📝 Add docstrings to feat/accurate-user-search

Docstrings generation was requested by @HenryXiaoYang.

* https://github.com/QuantumNous/new-api/pull/2535#issuecomment-3693274850

The following files were modified:

* `controller/user.go`
* `model/user.go`

.env.example

@@ -9,6 +9,14 @@
 # ENABLE_PPROF=true
 # 启用调试模式
 # DEBUG=true
+# Pyroscope 配置
+# PYROSCOPE_URL=http://localhost:4040
+# PYROSCOPE_APP_NAME=new-api
+# PYROSCOPE_BASIC_AUTH_USER=your-user
+# PYROSCOPE_BASIC_AUTH_PASSWORD=your-password
+# PYROSCOPE_MUTEX_RATE=5
+# PYROSCOPE_BLOCK_RATE=5
+# HOSTNAME=your-hostname
 
 # 数据库相关配置
 # 数据库连接字符串

.gitignore

@@ -16,9 +16,11 @@ new-api
 tiktoken_cache
 .eslintcache
 .gocache
+.gomodcache/
 .cache
 web/bun.lock
 
 electron/node_modules
 electron/dist
 data/
+.gomodcache/

Dockerfile

@@ -28,7 +28,7 @@ RUN go build -ldflags "-s -w -X 'github.com/QuantumNous/new-api/common.Version=$
 FROM debian:bookworm-slim
 
 RUN apt-get update \
-    && apt-get install -y --no-install-recommends ca-certificates tzdata libasan8 \
+    && apt-get install -y --no-install-recommends ca-certificates tzdata libasan8 wget \
     && rm -rf /var/lib/apt/lists/* \
     && update-ca-certificates
 

README.en.md

@@ -146,7 +146,7 @@ docker run --name new-api -d --restart always \
 
 🎉 After deployment is complete, visit `http://localhost:3000` to start using!
 
-📖 For more deployment methods, please refer to [Deployment Guide](https://docs.newapi.pro/installation)
+📖 For more deployment methods, please refer to [Deployment Guide](https://docs.newapi.pro/en/docs/installation)
 
 ---
 
@@ -154,7 +154,7 @@ docker run --name new-api -d --restart always \
 
 <div align="center">
 
-### 📖 [Official Documentation](https://docs.newapi.pro/) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
+### 📖 [Official Documentation](https://docs.newapi.pro/en/docs) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
 
 </div>
 
@@ -162,17 +162,17 @@ docker run --name new-api -d --restart always \
 
 | Category | Link |
 |------|------|
-| 🚀 Deployment Guide | [Installation Documentation](https://docs.newapi.pro/installation) |
-| ⚙️ Environment Configuration | [Environment Variables](https://docs.newapi.pro/installation/environment-variables) |
-| 📡 API Documentation | [API Documentation](https://docs.newapi.pro/api) |
-| ❓ FAQ | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 Community Interaction | [Communication Channels](https://docs.newapi.pro/support/community-interaction) |
+| 🚀 Deployment Guide | [Installation Documentation](https://docs.newapi.pro/en/docs/installation) |
+| ⚙️ Environment Configuration | [Environment Variables](https://docs.newapi.pro/en/docs/installation/config-maintenance/environment-variables) |
+| 📡 API Documentation | [API Documentation](https://docs.newapi.pro/en/docs/api) |
+| ❓ FAQ | [FAQ](https://docs.newapi.pro/en/docs/support/faq) |
+| 💬 Community Interaction | [Communication Channels](https://docs.newapi.pro/en/docs/support/community-interaction) |
 
 ---
 
 ## ✨ Key Features
 
-> For detailed features, please refer to [Features Introduction](https://docs.newapi.pro/wiki/features-introduction)
+> For detailed features, please refer to [Features Introduction](https://docs.newapi.pro/en/docs/guide/wiki/basic-concepts/features-introduction)
 
 ### 🎨 Core Functions
 
@@ -201,11 +201,11 @@ docker run --name new-api -d --restart always \
 ### 🚀 Advanced Features
 
 **API Format Support:**
-- ⚡ [OpenAI Responses](https://docs.newapi.pro/api/openai-responses)
-- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/api/openai-realtime) (including Azure)
-- ⚡ [Claude Messages](https://docs.newapi.pro/api/anthropic-chat)
-- ⚡ [Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
-- 🔄 [Rerank Models](https://docs.newapi.pro/api/jinaai-rerank) (Cohere, Jina)
+- ⚡ [OpenAI Responses](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-response)
+- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/en/docs/api/ai-model/realtime/create-realtime-session) (including Azure)
+- ⚡ [Claude Messages](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message)
+- ⚡ [Google Gemini](https://doc.newapi.pro/en/api/google-gemini-chat)
+- 🔄 [Rerank Models](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank) (Cohere, Jina)
 
 **Intelligent Routing:**
 - ⚖️ Channel weighted random
@@ -246,16 +246,16 @@ docker run --name new-api -d --restart always \
 
 ## 🤖 Model Support
 
-> For details, please refer to [API Documentation - Relay Interface](https://docs.newapi.pro/api)
+> For details, please refer to [API Documentation - Relay Interface](https://docs.newapi.pro/en/docs/api)
 
 | Model Type | Description | Documentation |
 |---------|------|------|
 | 🤖 OpenAI GPTs | gpt-4-gizmo-* series | - |
-| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [Documentation](https://docs.newapi.pro/api/midjourney-proxy-image) |
-| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [Documentation](https://docs.newapi.pro/api/suno-music) |
-| 🔄 Rerank | Cohere, Jina | [Documentation](https://docs.newapi.pro/api/jinaai-rerank) |
-| 💬 Claude | Messages format | [Documentation](https://docs.newapi.pro/api/anthropic-chat) |
-| 🌐 Gemini | Google Gemini format | [Documentation](https://docs.newapi.pro/api/google-gemini-chat/) |
+| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [Documentation](https://doc.newapi.pro/en/api/midjourney-proxy-image) |
+| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [Documentation](https://doc.newapi.pro/en/api/suno-music) |
+| 🔄 Rerank | Cohere, Jina | [Documentation](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank) |
+| 💬 Claude | Messages format | [Documentation](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message) |
+| 🌐 Gemini | Google Gemini format | [Documentation](https://doc.newapi.pro/en/api/google-gemini-chat) |
 | 🔧 Dify | ChatFlow mode | - |
 | 🎯 Custom | Supports complete call address | - |
 
@@ -264,16 +264,16 @@ docker run --name new-api -d --restart always \
 <details>
 <summary>View complete interface list</summary>
 
-- [Chat Interface (Chat Completions)](https://docs.newapi.pro/api/openai-chat)
-- [Response Interface (Responses)](https://docs.newapi.pro/api/openai-responses)
-- [Image Interface (Image)](https://docs.newapi.pro/api/openai-image)
-- [Audio Interface (Audio)](https://docs.newapi.pro/api/openai-audio)
-- [Video Interface (Video)](https://docs.newapi.pro/api/openai-video)
-- [Embedding Interface (Embeddings)](https://docs.newapi.pro/api/openai-embeddings)
-- [Rerank Interface (Rerank)](https://docs.newapi.pro/api/jinaai-rerank)
-- [Realtime Conversation (Realtime)](https://docs.newapi.pro/api/openai-realtime)
-- [Claude Chat](https://docs.newapi.pro/api/anthropic-chat)
-- [Google Gemini Chat](https://docs.newapi.pro/api/google-gemini-chat/)
+- [Chat Interface (Chat Completions)](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-chat-completion)
+- [Response Interface (Responses)](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-response)
+- [Image Interface (Image)](https://docs.newapi.pro/en/docs/api/ai-model/images/openai/v1-images-generations--post)
+- [Audio Interface (Audio)](https://docs.newapi.pro/en/docs/api/ai-model/audio/openai/create-transcription)
+- [Video Interface (Video)](https://docs.newapi.pro/en/docs/api/ai-model/videos/create-video-generation)
+- [Embedding Interface (Embeddings)](https://docs.newapi.pro/en/docs/api/ai-model/embeddings/create-embedding)
+- [Rerank Interface (Rerank)](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank)
+- [Realtime Conversation (Realtime)](https://docs.newapi.pro/en/docs/api/ai-model/realtime/create-realtime-session)
+- [Claude Chat](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message)
+- [Google Gemini Chat](https://doc.newapi.pro/en/api/google-gemini-chat)
 
