Merge c3b50a9c93 into 995bef73de

add: model gpt-5

app/client/platforms/openai.ts

@@ -200,6 +200,7 @@ export class ChatGPTApi implements LLMApi {
       options.config.model.startsWith("o1") ||
       options.config.model.startsWith("o3") ||
       options.config.model.startsWith("o4-mini");
+    const isGpt5 =  options.config.model.startsWith("gpt-5");
     if (isDalle3) {
       const prompt = getMessageTextContent(
         options.messages.slice(-1)?.pop() as any,
@@ -230,7 +231,7 @@ export class ChatGPTApi implements LLMApi {
         messages,
         stream: options.config.stream,
         model: modelConfig.model,
-        temperature: !isO1OrO3 ? modelConfig.temperature : 1,
+        temperature: (!isO1OrO3 && !isGpt5) ? modelConfig.temperature : 1,
         presence_penalty: !isO1OrO3 ? modelConfig.presence_penalty : 0,
         frequency_penalty: !isO1OrO3 ? modelConfig.frequency_penalty : 0,
         top_p: !isO1OrO3 ? modelConfig.top_p : 1,
@@ -238,7 +239,13 @@ 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.
       };
 
-      if (isO1OrO3) {
+      if (isGpt5) {
+  	// Remove max_tokens if present
+  	delete requestPayload.max_tokens;
+  	// Add max_completion_tokens (or max_completion_tokens if that's what you meant)
+  	requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
+
+      } else 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)
@@ -251,8 +258,9 @@ export class ChatGPTApi implements LLMApi {
         requestPayload["max_completion_tokens"] = modelConfig.max_tokens;
       }
 
+
       // add max_tokens to vision model
-      if (visionModel && !isO1OrO3) {
+      if (visionModel && !isO1OrO3 && ! isGpt5) {
         requestPayload["max_tokens"] = Math.max(modelConfig.max_tokens, 4000);
       }
     }

app/components/tts-config.tsx

@@ -1,19 +1,76 @@
 import { TTSConfig, TTSConfigValidator } from "../store";
+import React, { useState } from "react";
 
 import Locale from "../locales";
 import { ListItem, Select } from "./ui-lib";
 import {
+  ModelProvider,
   DEFAULT_TTS_ENGINE,
   DEFAULT_TTS_ENGINES,
   DEFAULT_TTS_MODELS,
   DEFAULT_TTS_VOICES,
 } from "../constant";
 import { InputRange } from "./input-range";
+import { IconButton } from "./button";
+import SpeakIcon from "../icons/speak.svg";
+import SpeakStopIcon from "../icons/speak-stop.svg";
+import { createTTSPlayer } from "../utils/audio";
+import { useAppConfig } from "../store";
+import { ClientApi } from "../client/api";
+import { showToast } from "../components/ui-lib";
 
+const ttsPlayer = createTTSPlayer();
 export function TTSConfigList(props: {
   ttsConfig: TTSConfig;
   updateConfig: (updater: (config: TTSConfig) => void) => void;
 }) {
+  const [speechLoading, setSpeechLoading] = useState(false);
+  const [speechStatus, setSpeechStatus] = useState(false);
+
+  const config = useAppConfig.getState();
+
+  function stopSpeech() {
+    ttsPlayer.stop();
+    setSpeechStatus(false);
+  }
+
+  async function playSpeech(text: string, ttsConfig: TTSConfig) {
+    try {
+      const api = new ClientApi(ModelProvider.GPT);
+      setSpeechLoading(true);
+      ttsPlayer.init();
+
+      const audioBuffer = await api.llm.speech({
+        model: ttsConfig.model,
+        input: text,
+        voice: ttsConfig.voice,
+        speed: ttsConfig.speed,
+      });
+
+      setSpeechStatus(true);
+      await ttsPlayer.play(audioBuffer, () => {
+        setSpeechStatus(false);
+      });
+    } catch (error) {
+      console.error("[OpenAI Speech]", error);
+      setSpeechStatus(false);
+      // Implement user-facing error notification here
+      if (typeof (error as Error).message === "string") {
+        showToast((error as Error).message);
+      }
+    } finally {
+      setSpeechLoading(false);
+    }
+  }
+
+  async function openaiSpeech(text: string) {
+    if (speechStatus) {
+      stopSpeech();
+    } else {
+      await playSpeech(text, config.ttsConfig);
+    }
+  }
+
   return (
     <>
       <ListItem
@@ -88,15 +145,32 @@ export function TTSConfigList(props: {
             title={Locale.Settings.TTS.Voice.Title}
             subTitle={Locale.Settings.TTS.Voice.SubTitle}
           >
+            <div style={{ display: "flex", gap: "10px" }}>
+              <IconButton
+                aria={Locale.Chat.Actions.Speech}
+                icon={speechStatus ? <SpeakStopIcon /> : <SpeakIcon />}
+                text={
+                  speechLoading
+                    ? "Loading..."
+                    : speechStatus
+                    ? Locale.Chat.Actions.Stop
+                    : Locale.Chat.Actions.Speech
+                }
+                onClick={() => {
+                  openaiSpeech(
+                    "NextChat,Unleash your imagination, experience the future of AI conversation.",
+                  );
+                }}
+              />
+
               <Select
                 value={props.ttsConfig.voice}
                 onChange={(e) => {
-                props.updateConfig(
-                  (config) =>
-                    (config.voice = TTSConfigValidator.voice(
+                  props.updateConfig((config) => {
+                    config.voice = TTSConfigValidator.voice(
                       e.currentTarget.value,
-                    )),
                     );
+                  });
                 }}
               >
                 {DEFAULT_TTS_VOICES.map((v, i) => (
@@ -105,6 +179,7 @@ export function TTSConfigList(props: {
                   </option>
                 ))}
               </Select>
+            </div>
           </ListItem>
           <ListItem
             title={Locale.Settings.TTS.Speed.Title}

app/constant.ts

@@ -493,6 +493,7 @@ export const VISION_MODEL_REGEXES = [
   /o3/,
   /o4-mini/,
   /grok-4/i,
+  /gpt-5/
 ];
 
 export const EXCLUDE_VISION_MODEL_REGEXES = [/claude-3-5-haiku-20241022/];
@@ -517,6 +518,11 @@ const openaiModels = [
   "gpt-4.1-nano-2025-04-14",
   "gpt-4.5-preview",
   "gpt-4.5-preview-2025-02-27",
+  "gpt-5-chat",
+  "gpt-5-mini",
+  "gpt-5-nano",
+  "gpt-5",
+  "gpt-5-chat-2025-01-01-preview",
   "gpt-4o",
   "gpt-4o-2024-05-13",
   "gpt-4o-2024-08-06",