  "grok-3-mini-fast",
+  "grok-3-mini-fast-latest",
+  "grok-3-mini-beta",
+  "grok-3-mini",
+  "grok-3-mini-latest",
+  "grok-3-fast-beta",
+  "grok-3-fast",
+  "grok-3-fast-latest",
+  "grok-3-beta",
+  "grok-3",
+  "grok-3-latest",
 ];
 
 const chatglmModels = [

app/utils/model.ts

@@ -66,8 +66,8 @@ export function collectModelTable(
 
   // default models
   models.forEach((m) => {
-    // using <modelName>@<providerId> as fullName
-    modelTable[`${m.name}@${m?.provider?.id}`] = {
+    // using <modelName>@<providerType> as fullName
+    modelTable[`${m.name}@${m?.provider?.providerType}`] = {
       ...m,
       displayName: m.name, // 'provider' is copied over if it exists
     };
@@ -121,12 +121,14 @@ export function collectModelTable(
           if (displayName && provider.providerName == "ByteDance") {
             [customModelName, displayName] = [displayName, customModelName];
           }
-          modelTable[`${customModelName}@${provider?.id}`] = {
+          modelTable[`${customModelName}@${provider?.providerType}`] = {
             name: customModelName,
             displayName: displayName || customModelName,
             available,
             provider, // Use optional chaining
-            sorted: CustomSeq.next(`${customModelName}@${provider?.id}`),
+            sorted: CustomSeq.next(
+              `${customModelName}@${provider?.providerType}`,
+            ),
           };
         }
       }

test/model-available.test.ts

@@ -53,7 +53,7 @@ describe("isModelNotavailableInServer", () => {
     expect(result).toBe(true);
   });
 
-  // FIXME: 这个测试用例有问题，需要修复
+  // 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";
@@ -69,7 +69,19 @@ describe("isModelNotavailableInServer", () => {
   test("test custom model without setting provider", () => {
     const customModels = "-all,mistral-large";
     const modelName = "mistral-large";
-    const providerNames = modelName;
+    const providerNames = "custom";
+    const result = isModelNotavailableInServer(
+      customModels,
+      modelName,
+      providerNames,
+    );
+    expect(result).toBe(false);
+  });
+
+  test("test custom model with non-standard provider", () => {
+    const customModels = "-all,deepseek-chat@DeepSeek";
+    const modelName = "deepseek-chat";
+    const providerNames = "custom";
     const result = isModelNotavailableInServer(
       customModels,
       modelName,