 </details>
 
@@ -305,10 +305,18 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis connection string | - |
 | `STREAMING_TIMEOUT` | Streaming timeout (seconds) | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | Max per-line buffer (MB) for the stream scanner; increase when upstream sends huge image/base64 payloads | `64` |
+| `MAX_REQUEST_BODY_MB` | Max request body size (MB, counted **after decompression**; prevents huge requests/zip bombs from exhausting memory). Exceeding it returns `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure API version | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Error log switch | `false` |
+| `PYROSCOPE_URL` | Pyroscope server address | - |
+| `PYROSCOPE_APP_NAME` | Pyroscope application name | `new-api` |
+| `PYROSCOPE_BASIC_AUTH_USER` | Pyroscope basic auth user | - |
+| `PYROSCOPE_BASIC_AUTH_PASSWORD` | Pyroscope basic auth password | - |
+| `PYROSCOPE_MUTEX_RATE` | Pyroscope mutex sampling rate | `5` |
+| `PYROSCOPE_BLOCK_RATE` | Pyroscope block sampling rate | `5` |
+| `HOSTNAME` | Hostname tag for Pyroscope | `new-api` |
 
-📖 **Complete configuration:** [Environment Variables Documentation](https://docs.newapi.pro/installation/environment-variables)
+📖 **Complete configuration:** [Environment Variables Documentation](https://docs.newapi.pro/en/docs/installation/config-maintenance/environment-variables)
 
 </details>
 
@@ -410,10 +418,10 @@ docker run --name new-api -d --restart always \
 
 | Resource | Link |
 |------|------|
-| 📘 FAQ | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 Community Interaction | [Communication Channels](https://docs.newapi.pro/support/community-interaction) |
-| 🐛 Issue Feedback | [Issue Feedback](https://docs.newapi.pro/support/feedback-issues) |
-| 📚 Complete Documentation | [Official Documentation](https://docs.newapi.pro/support) |
+| 📘 FAQ | [FAQ](https://docs.newapi.pro/en/docs/support/faq) |
+| 💬 Community Interaction | [Communication Channels](https://docs.newapi.pro/en/docs/support/community-interaction) |
+| 🐛 Issue Feedback | [Issue Feedback](https://docs.newapi.pro/en/docs/support/feedback-issues) |
+| 📚 Complete Documentation | [Official Documentation](https://docs.newapi.pro/en/docs) |
 
 ### 🤝 Contribution Guide
 
@@ -442,7 +450,7 @@ Welcome all forms of contribution!
 
 If this project is helpful to you, welcome to give us a ⭐️ Star！
 
-**[Official Documentation](https://docs.newapi.pro/)** • **[Issue Feedback](https://github.com/Calcium-Ion/new-api/issues)** • **[Latest Release](https://github.com/Calcium-Ion/new-api/releases)**
+**[Official Documentation](https://docs.newapi.pro/en/docs)** • **[Issue Feedback](https://github.com/Calcium-Ion/new-api/issues)** • **[Latest Release](https://github.com/Calcium-Ion/new-api/releases)**
 
 <sub>Built with ❤️ by QuantumNous</sub>
 

README.fr.md

@@ -146,7 +146,7 @@ docker run --name new-api -d --restart always \
 
 🎉 Après le déploiement, visitez `http://localhost:3000` pour commencer à utiliser!
 
-📖 Pour plus de méthodes de déploiement, veuillez vous référer à [Guide de déploiement](https://docs.newapi.pro/installation)
+📖 Pour plus de méthodes de déploiement, veuillez vous référer à [Guide de déploiement](https://docs.newapi.pro/en/docs/installation)
 
 ---
 
@@ -154,7 +154,7 @@ docker run --name new-api -d --restart always \
 
 <div align="center">
 
-### 📖 [Documentation officielle](https://docs.newapi.pro/) | [![Demander à DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
+### 📖 [Documentation officielle](https://docs.newapi.pro/en/docs) | [![Demander à DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
 
 </div>
 
@@ -162,17 +162,17 @@ docker run --name new-api -d --restart always \
 
 | Catégorie | Lien |
 |------|------|
-| 🚀 Guide de déploiement | [Documentation d'installation](https://docs.newapi.pro/installation) |
-| ⚙️ Configuration de l'environnement | [Variables d'environnement](https://docs.newapi.pro/installation/environment-variables) |
-| 📡 Documentation de l'API | [Documentation de l'API](https://docs.newapi.pro/api) |
-| ❓ FAQ | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 Interaction avec la communauté | [Canaux de communication](https://docs.newapi.pro/support/community-interaction) |
+| 🚀 Guide de déploiement | [Documentation d'installation](https://docs.newapi.pro/en/docs/installation) |
+| ⚙️ Configuration de l'environnement | [Variables d'environnement](https://docs.newapi.pro/en/docs/installation/config-maintenance/environment-variables) |
+| 📡 Documentation de l'API | [Documentation de l'API](https://docs.newapi.pro/en/docs/api) |
+| ❓ FAQ | [FAQ](https://docs.newapi.pro/en/docs/support/faq) |
+| 💬 Interaction avec la communauté | [Canaux de communication](https://docs.newapi.pro/en/docs/support/community-interaction) |
 
 ---
 
 ## ✨ Fonctionnalités clés
 
-> Pour les fonctionnalités détaillées, veuillez vous référer à [Présentation des fonctionnalités](https://docs.newapi.pro/wiki/features-introduction) |
+> Pour les fonctionnalités détaillées, veuillez vous référer à [Présentation des fonctionnalités](https://docs.newapi.pro/en/docs/guide/wiki/basic-concepts/features-introduction) |
 
 ### 🎨 Fonctions principales
 
@@ -200,11 +200,11 @@ docker run --name new-api -d --restart always \
 ### 🚀 Fonctionnalités avancées
 
 **Prise en charge des formats d'API:**
-- ⚡ [OpenAI Responses](https://docs.newapi.pro/api/openai-responses)
-- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/api/openai-realtime) (y compris Azure)
-- ⚡ [Claude Messages](https://docs.newapi.pro/api/anthropic-chat)
-- ⚡ [Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
-- 🔄 [Modèles Rerank](https://docs.newapi.pro/api/jinaai-rerank) (Cohere, Jina)
+- ⚡ [OpenAI Responses](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-response)
+- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/en/docs/api/ai-model/realtime/create-realtime-session) (y compris Azure)
+- ⚡ [Claude Messages](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message)
+- ⚡ [Google Gemini](https://doc.newapi.pro/en/api/google-gemini-chat)
+- 🔄 [Modèles Rerank](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank) (Cohere, Jina)
 
 **Routage intelligent:**
 - ⚖️ Sélection aléatoire pondérée des canaux
@@ -242,16 +242,16 @@ docker run --name new-api -d --restart always \
 
 ## 🤖 Prise en charge des modèles
 
-> Pour les détails, veuillez vous référer à [Documentation de l'API - Interface de relais](https://docs.newapi.pro/api)
+> Pour les détails, veuillez vous référer à [Documentation de l'API - Interface de relais](https://docs.newapi.pro/en/docs/api)
 
 | Type de modèle | Description | Documentation |
 |---------|------|------|
 | 🤖 OpenAI GPTs | série gpt-4-gizmo-* | - |
-| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [Documentation](https://docs.newapi.pro/api/midjourney-proxy-image) |
-| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [Documentation](https://docs.newapi.pro/api/suno-music) |
-| 🔄 Rerank | Cohere, Jina | [Documentation](https://docs.newapi.pro/api/jinaai-rerank) |
-| 💬 Claude | Format Messages | [Documentation](https://docs.newapi.pro/api/anthropic-chat) |
-| 🌐 Gemini | Format Google Gemini | [Documentation](https://docs.newapi.pro/api/google-gemini-chat/) |
+| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [Documentation](https://doc.newapi.pro/en/api/midjourney-proxy-image) |
+| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [Documentation](https://doc.newapi.pro/en/api/suno-music) |
+| 🔄 Rerank | Cohere, Jina | [Documentation](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank) |
+| 💬 Claude | Format Messages | [Documentation](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message) |
+| 🌐 Gemini | Format Google Gemini | [Documentation](https://doc.newapi.pro/en/api/google-gemini-chat) |
 | 🔧 Dify | Mode ChatFlow | - |
 | 🎯 Personnalisé | Prise en charge de l'adresse d'appel complète | - |
 
@@ -260,16 +260,16 @@ docker run --name new-api -d --restart always \
 <details>
 <summary>Voir la liste complète des interfaces</summary>
 
