Merge ac7b720b5b into 48469bd8ca

.gitignore

@@ -49,3 +49,4 @@ masks.json
 
 # mcp config
 app/mcp/mcp_config.json
+Dockerfile.local

app/client/platforms/anthropic.ts

@@ -71,8 +71,6 @@ const ClaudeMapper = {
   system: "user",
 } as const;
 
-const keys = ["claude-2, claude-instant-1"];
-
 export class ClaudeApi implements LLMApi {
   speech(options: SpeechOptions): Promise<ArrayBuffer> {
     throw new Error("Method not implemented.");

app/client/platforms/google.ts

@@ -197,8 +197,6 @@ export class GeminiProApi implements LLMApi {
         signal: controller.signal,
         headers: getHeaders(),
       };
-
-      const isThinking = options.config.model.includes("-thinking");
       // make a fetch request
       const requestTimeoutId = setTimeout(
         () => controller.abort(),

app/client/platforms/openai.ts

@@ -67,6 +67,8 @@ export interface RequestPayload {
   top_p: number;
   max_tokens?: number;
   max_completion_tokens?: number;
+  reasoning_effort?: string;
+  // O3 only
 }
 
 export interface DalleRequestPayload {
@@ -196,9 +198,9 @@ export class ChatGPTApi implements LLMApi {
     let requestPayload: RequestPayload | DalleRequestPayload;
 
     const isDalle3 = _isDalle3(options.config.model);
-    const isO1OrO3 =
-      options.config.model.startsWith("o1") ||
-      options.config.model.startsWith("o3");
+    const isO1 = options.config.model.startsWith("o1");
+    const isO3 = options.config.model.startsWith("o3");
+    const isO1OrO3 = isO1 || isO3;
     if (isDalle3) {
       const prompt = getMessageTextContent(
         options.messages.slice(-1)?.pop() as any,
@@ -242,9 +244,18 @@ export class ChatGPTApi implements LLMApi {
         requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
       }
 
+      if (isO3) {
+        requestPayload["reasoning_effort"] = "high";
+        // make o3-mini defaults to high reasoning effort
+      }
+
       // add max_tokens to vision model
       if (visionModel) {
-        requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
+        if (isO1) {
+          requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
+        } else {
+          requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
+        }
       }
     }
 
@@ -286,6 +297,11 @@ export class ChatGPTApi implements LLMApi {
           isDalle3 ? OpenaiPath.ImagePath : OpenaiPath.ChatPath,
         );
       }
+      // make a fetch request
+      const requestTimeoutId = setTimeout(
+        () => controller.abort(),
+        getTimeoutMSByModel(options.config.model),
+      );
       if (shouldStream) {
         let index = -1;
         const [tools, funcs] = usePluginStore
@@ -393,12 +409,6 @@ export class ChatGPTApi implements LLMApi {
           headers: getHeaders(),
         };
 
-        // make a fetch request
-        const requestTimeoutId = setTimeout(
-          () => controller.abort(),
-          getTimeoutMSByModel(options.config.model),
-        );
-
         const res = await fetch(chatPath, chatPayload);
         clearTimeout(requestTimeoutId);
 

app/constant.ts

@@ -408,35 +408,20 @@ You are an AI assistant with access to system tools. Your role is to help users
    
 `;
 
-export const SUMMARIZE_MODEL = "gpt-4o-mini";
-export const GEMINI_SUMMARIZE_MODEL = "gemini-pro";
+export const SUMMARIZE_MODEL = "gpt-4.1-mini";
+export const GEMINI_SUMMARIZE_MODEL = "gemini-2.0-flash";
 export const DEEPSEEK_SUMMARIZE_MODEL = "deepseek-chat";
 
 export const KnowledgeCutOffDate: Record<string, string> = {
-  default: "2021-09",
-  "gpt-4-turbo": "2023-12",
-  "gpt-4-turbo-2024-04-09": "2023-12",
-  "gpt-4-turbo-preview": "2023-12",
-  "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",
-  "gpt-4-vision-preview": "2023-04",
-  "o1-mini-2024-09-12": "2023-10",
-  "o1-mini": "2023-10",
-  "o1-preview-2024-09-12": "2023-10",
-  "o1-preview": "2023-10",
-  "o1-2024-12-17": "2023-10",
-  o1: "2023-10",
-  "o3-mini-2025-01-31": "2023-10",
-  "o3-mini": "2023-10",
+  default: "2023-10",
   // After improvements,
   // 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",
+  "gemini-2.5-pro-exp-03-25": "2025-01",
+  "gemini-2.0-flash": "2024-08",
+  "claude-3-7-sonnet-latest": "2024-10",
+  "claude-3-5-haiku-latest": "2024-10",
+  "gpt-4.1": "2024-06",
+  "gpt-4.1-mini": "2024-06",
   "deepseek-chat": "2024-07",
   "deepseek-coder": "2024-07",
 };
@@ -457,11 +442,12 @@ export const DEFAULT_TTS_VOICES = [
 
