Merge branch 'main' into update_knowledge_cutoff_date

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("o1")) ||
+        !(m.id.startsWith("gpt-4") || m.id.startsWith("chatgpt-4o") || m.id.startsWith("o1") || m.id.startsWith("o3")) ||
         m.id.startsWith("gpt-4o-mini"),
     );
   }

app/client/platforms/openai.ts

@@ -195,7 +195,7 @@ export class ChatGPTApi implements LLMApi {
     let requestPayload: RequestPayload | DalleRequestPayload;
 
     const isDalle3 = _isDalle3(options.config.model);
-    const isO1 = options.config.model.startsWith("o1");
+    const isO1OrO3 = options.config.model.startsWith("o1") || options.config.model.startsWith("o3");
     if (isDalle3) {
       const prompt = getMessageTextContent(
         options.messages.slice(-1)?.pop() as any,
@@ -217,7 +217,7 @@ export class ChatGPTApi implements LLMApi {
         const content = visionModel
           ? await preProcessImageContent(v.content)
           : getMessageTextContent(v);
-        if (!(isO1 && v.role === "system"))
+        if (!(isO1OrO3 && v.role === "system"))
           messages.push({ role: v.role, content });
       }
 
@@ -226,16 +226,16 @@ export class ChatGPTApi implements LLMApi {
         messages,
         stream: options.config.stream,
         model: modelConfig.model,
-        temperature: !isO1 ? modelConfig.temperature : 1,
-        presence_penalty: !isO1 ? modelConfig.presence_penalty : 0,
-        frequency_penalty: !isO1 ? modelConfig.frequency_penalty : 0,
-        top_p: !isO1 ? modelConfig.top_p : 1,
+        temperature: !isO1OrO3 ? modelConfig.temperature : 1,
+        presence_penalty: !isO1OrO3 ? modelConfig.presence_penalty : 0,
+        frequency_penalty: !isO1OrO3 ? modelConfig.frequency_penalty : 0,
+        top_p: !isO1OrO3 ? modelConfig.top_p : 1,
         // 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.
       };
 
       // O1 使用 max_completion_tokens 控制token数 (https://platform.openai.com/docs/guides/reasoning#controlling-costs)
-      if (isO1) {
+      if (isO1OrO3) {
         requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
       }
 
@@ -359,7 +359,7 @@ export class ChatGPTApi implements LLMApi {
         // make a fetch request
         const requestTimeoutId = setTimeout(
           () => controller.abort(),
-          isDalle3 || isO1 ? REQUEST_TIMEOUT_MS * 4 : REQUEST_TIMEOUT_MS, // dalle3 using b64_json is slow.
+          isDalle3 || isO1OrO3 ? REQUEST_TIMEOUT_MS * 4 : REQUEST_TIMEOUT_MS, // dalle3 using b64_json is slow.
         );
 
         const res = await fetch(chatPath, chatPayload);

app/components/emoji.tsx

@@ -38,7 +38,8 @@ export function Avatar(props: { model?: ModelType; avatar?: string }) {
       <div className="no-dark">
         {props.model?.startsWith("gpt-4") ||
         props.model?.startsWith("chatgpt-4o") ||
-        props.model?.startsWith("o1") ? (
+        props.model?.startsWith("o1") || 
+        props.model?.startsWith("o3") ? (
           <BlackBotIcon className="user-avatar" />
         ) : (
           <BotIcon className="user-avatar" />

app/constant.ts

@@ -500,7 +500,9 @@ const googleModels = [
   "gemini-exp-1121",
   "gemini-exp-1206",
   "gemini-2.0-flash-exp",
+  "gemini-2.0-flash-thinking-exp",
   "gemini-2.0-flash-thinking-exp-1219",
+  "gemini-2.0-flash-thinking-exp-01-21",
 ];
 
 const anthropicModels = [

app/utils/model.ts

@@ -1,4 +1,4 @@
-import { DEFAULT_MODELS } from "../constant";
+import { DEFAULT_MODELS, ServiceProvider } from "../constant";
 import { LLMModel } from "../client/api";
 
 const CustomSeq = {
@@ -246,6 +246,11 @@ export function isModelNotavailableInServer(
     ? providerNames
     : [providerNames];
   for (const providerName of providerNamesArray) {
+    // if model provider is bytedance, use model config name to check if not avaliable
+    if (providerName === ServiceProvider.ByteDance) {
+      return !Object.values(modelTable).filter((v) => v.name === modelName)?.[0]
+        ?.available;
+    }
     const fullName = `${modelName}@${providerName.toLowerCase()}`;
     if (modelTable?.[fullName]?.available === true) return false;
   }