-- [Interface de discussion (Chat Completions)](https://docs.newapi.pro/api/openai-chat)
-- [Interface de réponse (Responses)](https://docs.newapi.pro/api/openai-responses)
-- [Interface d'image (Image)](https://docs.newapi.pro/api/openai-image)
-- [Interface audio (Audio)](https://docs.newapi.pro/api/openai-audio)
-- [Interface vidéo (Video)](https://docs.newapi.pro/api/openai-video)
-- [Interface d'incorporation (Embeddings)](https://docs.newapi.pro/api/openai-embeddings)
-- [Interface de rerank (Rerank)](https://docs.newapi.pro/api/jinaai-rerank)
-- [Conversation en temps réel (Realtime)](https://docs.newapi.pro/api/openai-realtime)
-- [Discussion Claude](https://docs.newapi.pro/api/anthropic-chat)
-- [Discussion Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
+- [Interface de discussion (Chat Completions)](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-chat-completion)
+- [Interface de réponse (Responses)](https://docs.newapi.pro/en/docs/api/ai-model/chat/openai/create-response)
+- [Interface d'image (Image)](https://docs.newapi.pro/en/docs/api/ai-model/images/openai/v1-images-generations--post)
+- [Interface audio (Audio)](https://docs.newapi.pro/en/docs/api/ai-model/audio/openai/create-transcription)
+- [Interface vidéo (Video)](https://docs.newapi.pro/en/docs/api/ai-model/videos/create-video-generation)
+- [Interface d'incorporation (Embeddings)](https://docs.newapi.pro/en/docs/api/ai-model/embeddings/create-embedding)
+- [Interface de rerank (Rerank)](https://docs.newapi.pro/en/docs/api/ai-model/rerank/create-rerank)
+- [Conversation en temps réel (Realtime)](https://docs.newapi.pro/en/docs/api/ai-model/realtime/create-realtime-session)
+- [Discussion Claude](https://docs.newapi.pro/en/docs/api/ai-model/chat/create-message)
+- [Discussion Google Gemini](https://doc.newapi.pro/en/api/google-gemini-chat)
 
 </details>
 
@@ -301,10 +301,18 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Chaine de connexion Redis | - |
 | `STREAMING_TIMEOUT` | Délai d'expiration du streaming (secondes) | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | Taille max du buffer par ligne (Mo) pour le scanner SSE ; à augmenter quand les sorties image/base64 sont très volumineuses (ex. images 4K) | `64` |
+| `MAX_REQUEST_BODY_MB` | Taille maximale du corps de requête (Mo, comptée **après décompression** ; évite les requêtes énormes/zip bombs qui saturent la mémoire). Dépassement ⇒ `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Version de l'API Azure | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Interrupteur du journal d'erreurs | `false` |
+| `PYROSCOPE_URL` | Adresse du serveur Pyroscope | - |
+| `PYROSCOPE_APP_NAME` | Nom de l'application Pyroscope | `new-api` |
+| `PYROSCOPE_BASIC_AUTH_USER` | Utilisateur Basic Auth Pyroscope | - |
+| `PYROSCOPE_BASIC_AUTH_PASSWORD` | Mot de passe Basic Auth Pyroscope | - |
+| `PYROSCOPE_MUTEX_RATE` | Taux d'échantillonnage mutex Pyroscope | `5` |
+| `PYROSCOPE_BLOCK_RATE` | Taux d'échantillonnage block Pyroscope | `5` |
+| `HOSTNAME` | Nom d'hôte tagué pour Pyroscope | `new-api` |
 
-📖 **Configuration complète:** [Documentation des variables d'environnement](https://docs.newapi.pro/installation/environment-variables)
+📖 **Configuration complète:** [Documentation des variables d'environnement](https://docs.newapi.pro/en/docs/installation/config-maintenance/environment-variables)
 
 </details>
 
@@ -404,10 +412,10 @@ docker run --name new-api -d --restart always \
 
 | Ressource | Lien |
 |------|------|
-| 📘 FAQ | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 Interaction avec la communauté | [Canaux de communication](https://docs.newapi.pro/support/community-interaction) |
-| 🐛 Commentaires sur les problèmes | [Commentaires sur les problèmes](https://docs.newapi.pro/support/feedback-issues) |
-| 📚 Documentation complète | [Documentation officielle](https://docs.newapi.pro/support) |
+| 📘 FAQ | [FAQ](https://docs.newapi.pro/en/docs/support/faq) |
+| 💬 Interaction avec la communauté | [Canaux de communication](https://docs.newapi.pro/en/docs/support/community-interaction) |
+| 🐛 Commentaires sur les problèmes | [Commentaires sur les problèmes](https://docs.newapi.pro/en/docs/support/feedback-issues) |
+| 📚 Documentation complète | [Documentation officielle](https://docs.newapi.pro/en/docs) |
 
 ### 🤝 Guide de contribution
 
@@ -436,7 +444,7 @@ Bienvenue à toutes les formes de contribution!
 
 Si ce projet vous est utile, bienvenue à nous donner une ⭐️ Étoile！
 
-**[Documentation officielle](https://docs.newapi.pro/)** • **[Commentaires sur les problèmes](https://github.com/Calcium-Ion/new-api/issues)** • **[Dernière version](https://github.com/Calcium-Ion/new-api/releases)**
+**[Documentation officielle](https://docs.newapi.pro/en/docs)** • **[Commentaires sur les problèmes](https://github.com/Calcium-Ion/new-api/issues)** • **[Dernière version](https://github.com/Calcium-Ion/new-api/releases)**
 
 <sub>Construit avec ❤️ par QuantumNous</sub>
 

README.ja.md

@@ -146,7 +146,7 @@ docker run --name new-api -d --restart always \
 
 🎉 デプロイが完了したら、`http://localhost:3000` にアクセスして使用を開始してください！
 
-📖 その他のデプロイ方法については[デプロイガイド](https://docs.newapi.pro/installation)を参照してください。
+📖 その他のデプロイ方法については[デプロイガイド](https://docs.newapi.pro/ja/docs/installation)を参照してください。
 
 ---
 
@@ -154,7 +154,7 @@ docker run --name new-api -d --restart always \
 
 <div align="center">
 
-### 📖 [公式ドキュメント](https://docs.newapi.pro/) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
+### 📖 [公式ドキュメント](https://docs.newapi.pro/ja/docs) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
 
 </div>
 
@@ -162,17 +162,17 @@ docker run --name new-api -d --restart always \
 
 | カテゴリ | リンク |
 |------|------|
-| 🚀 デプロイガイド | [インストールドキュメント](https://docs.newapi.pro/installation) |
-| ⚙️ 環境設定 | [環境変数](https://docs.newapi.pro/installation/environment-variables) |
-| 📡 APIドキュメント | [APIドキュメント](https://docs.newapi.pro/api) |
-| ❓ よくある質問 | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 コミュニティ交流 | [交流チャネル](https://docs.newapi.pro/support/community-interaction) |
+| 🚀 デプロイガイド | [インストールドキュメント](https://docs.newapi.pro/ja/docs/installation) |
+| ⚙️ 環境設定 | [環境変数](https://docs.newapi.pro/ja/docs/installation/config-maintenance/environment-variables) |
+| 📡 APIドキュメント | [APIドキュメント](https://docs.newapi.pro/ja/docs/api) |
+| ❓ よくある質問 | [FAQ](https://docs.newapi.pro/ja/docs/support/faq) |
+| 💬 コミュニティ交流 | [交流チャネル](https://docs.newapi.pro/ja/docs/support/community-interaction) |
 
 ---
 
 ## ✨ 主な機能
 
-> 詳細な機能については[機能説明](https://docs.newapi.pro/wiki/features-introduction)を参照してください。
+> 詳細な機能については[機能説明](https://docs.newapi.pro/ja/docs/guide/wiki/basic-concepts/features-introduction)を参照してください。
 
 ### 🎨 コア機能
 
@@ -202,15 +202,15 @@ docker run --name new-api -d --restart always \
 ### 🚀 高度な機能
 
 **APIフォーマットサポート:**
-- ⚡ [OpenAI Responses](https://docs.newapi.pro/api/openai-responses)
-- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/api/openai-realtime)（Azureを含む）
-- ⚡ [Claude Messages](https://docs.newapi.pro/api/anthropic-chat)
-- ⚡ [Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
-- 🔄 [Rerankモデル](https://docs.newapi.pro/api/jinaai-rerank)
-- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/api/openai-realtime)
-- ⚡ [Claude Messages](https://docs.newapi.pro/api/anthropic-chat)
-- ⚡ [Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
-- 🔄 [Rerankモデル](https://docs.newapi.pro/api/jinaai-rerank)（Cohere、Jina）
+- ⚡ [OpenAI Responses](https://docs.newapi.pro/ja/docs/api/ai-model/chat/openai/create-response)
+- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/ja/docs/api/ai-model/realtime/create-realtime-session)（Azureを含む）
+- ⚡ [Claude Messages](https://docs.newapi.pro/ja/docs/api/ai-model/chat/create-message)
+- ⚡ [Google Gemini](https://doc.newapi.pro/ja/api/google-gemini-chat)
+- 🔄 [Rerankモデル](https://docs.newapi.pro/ja/docs/api/ai-model/rerank/create-rerank)
+- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/ja/docs/api/ai-model/realtime/create-realtime-session)
+- ⚡ [Claude Messages](https://docs.newapi.pro/ja/docs/api/ai-model/chat/create-message)
+- ⚡ [Google Gemini](https://doc.newapi.pro/ja/api/google-gemini-chat)
+- 🔄 [Rerankモデル](https://docs.newapi.pro/ja/docs/api/ai-model/rerank/create-rerank)（Cohere、Jina）
 