 export const VISION_MODEL_REGEXES = [
   /vision/,
-  /gpt-4o/,
+  /gpt-4\.1/,
   /claude-3/,
   /gemini-1\.5/,
   /gemini-exp/,
   /gemini-2\.0/,
+  /gemini-2\.5-pro/,
   /learnlm/,
   /qwen-vl/,
   /qwen2-vl/,
@@ -469,78 +455,23 @@ export const VISION_MODEL_REGEXES = [
   /^dall-e-3$/, // Matches exactly "dall-e-3"
   /glm-4v/,
   /vl/i,
+  /o1/,
 ];
 
 export const EXCLUDE_VISION_MODEL_REGEXES = [/claude-3-5-haiku-20241022/];
 
-const openaiModels = [
-  // As of July 2024, gpt-4o-mini should be used in place of gpt-3.5-turbo,
-  // as it is cheaper, more capable, multimodal, and just as fast. gpt-3.5-turbo is still available for use in the API.
-  "gpt-3.5-turbo",
-  "gpt-3.5-turbo-1106",
-  "gpt-3.5-turbo-0125",
-  "gpt-4",
-  "gpt-4-0613",
-  "gpt-4-32k",
-  "gpt-4-32k-0613",
-  "gpt-4-turbo",
-  "gpt-4-turbo-preview",
-  "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",
-  "gpt-4-vision-preview",
-  "gpt-4-turbo-2024-04-09",
-  "gpt-4-1106-preview",
-  "dall-e-3",
-  "o1-mini",
-  "o1-preview",
-  "o3-mini",
-];
+const openaiModels = ["dall-e-3", "o1", "o3-mini", "gpt-4.1", "gpt-4.1-mini"];
 
 const googleModels = [
-  "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",
-  "gemini-exp-1206",
   "gemini-2.0-flash",
-  "gemini-2.0-flash-exp",
-  "gemini-2.0-flash-lite-preview-02-05",
-  "gemini-2.0-flash-thinking-exp",
-  "gemini-2.0-flash-thinking-exp-1219",
-  "gemini-2.0-flash-thinking-exp-01-21",
-  "gemini-2.0-pro-exp",
-  "gemini-2.0-pro-exp-02-05",
+  "gemini-2.0-flash-lite",
+  "gemini-2.5-pro-exp-03-25",
 ];
 
 const anthropicModels = [
-  "claude-instant-1.2",
-  "claude-2.0",
-  "claude-2.1",
-  "claude-3-sonnet-20240229",
-  "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-3-7-sonnet-20250219",
   "claude-3-7-sonnet-latest",
 ];
 

app/store/config.ts

@@ -66,14 +66,14 @@ export const DEFAULT_CONFIG = {
   modelConfig: {
     model: "gpt-4o-mini" as ModelType,
     providerName: "OpenAI" as ServiceProvider,
-    temperature: 0.5,
+    temperature: 0.2,
     top_p: 1,
     max_tokens: 4000,
     presence_penalty: 0,
     frequency_penalty: 0,
     sendMemory: true,
-    historyMessageCount: 4,
-    compressMessageLengthThreshold: 1000,
+    historyMessageCount: 20,
+    compressMessageLengthThreshold: 5000,
     compressModel: "",
     compressProviderName: "",
     enableInjectSystemPrompts: true,

app/utils.ts

@@ -304,9 +304,18 @@ export function getTimeoutMSByModel(model: string) {
     model.startsWith("o1") ||
     model.startsWith("o3") ||
     model.includes("deepseek-r") ||
-    model.includes("-thinking")
-  )
+    model.includes("-thinking") ||
+    model.includes("pro")
+  ) {
+    console.log(
+      "thinking model is " +
+        model +
+        " timeout is " +
+        REQUEST_TIMEOUT_MS_FOR_THINKING,
+    );
     return REQUEST_TIMEOUT_MS_FOR_THINKING;
+  }
+  console.log("normal model is " + model + " timeout is " + REQUEST_TIMEOUT_MS);
   return REQUEST_TIMEOUT_MS;
 }
 

test/model-available.test.ts

@@ -3,7 +3,7 @@ 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 modelName = "gpt-4.1";
     const providerNames = "OpenAI";
     const result = isModelNotavailableInServer(
       customModels,

test/vision-model-checker.test.ts

@@ -15,10 +15,11 @@ describe("isVisionModel", () => {
 
   test("should identify vision models using regex patterns", () => {
     const visionModels = [
-      "gpt-4-vision",
+      "gpt-4.1",
       "claude-3-opus",
       "gemini-1.5-pro",
       "gemini-2.0",
+      "gemini-2.5-pro",
       "gemini-exp-vision",
       "learnlm-vision",
       "qwen-vl-max",