 **インテリジェントルーティング:**
 - ⚖️ チャネル重み付けランダム
@@ -251,16 +251,16 @@ docker run --name new-api -d --restart always \
 
 ## 🤖 モデルサポート
 
-> 詳細については[APIドキュメント - 中継インターフェース](https://docs.newapi.pro/api)
+> 詳細については[APIドキュメント - 中継インターフェース](https://docs.newapi.pro/ja/docs/api)
 
 | モデルタイプ | 説明 | ドキュメント |
 |---------|------|------|
 | 🤖 OpenAI GPTs | gpt-4-gizmo-* シリーズ | - |
-| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [ドキュメント](https://docs.newapi.pro/api/midjourney-proxy-image) |
-| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [ドキュメント](https://docs.newapi.pro/api/suno-music) |
-| 🔄 Rerank | Cohere、Jina | [ドキュメント](https://docs.newapi.pro/api/jinaai-rerank) |
-| 💬 Claude | Messagesフォーマット | [ドキュメント](https://docs.newapi.pro/api/suno-music) |
-| 🌐 Gemini | Google Geminiフォーマット | [ドキュメント](https://docs.newapi.pro/api/google-gemini-chat/) |
+| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [ドキュメント](https://doc.newapi.pro/ja/api/midjourney-proxy-image) |
+| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [ドキュメント](https://doc.newapi.pro/ja/api/suno-music) |
+| 🔄 Rerank | Cohere、Jina | [ドキュメント](https://docs.newapi.pro/ja/docs/api/ai-model/rerank/create-rerank) |
+| 💬 Claude | Messagesフォーマット | [ドキュメント](https://docs.newapi.pro/ja/docs/api/ai-model/chat/create-message) |
+| 🌐 Gemini | Google Geminiフォーマット | [ドキュメント](https://doc.newapi.pro/ja/api/google-gemini-chat) |
 | 🔧 Dify | ChatFlowモード | - |
 | 🎯 カスタム | 完全な呼び出しアドレスの入力をサポート | - |
 
@@ -269,16 +269,16 @@ docker run --name new-api -d --restart always \
 <details>
 <summary>完全なインターフェースリストを表示</summary>
 
-- [チャットインターフェース (Chat Completions)](https://docs.newapi.pro/api/openai-chat)
-- [レスポンスインターフェース (Responses)](https://docs.newapi.pro/api/openai-responses)
-- [イメージインターフェース (Image)](https://docs.newapi.pro/api/openai-image)
-- [オーディオインターフェース (Audio)](https://docs.newapi.pro/api/openai-audio)
-- [ビデオインターフェース (Video)](https://docs.newapi.pro/api/openai-video)
-- [エンベッドインターフェース (Embeddings)](https://docs.newapi.pro/api/openai-embeddings)
-- [再ランク付けインターフェース (Rerank)](https://docs.newapi.pro/api/jinaai-rerank)
-- [リアルタイム対話インターフェース (Realtime)](https://docs.newapi.pro/api/openai-realtime)
-- [Claudeチャット](https://docs.newapi.pro/api/anthropic-chat)
-- [Google Geminiチャット](https://docs.newapi.pro/api/google-gemini-chat/)
+- [チャットインターフェース (Chat Completions)](https://docs.newapi.pro/ja/docs/api/ai-model/chat/openai/create-chat-completion)
+- [レスポンスインターフェース (Responses)](https://docs.newapi.pro/ja/docs/api/ai-model/chat/openai/create-response)
+- [イメージインターフェース (Image)](https://docs.newapi.pro/ja/docs/api/ai-model/images/openai/v1-images-generations--post)
+- [オーディオインターフェース (Audio)](https://docs.newapi.pro/ja/docs/api/ai-model/audio/openai/create-transcription)
+- [ビデオインターフェース (Video)](https://docs.newapi.pro/ja/docs/api/ai-model/videos/create-video-generation)
+- [エンベッドインターフェース (Embeddings)](https://docs.newapi.pro/ja/docs/api/ai-model/embeddings/create-embedding)
+- [再ランク付けインターフェース (Rerank)](https://docs.newapi.pro/ja/docs/api/ai-model/rerank/create-rerank)
+- [リアルタイム対話インターフェース (Realtime)](https://docs.newapi.pro/ja/docs/api/ai-model/realtime/create-realtime-session)
+- [Claudeチャット](https://docs.newapi.pro/ja/docs/api/ai-model/chat/create-message)
+- [Google Geminiチャット](https://doc.newapi.pro/ja/api/google-gemini-chat)
 
 </details>
 
@@ -310,10 +310,18 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis接続文字列 | - |
 | `STREAMING_TIMEOUT` | ストリーミング応答のタイムアウト時間（秒） | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | ストリームスキャナの1行あたりバッファ上限（MB）。4K画像など巨大なbase64 `data:` ペイロードを扱う場合は値を増加させてください | `64` |
+| `MAX_REQUEST_BODY_MB` | リクエストボディ最大サイズ（MB、**解凍後**に計測。巨大リクエスト/zip bomb によるメモリ枯渇を防止）。超過時は `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure APIバージョン | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | エラーログスイッチ | `false` |
+| `PYROSCOPE_URL` | Pyroscopeサーバーのアドレス | - |
+| `PYROSCOPE_APP_NAME` | Pyroscopeアプリ名 | `new-api` |
+| `PYROSCOPE_BASIC_AUTH_USER` | Pyroscope Basic Authユーザー | - |
+| `PYROSCOPE_BASIC_AUTH_PASSWORD` | Pyroscope Basic Authパスワード | - |
+| `PYROSCOPE_MUTEX_RATE` | Pyroscope mutexサンプリング率 | `5` |
+| `PYROSCOPE_BLOCK_RATE` | Pyroscope blockサンプリング率 | `5` |
+| `HOSTNAME` | Pyroscope用のホスト名タグ | `new-api` |
 
-📖 **完全な設定:** [環境変数ドキュメント](https://docs.newapi.pro/installation/environment-variables)
+📖 **完全な設定:** [環境変数ドキュメント](https://docs.newapi.pro/ja/docs/installation/config-maintenance/environment-variables)
 
 </details>
 
@@ -413,10 +421,10 @@ docker run --name new-api -d --restart always \
 
 | リソース | リンク |
 |------|------|
-| 📘 よくある質問 | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 コミュニティ交流 | [交流チャネル](https://docs.newapi.pro/support/community-interaction) |
-| 🐛 問題のフィードバック | [問題フィードバック](https://docs.newapi.pro/support/feedback-issues) |
-| 📚 完全なドキュメント | [公式ドキュメント](https://docs.newapi.pro/support) |
+| 📘 よくある質問 | [FAQ](https://docs.newapi.pro/ja/docs/support/faq) |
+| 💬 コミュニティ交流 | [交流チャネル](https://docs.newapi.pro/ja/docs/support/community-interaction) |
+| 🐛 問題のフィードバック | [問題フィードバック](https://docs.newapi.pro/ja/docs/support/feedback-issues) |
+| 📚 完全なドキュメント | [公式ドキュメント](https://docs.newapi.pro/ja/docs) |
 
 ### 🤝 貢献ガイド
 
@@ -445,7 +453,7 @@ docker run --name new-api -d --restart always \
 
 このプロジェクトがあなたのお役に立てたなら、ぜひ ⭐️ スターをください！
 
-**[公式ドキュメント](https://docs.newapi.pro/)** • **[問題フィードバック](https://github.com/Calcium-Ion/new-api/issues)** • **[最新リリース](https://github.com/Calcium-Ion/new-api/releases)**
+**[公式ドキュメント](https://docs.newapi.pro/ja/docs)** • **[問題フィードバック](https://github.com/Calcium-Ion/new-api/issues)** • **[最新リリース](https://github.com/Calcium-Ion/new-api/releases)**
 
 <sub>❤️ で構築された QuantumNous</sub>
 

README.md

@@ -146,7 +146,7 @@ docker run --name new-api -d --restart always \
 
 🎉 部署完成后，访问 `http://localhost:3000` 即可使用！
 
-📖 更多部署方式请参考 [部署指南](https://docs.newapi.pro/installation)
+📖 更多部署方式请参考 [部署指南](https://docs.newapi.pro/zh/docs/installation)
 
 ---
 
@@ -154,7 +154,7 @@ docker run --name new-api -d --restart always \
 
 <div align="center">
 
-### 📖 [官方文档](https://docs.newapi.pro/) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
+### 📖 [官方文档](https://docs.newapi.pro/zh/docs) | [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/QuantumNous/new-api)
 
 </div>
 
@@ -162,17 +162,17 @@ docker run --name new-api -d --restart always \
 
 | 分类 | 链接 |
 |------|------|
-| 🚀 部署指南 | [安装文档](https://docs.newapi.pro/installation) |
-| ⚙️ 环境配置 | [环境变量](https://docs.newapi.pro/installation/environment-variables) |
-| 📡 接口文档 | [API 文档](https://docs.newapi.pro/api) |
-| ❓ 常见问题 | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 社区交流 | [交流渠道](https://docs.newapi.pro/support/community-interaction) |
+| 🚀 部署指南 | [安装文档](https://docs.newapi.pro/zh/docs/installation) |
+| ⚙️ 环境配置 | [环境变量](https://docs.newapi.pro/zh/docs/installation/config-maintenance/environment-variables) |
+| 📡 接口文档 | [API 文档](https://docs.newapi.pro/zh/docs/api) |
+| ❓ 常见问题 | [FAQ](https://docs.newapi.pro/zh/docs/support/faq) |
+| 💬 社区交流 | [交流渠道](https://docs.newapi.pro/zh/docs/support/community-interaction) |
 
 ---
 
 ## ✨ 主要特性
 
-> 详细特性请参考 [特性说明](https://docs.newapi.pro/wiki/features-introduction)
+> 详细特性请参考 [特性说明](https://docs.newapi.pro/zh/docs/guide/wiki/basic-concepts/features-introduction)
 
 ### 🎨 核心功能
 
@@ -202,11 +202,11 @@ docker run --name new-api -d --restart always \
 ### 🚀 高级功能
 
 **API 格式支持：**
-- ⚡ [OpenAI Responses](https://docs.newapi.pro/api/openai-responses)
-- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/api/openai-realtime)（含 Azure）
-- ⚡ [Claude Messages](https://docs.newapi.pro/api/anthropic-chat)
-- ⚡ [Google Gemini](https://docs.newapi.pro/api/google-gemini-chat/)
-- 🔄 [Rerank 模型](https://docs.newapi.pro/api/jinaai-rerank)（Cohere、Jina）
+- ⚡ [OpenAI Responses](https://docs.newapi.pro/zh/docs/api/ai-model/chat/openai/create-response)
+- ⚡ [OpenAI Realtime API](https://docs.newapi.pro/zh/docs/api/ai-model/realtime/create-realtime-session)（含 Azure）
+- ⚡ [Claude Messages](https://docs.newapi.pro/zh/docs/api/ai-model/chat/create-message)
+- ⚡ [Google Gemini](https://doc.newapi.pro/api/google-gemini-chat)
+- 🔄 [Rerank 模型](https://docs.newapi.pro/zh/docs/api/ai-model/rerank/create-rerank)（Cohere、Jina）
 
 **智能路由：**
 - ⚖️ 渠道加权随机
@@ -247,16 +247,16 @@ docker run --name new-api -d --restart always \
 
 ## 🤖 模型支持
 
-> 详情请参考 [接口文档 - 中继接口](https://docs.newapi.pro/api)
+> 详情请参考 [接口文档 - 中继接口](https://docs.newapi.pro/zh/docs/api)
 
 | 模型类型 | 说明 | 文档 |
 |---------|------|------|
 | 🤖 OpenAI GPTs | gpt-4-gizmo-* 系列 | - |
-| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [文档](https://docs.newapi.pro/api/midjourney-proxy-image) |
-| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [文档](https://docs.newapi.pro/api/suno-music) |
-| 🔄 Rerank | Cohere、Jina | [文档](https://docs.newapi.pro/api/jinaai-rerank) |
-| 💬 Claude | Messages 格式 | [文档](https://docs.newapi.pro/api/anthropic-chat) |
-| 🌐 Gemini | Google Gemini 格式 | [文档](https://docs.newapi.pro/api/google-gemini-chat/) |
+| 🎨 Midjourney-Proxy | [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) | [文档](https://doc.newapi.pro/api/midjourney-proxy-image) |
+| 🎵 Suno-API | [Suno API](https://github.com/Suno-API/Suno-API) | [文档](https://doc.newapi.pro/api/suno-music) |
+| 🔄 Rerank | Cohere、Jina | [文档](https://docs.newapi.pro/zh/docs/api/ai-model/rerank/create-rerank) |
+| 💬 Claude | Messages 格式 | [文档](https://docs.newapi.pro/zh/docs/api/ai-model/chat/create-message) |
+| 🌐 Gemini | Google Gemini 格式 | [文档](https://doc.newapi.pro/api/google-gemini-chat) |
 | 🔧 Dify | ChatFlow 模式 | - |
 | 🎯 自定义 | 支持完整调用地址 | - |
 
@@ -265,16 +265,16 @@ docker run --name new-api -d --restart always \
 <details>
 <summary>查看完整接口列表</summary>
 
-- [聊天接口 (Chat Completions)](https://docs.newapi.pro/api/openai-chat)
-- [响应接口 (Responses)](https://docs.newapi.pro/api/openai-responses)
-- [图像接口 (Image)](https://docs.newapi.pro/api/openai-image)
-- [音频接口 (Audio)](https://docs.newapi.pro/api/openai-audio)
-- [视频接口 (Video)](https://docs.newapi.pro/api/openai-video)
-- [嵌入接口 (Embeddings)](https://docs.newapi.pro/api/openai-embeddings)
-- [重排序接口 (Rerank)](https://docs.newapi.pro/api/jinaai-rerank)
-- [实时对话 (Realtime)](https://docs.newapi.pro/api/openai-realtime)
-- [Claude 聊天](https://docs.newapi.pro/api/anthropic-chat)
-- [Google Gemini 聊天](https://docs.newapi.pro/api/google-gemini-chat)
+- [聊天接口 (Chat Completions)](https://docs.newapi.pro/zh/docs/api/ai-model/chat/openai/create-chat-completion)
+- [响应接口 (Responses)](https://docs.newapi.pro/zh/docs/api/ai-model/chat/openai/create-response)
+- [图像接口 (Image)](https://docs.newapi.pro/zh/docs/api/ai-model/images/openai/v1-images-generations--post)
+- [音频接口 (Audio)](https://docs.newapi.pro/zh/docs/api/ai-model/audio/openai/create-transcription)
+- [视频接口 (Video)](https://docs.newapi.pro/zh/docs/api/ai-model/videos/create-video-generation)
+- [嵌入接口 (Embeddings)](https://docs.newapi.pro/zh/docs/api/ai-model/embeddings/create-embedding)
+- [重排序接口 (Rerank)](https://docs.newapi.pro/zh/docs/api/ai-model/rerank/create-rerank)
+- [实时对话 (Realtime)](https://docs.newapi.pro/zh/docs/api/ai-model/realtime/create-realtime-session)
+- [Claude 聊天](https://docs.newapi.pro/zh/docs/api/ai-model/chat/create-message)
+- [Google Gemini 聊天](https://doc.newapi.pro/api/google-gemini-chat)
 
 </details>
 
@@ -306,10 +306,18 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis 连接字符串                                                  | - |
 | `STREAMING_TIMEOUT` | 流式超时时间（秒）                                                    | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | 流式扫描器单行最大缓冲（MB），图像生成等超大 `data:` 片段（如 4K 图片 base64）需适当调大 | `64` |
+| `MAX_REQUEST_BODY_MB` | 请求体最大大小（MB，**解压后**计；防止超大请求/zip bomb 导致内存暴涨），超过将返回 `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure API 版本                                                 | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | 错误日志开关                                                       | `false` |
+| `PYROSCOPE_URL` | Pyroscope 服务地址                                            | - |
+| `PYROSCOPE_APP_NAME` | Pyroscope 应用名                                        | `new-api` |
+| `PYROSCOPE_BASIC_AUTH_USER` | Pyroscope Basic Auth 用户名                        | - |
+| `PYROSCOPE_BASIC_AUTH_PASSWORD` | Pyroscope Basic Auth 密码                  | - |
+| `PYROSCOPE_MUTEX_RATE` | Pyroscope mutex 采样率                               | `5` |
+| `PYROSCOPE_BLOCK_RATE` | Pyroscope block 采样率                               | `5` |
+| `HOSTNAME` | Pyroscope 标签里的主机名                                          | `new-api` |
 
-📖 **完整配置：** [环境变量文档](https://docs.newapi.pro/installation/environment-variables)
+📖 **完整配置：** [环境变量文档](https://docs.newapi.pro/zh/docs/installation/config-maintenance/environment-variables)
 
 </details>
 
@@ -411,10 +419,10 @@ docker run --name new-api -d --restart always \
 
 | 资源 | 链接 |
 |------|------|
-| 📘 常见问题 | [FAQ](https://docs.newapi.pro/support/faq) |
-| 💬 社区交流 | [交流渠道](https://docs.newapi.pro/support/community-interaction) |
-| 🐛 反馈问题 | [问题反馈](https://docs.newapi.pro/support/feedback-issues) |
-| 📚 完整文档 | [官方文档](https://docs.newapi.pro/support) |
+| 📘 常见问题 | [FAQ](https://docs.newapi.pro/zh/docs/support/faq) |
+| 💬 社区交流 | [交流渠道](https://docs.newapi.pro/zh/docs/support/community-interaction) |
+| 🐛 反馈问题 | [问题反馈](https://docs.newapi.pro/zh/docs/support/feedback-issues) |
+| 📚 完整文档 | [官方文档](https://docs.newapi.pro/zh/docs) |
 
 ### 🤝 贡献指南
 
@@ -443,7 +451,7 @@ docker run --name new-api -d --restart always \
 
 如果这个项目对你有帮助，欢迎给我们一个 ⭐️ Star！
 
-**[官方文档](https://docs.newapi.pro/)** • **[问题反馈](https://github.com/Calcium-Ion/new-api/issues)** • **[最新发布](https://github.com/Calcium-Ion/new-api/releases)**
+**[官方文档](https://docs.newapi.pro/zh/docs)** • **[问题反馈](https://github.com/Calcium-Ion/new-api/issues)** • **[最新发布](https://github.com/Calcium-Ion/new-api/releases)**
 
 <sub>Built with ❤️ by QuantumNous</sub>
 

common/audio.go

@@ -71,15 +71,66 @@ func getMP3Duration(r io.Reader) (float64, error) {
 
 // getWAVDuration 解析 WAV 文件头以获取时长。
 func getWAVDuration(r io.ReadSeeker) (float64, error) {
+	// 1. 强制复位指针
+	r.Seek(0, io.SeekStart)
+
 	dec := wav.NewDecoder(r)
+
+	// IsValidFile 会读取 fmt 块
 	if !dec.IsValidFile() {
 		return 0, errors.New("invalid wav file")
 	}
-	d, err := dec.Duration()
-	if err != nil {
-		return 0, errors.Wrap(err, "failed to get wav duration")
+
+	// 尝试寻找 data 块
+	if err := dec.FwdToPCM(); err != nil {
+		return 0, errors.Wrap(err, "failed to find PCM data chunk")
 	}
-	return d.Seconds(), nil
+
+	pcmSize := int64(dec.PCMSize)
+
+	// 如果读出来的 Size 是 0，尝试用文件大小反推
+	if pcmSize == 0 {
+		// 获取文件总大小
+		currentPos, _ := r.Seek(0, io.SeekCurrent) // 当前通常在 data chunk header 之后
+		endPos, _ := r.Seek(0, io.SeekEnd)
+		fileSize := endPos
+
+		// 恢复位置（虽然如果不继续读也没关系）
+		r.Seek(currentPos, io.SeekStart)
+
+		// 数据区大小 ≈ 文件总大小 - 当前指针位置(即Header大小)
+		// 注意：FwdToPCM 成功后，CurrentPos 应该刚好指向 Data 区数据的开始
+		// 或者是 Data Chunk ID + Size 之后。
+		// WAV Header 一般 44 字节。
+		if fileSize > 44 {
+			// 如果 FwdToPCM 成功，Reader 应该位于 data 块的数据起始处
+			// 所以剩余的所有字节理论上都是音频数据
+			pcmSize = fileSize - currentPos
+
+			// 简单的兜底：如果算出来还是负数或0，强制按文件大小-44计算
+			if pcmSize <= 0 {
+				pcmSize = fileSize - 44
+			}
+		}
+	}
+
+	numChans := int64(dec.NumChans)
+	bitDepth := int64(dec.BitDepth)
+	sampleRate := float64(dec.SampleRate)
+
+	if sampleRate == 0 || numChans == 0 || bitDepth == 0 {
+		return 0, errors.New("invalid wav header metadata")
+	}
+
+	bytesPerFrame := numChans * (bitDepth / 8)
+	if bytesPerFrame == 0 {
+		return 0, errors.New("invalid byte depth calculation")
+	}
+
+	totalFrames := pcmSize / bytesPerFrame
+
+	durationSeconds := float64(totalFrames) / sampleRate
+	return durationSeconds, nil
 }
 
 // getFLACDuration 解析 FLAC 文件的 STREAMINFO 块。

common/email.go

@@ -32,7 +32,7 @@ func SendEmail(subject string, receiver string, content string) error {
 	}
 	encodedSubject := fmt.Sprintf("=?UTF-8?B?%s?=", base64.StdEncoding.EncodeToString([]byte(subject)))
 	mail := []byte(fmt.Sprintf("To: %s\r\n"+
-		"From: %s<%s>\r\n"+
+		"From: %s <%s>\r\n"+
 		"Subject: %s\r\n"+
 		"Date: %s\r\n"+
 		"Message-ID: %s\r\n"+ // 添加 Message-ID 头

common/gin.go

@@ -2,7 +2,7 @@ package common
 
 import (
 	"bytes"
-	"errors"
+	"fmt"
 	"io"
 	"mime"
 	"mime/multipart"
@@ -12,24 +12,61 @@ import (
 	"time"
 
 	"github.com/QuantumNous/new-api/constant"
+	"github.com/pkg/errors"
 
 	"github.com/gin-gonic/gin"
 )
 
 const KeyRequestBody = "key_request_body"
 
-func GetRequestBody(c *gin.Context) ([]byte, error) {
-	requestBody, _ := c.Get(KeyRequestBody)
-	if requestBody != nil {
-		return requestBody.([]byte), nil
+var ErrRequestBodyTooLarge = errors.New("request body too large")
+
+func IsRequestBodyTooLargeError(err error) bool {
+	if err == nil {
+		return false
 	}
-	requestBody, err := io.ReadAll(c.Request.Body)
+	if errors.Is(err, ErrRequestBodyTooLarge) {
+		return true
+	}
+	var mbe *http.MaxBytesError
+	return errors.As(err, &mbe)
+}
+
+func GetRequestBody(c *gin.Context) ([]byte, error) {
+	cached, exists := c.Get(KeyRequestBody)
+	if exists && cached != nil {
+		if b, ok := cached.([]byte); ok {
+			return b, nil
+		}
+	}
+	maxMB := constant.MaxRequestBodyMB
+	if maxMB < 0 {
+		// no limit
+		body, err := io.ReadAll(c.Request.Body)
+		_ = c.Request.Body.Close()
+		if err != nil {
+			return nil, err
+		}
+		c.Set(KeyRequestBody, body)
+		return body, nil
+	}
+	maxBytes := int64(maxMB) << 20
+
+	limited := io.LimitReader(c.Request.Body, maxBytes+1)
+	body, err := io.ReadAll(limited)
 	if err != nil {
+		_ = c.Request.Body.Close()
+		if IsRequestBodyTooLargeError(err) {
+			return nil, errors.Wrap(ErrRequestBodyTooLarge, fmt.Sprintf("request body exceeds %d MB", maxMB))
+		}
 		return nil, err
 	}
 	_ = c.Request.Body.Close()
-	c.Set(KeyRequestBody, requestBody)
-	return requestBody.([]byte), nil
+	if int64(len(body)) > maxBytes {
+		return nil, errors.Wrap(ErrRequestBodyTooLarge, fmt.Sprintf("request body exceeds %d MB", maxMB))
+	}
+	c.Set(KeyRequestBody, body)
+	return body, nil
 }
 
 func UnmarshalBodyReusable(c *gin.Context, v any) error {

common/init.go

@@ -117,6 +117,8 @@ func initConstantEnv() {
 	constant.DifyDebug = GetEnvOrDefaultBool("DIFY_DEBUG", true)
 	constant.MaxFileDownloadMB = GetEnvOrDefault("MAX_FILE_DOWNLOAD_MB", 20)
 	constant.StreamScannerMaxBufferMB = GetEnvOrDefault("STREAM_SCANNER_MAX_BUFFER_MB", 64)
+	// MaxRequestBodyMB 请求体最大大小（解压后），用于防止超大请求/zip bomb导致内存暴涨
+	constant.MaxRequestBodyMB = GetEnvOrDefault("MAX_REQUEST_BODY_MB", 64)
 	// ForceStreamOption 覆盖请求参数，强制返回usage信息
 	constant.ForceStreamOption = GetEnvOrDefaultBool("FORCE_STREAM_OPTION", true)
 	constant.CountToken = GetEnvOrDefaultBool("CountToken", true)

common/ip.go

@@ -2,6 +2,15 @@ package common
 
 import "net"
 
+func IsIP(s string) bool {
+	ip := net.ParseIP(s)
+	return ip != nil
+}
+
+func ParseIP(s string) net.IP {
+	return net.ParseIP(s)
+}
+
 func IsPrivateIP(ip net.IP) bool {
 	if ip.IsLoopback() || ip.IsLinkLocalUnicast() || ip.IsLinkLocalMulticast() {
 		return true
@@ -20,3 +29,23 @@ func IsPrivateIP(ip net.IP) bool {
 	}
 	return false
 }
+
+func IsIpInCIDRList(ip net.IP, cidrList []string) bool {
+	for _, cidr := range cidrList {
+		_, network, err := net.ParseCIDR(cidr)
+		if err != nil {
+			// 尝试作为单个IP处理
+			if whitelistIP := net.ParseIP(cidr); whitelistIP != nil {
+				if ip.Equal(whitelistIP) {
+					return true
+				}
+			}
+			continue
+		}
+
+		if network.Contains(ip) {
+			return true
+		}
+	}
+	return false
+}

common/pyro.go

@@ -0,0 +1,56 @@
+package common
+
+import (
+	"runtime"
+
+	"github.com/grafana/pyroscope-go"
+)
+
+func StartPyroScope() error {
+
+	pyroscopeUrl := GetEnvOrDefaultString("PYROSCOPE_URL", "")
+	if pyroscopeUrl == "" {
+		return nil
+	}
+
+	pyroscopeAppName := GetEnvOrDefaultString("PYROSCOPE_APP_NAME", "new-api")
+	pyroscopeBasicAuthUser := GetEnvOrDefaultString("PYROSCOPE_BASIC_AUTH_USER", "")
+	pyroscopeBasicAuthPassword := GetEnvOrDefaultString("PYROSCOPE_BASIC_AUTH_PASSWORD", "")
+	pyroscopeHostname := GetEnvOrDefaultString("HOSTNAME", "new-api")
+
+	mutexRate := GetEnvOrDefault("PYROSCOPE_MUTEX_RATE", 5)
+	blockRate := GetEnvOrDefault("PYROSCOPE_BLOCK_RATE", 5)
+
+	runtime.SetMutexProfileFraction(mutexRate)
+	runtime.SetBlockProfileRate(blockRate)
+
+	_, err := pyroscope.Start(pyroscope.Config{
+		ApplicationName: pyroscopeAppName,
+
+		ServerAddress:     pyroscopeUrl,
+		BasicAuthUser:     pyroscopeBasicAuthUser,
+		BasicAuthPassword: pyroscopeBasicAuthPassword,
+
+		Logger: nil,
+
+		Tags: map[string]string{"hostname": pyroscopeHostname},
+
+		ProfileTypes: []pyroscope.ProfileType{
+			pyroscope.ProfileCPU,
+			pyroscope.ProfileAllocObjects,
+			pyroscope.ProfileAllocSpace,
+			pyroscope.ProfileInuseObjects,
+			pyroscope.ProfileInuseSpace,
+
+			pyroscope.ProfileGoroutines,
+			pyroscope.ProfileMutexCount,
+			pyroscope.ProfileMutexDuration,
+			pyroscope.ProfileBlockCount,
+			pyroscope.ProfileBlockDuration,
+		},
+	})
+	if err != nil {
+		return err
+	}
+	return nil
+}

common/ssrf_protection.go

@@ -186,23 +186,7 @@ func isIPListed(ip net.IP, list []string) bool {
 		return false
 	}
 
-	for _, whitelistCIDR := range list {
-		_, network, err := net.ParseCIDR(whitelistCIDR)
-		if err != nil {
-			// 尝试作为单个IP处理
-			if whitelistIP := net.ParseIP(whitelistCIDR); whitelistIP != nil {
-				if ip.Equal(whitelistIP) {
-					return true
-				}
-			}
-			continue
-		}
-
-		if network.Contains(ip) {
-			return true
-		}
-	}
-	return false
+	return IsIpInCIDRList(ip, list)
 }
 
 // IsIPAccessAllowed 检查IP是否允许访问

common/utils.go

@@ -217,11 +217,6 @@ func IntMax(a int, b int) int {
 	}
 }
 
-func IsIP(s string) bool {
-	ip := net.ParseIP(s)
-	return ip != nil
-}
-
 func GetUUID() string {
 	code := uuid.New().String()
 	code = strings.Replace(code, "-", "", -1)

constant/context_key.go

@@ -18,6 +18,7 @@ const (
 	ContextKeyTokenSpecificChannelId ContextKey = "specific_channel_id"
 	ContextKeyTokenModelLimitEnabled ContextKey = "token_model_limit_enabled"
 	ContextKeyTokenModelLimit        ContextKey = "token_model_limit"
+	ContextKeyTokenCrossGroupRetry   ContextKey = "token_cross_group_retry"
 
 	/* channel related keys */
 	ContextKeyChannelId                ContextKey = "channel_id"
@@ -37,6 +38,10 @@ const (
 	ContextKeyChannelMultiKeyIndex     ContextKey = "channel_multi_key_index"
 	ContextKeyChannelKey               ContextKey = "channel_key"
 
+	ContextKeyAutoGroup           ContextKey = "auto_group"
+	ContextKeyAutoGroupIndex      ContextKey = "auto_group_index"
+	ContextKeyAutoGroupRetryIndex ContextKey = "auto_group_retry_index"
+
 	/* user related keys */
 	ContextKeyUserId      ContextKey = "id"
 	ContextKeyUserSetting ContextKey = "user_setting"

constant/env.go

@@ -9,6 +9,7 @@ var CountToken bool
 var GetMediaToken bool
 var GetMediaTokenNotStream bool
 var UpdateTask bool
+var MaxRequestBodyMB int
 var AzureDefaultAPIVersion string
 var GeminiVisionMaxImageNum int
 var NotifyLimitCount int

controller/billing.go

@@ -2,9 +2,9 @@ package controller
 
 import (
 	"github.com/QuantumNous/new-api/common"
-	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/setting/operation_setting"
+	"github.com/QuantumNous/new-api/types"
 	"github.com/gin-gonic/gin"
 )
 
@@ -29,7 +29,7 @@ func GetSubscription(c *gin.Context) {
 		expiredTime = 0
 	}
 	if err != nil {
-		openAIError := dto.OpenAIError{
+		openAIError := types.OpenAIError{
 			Message: err.Error(),
 			Type:    "upstream_error",
 		}
@@ -81,7 +81,7 @@ func GetUsage(c *gin.Context) {
 		quota, err = model.GetUserUsedQuota(userId)
 	}
 	if err != nil {
-		openAIError := dto.OpenAIError{
+		openAIError := types.OpenAIError{
 			Message: err.Error(),
 			Type:    "new_api_error",
 		}

controller/channel-test.go

@@ -97,6 +97,11 @@ func testChannel(channel *model.Channel, testModel string, endpointType string)
 		if channel.Type == constant.ChannelTypeVolcEngine && strings.Contains(testModel, "seedream") {
 			requestPath = "/v1/images/generations"
 		}
+
+		// responses-only models
+		if strings.Contains(strings.ToLower(testModel), "codex") {
+			requestPath = "/v1/responses"
+		}
 	}
 
 	c.Request = &http.Request{
@@ -176,7 +181,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string)
 		}
 	}
 
-	request := buildTestRequest(testModel, endpointType)
+	request := buildTestRequest(testModel, endpointType, channel)
 
 	info, err := relaycommon.GenRelayInfo(c, relayFormat, request, nil)
 
@@ -319,6 +324,16 @@ func testChannel(channel *model.Channel, testModel string, endpointType string)
 		httpResp = resp.(*http.Response)
 		if httpResp.StatusCode != http.StatusOK {
 			err := service.RelayErrorHandler(c.Request.Context(), httpResp, true)
+			common.SysError(fmt.Sprintf(
+				"channel test bad response: channel_id=%d name=%s type=%d model=%s endpoint_type=%s status=%d err=%v",
+				channel.Id,
+				channel.Name,
+				channel.Type,
+				testModel,
+				endpointType,
+				httpResp.StatusCode,
+				err,
+			))
 			return testResult{
 				context:     c,
 				localErr:    err,
@@ -389,7 +404,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string)
 	}
 }
 
-func buildTestRequest(model string, endpointType string) dto.Request {
+func buildTestRequest(model string, endpointType string, channel *model.Channel) dto.Request {
 	// 根据端点类型构建不同的测试请求
 	if endpointType != "" {
 		switch constant.EndpointType(endpointType) {
@@ -423,7 +438,7 @@ func buildTestRequest(model string, endpointType string) dto.Request {
 			}
 		case constant.EndpointTypeAnthropic, constant.EndpointTypeGemini, constant.EndpointTypeOpenAI:
 			// 返回 GeneralOpenAIRequest
-			maxTokens := uint(10)
+			maxTokens := uint(16)
 			if constant.EndpointType(endpointType) == constant.EndpointTypeGemini {
 				maxTokens = 3000
 			}
@@ -453,6 +468,14 @@ func buildTestRequest(model string, endpointType string) dto.Request {
 		}
 	}
 
+	// Responses-only models (e.g. codex series)
+	if strings.Contains(strings.ToLower(model), "codex") {
+		return &dto.OpenAIResponsesRequest{
+			Model: model,
+			Input: json.RawMessage("\"hi\""),
+		}
+	}
+
 	// Chat/Completion 请求 - 返回 GeneralOpenAIRequest
 	testRequest := &dto.GeneralOpenAIRequest{
 		Model:  model,
@@ -466,7 +489,7 @@ func buildTestRequest(model string, endpointType string) dto.Request {
 	}
 
 	if strings.HasPrefix(model, "o") {
-		testRequest.MaxCompletionTokens = 10
+		testRequest.MaxCompletionTokens = 16
 	} else if strings.Contains(model, "thinking") {
 		if !strings.Contains(model, "claude") {
 			testRequest.MaxTokens = 50
@@ -474,7 +497,7 @@ func buildTestRequest(model string, endpointType string) dto.Request {
 	} else if strings.Contains(model, "gemini") {
 		testRequest.MaxTokens = 3000
 	} else {
-		testRequest.MaxTokens = 10
+		testRequest.MaxTokens = 16
 	}
 
 	return testRequest

controller/discord.go

@@ -114,7 +114,7 @@ func DiscordOAuth(c *gin.Context) {
 		DiscordBind(c)
 		return
 	}
-		if !system_setting.GetDiscordSettings().Enabled {
+	if !system_setting.GetDiscordSettings().Enabled {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
 			"message": "管理员未开启通过 Discord 登录以及注册",

controller/model.go

@@ -18,6 +18,7 @@ import (
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/operation_setting"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
+	"github.com/QuantumNous/new-api/types"
 	"github.com/gin-gonic/gin"
 	"github.com/samber/lo"
 )
@@ -275,7 +276,7 @@ func RetrieveModel(c *gin.Context, modelType int) {
 			c.JSON(200, aiModel)
 		}
 	} else {
-		openAIError := dto.OpenAIError{
+		openAIError := types.OpenAIError{
 			Message: fmt.Sprintf("The model '%s' does not exist", modelId),
 			Type:    "invalid_request_error",
 			Param:   "model",

controller/model_sync.go

@@ -249,7 +249,9 @@ func ensureVendorID(vendorName string, vendorByName map[string]upstreamVendor, v
 	return 0
 }
 
-// SyncUpstreamModels 同步上游模型与供应商，仅对「未配置模型」生效
+// SyncUpstreamModels 同步上游模型与供应商：
+// - 默认仅创建「未配置模型」
+// - 可通过 overwrite 选择性覆盖更新本地已有模型的字段（前提：sync_official <> 0）
 func SyncUpstreamModels(c *gin.Context) {
 	var req syncRequest
 	// 允许空体
@@ -260,12 +262,26 @@ func SyncUpstreamModels(c *gin.Context) {
 		c.JSON(http.StatusOK, gin.H{"success": false, "message": err.Error()})
 		return
 	}
-	if len(missing) == 0 {
-		c.JSON(http.StatusOK, gin.H{"success": true, "data": gin.H{
-			"created_models":  0,
-			"created_vendors": 0,
-			"skipped_models":  []string{},
-		}})
+
+	// 若既无缺失模型需要创建，也未指定覆盖更新字段，则无需请求上游数据，直接返回
+	if len(missing) == 0 && len(req.Overwrite) == 0 {
+		modelsURL, vendorsURL := getUpstreamURLs(req.Locale)
+		c.JSON(http.StatusOK, gin.H{
+			"success": true,
+			"data": gin.H{
+				"created_models":  0,
+				"created_vendors": 0,
+				"updated_models":  0,
+				"skipped_models":  []string{},
+				"created_list":    []string{},
+				"updated_list":    []string{},
+				"source": gin.H{
+					"locale":      req.Locale,
+					"models_url":  modelsURL,
+					"vendors_url": vendorsURL,
+				},
+			},
+		})
 		return
 	}
 
@@ -315,9 +331,9 @@ func SyncUpstreamModels(c *gin.Context) {
 	createdModels := 0
 	createdVendors := 0
 	updatedModels := 0
-	var skipped []string
-	var createdList []string
-	var updatedList []string
+	skipped := make([]string, 0)
+	createdList := make([]string, 0)
+	updatedList := make([]string, 0)
 
 	// 本地缓存：vendorName -> id
 	vendorIDCache := make(map[string]int)

controller/playground.go

@@ -3,12 +3,10 @@ package controller
 import (
 	"errors"
 	"fmt"
-	"time"
 
-	"github.com/QuantumNous/new-api/common"
-	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/middleware"
 	"github.com/QuantumNous/new-api/model"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -31,8 +29,11 @@ func Playground(c *gin.Context) {
 		return
 	}
 
-	group := common.GetContextKeyString(c, constant.ContextKeyUsingGroup)
-	modelName := c.GetString("original_model")
+	relayInfo, err := relaycommon.GenRelayInfo(c, types.RelayFormatOpenAI, nil, nil)
+	if err != nil {
+		newAPIError = types.NewError(err, types.ErrorCodeInvalidRequest, types.ErrOptionWithSkipRetry())
+		return
+	}
 
 	userId := c.GetInt("id")
 
@@ -46,16 +47,10 @@ func Playground(c *gin.Context) {
 
 	tempToken := &model.Token{
 		UserId: userId,
-		Name:   fmt.Sprintf("playground-%s", group),
-		Group:  group,
+		Name:   fmt.Sprintf("playground-%s", relayInfo.UsingGroup),
+		Group:  relayInfo.UsingGroup,
 	}
 	_ = middleware.SetupContextForToken(c, tempToken)
-	_, newAPIError = getChannel(c, group, modelName, 0)
-	if newAPIError != nil {
-		return
-	}
-	//middleware.SetupContextForSelectedChannel(c, channel, playgroundRequest.Model)
-	common.SetContextKey(c, constant.ContextKeyRequestStartTime, time.Now())
 
 	Relay(c, types.RelayFormatOpenAI)
 }

controller/relay.go

@@ -2,6 +2,7 @@ package controller
 
 import (
 	"bytes"
+	"errors"
 	"fmt"
 	"io"
 	"log"
@@ -64,8 +65,8 @@ func geminiRelayHandler(c *gin.Context, info *relaycommon.RelayInfo) *types.NewA
 func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 	requestId := c.GetString(common.RequestIdKey)
-	group := common.GetContextKeyString(c, constant.ContextKeyUsingGroup)
-	originalModel := common.GetContextKeyString(c, constant.ContextKeyOriginalModel)
+	//group := common.GetContextKeyString(c, constant.ContextKeyUsingGroup)
+	//originalModel := common.GetContextKeyString(c, constant.ContextKeyOriginalModel)
 
 	var (
 		newAPIError *types.NewAPIError
@@ -104,7 +105,12 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 	request, err := helper.GetAndValidateRequest(c, relayFormat)
 	if err != nil {
-		newAPIError = types.NewError(err, types.ErrorCodeInvalidRequest)
+		// Map "request body too large" to 413 so clients can handle it correctly
+		if common.IsRequestBodyTooLargeError(err) || errors.Is(err, common.ErrRequestBodyTooLarge) {
+			newAPIError = types.NewErrorWithStatusCode(err, types.ErrorCodeReadRequestBodyFailed, http.StatusRequestEntityTooLarge, types.ErrOptionWithSkipRetry())
+		} else {
+			newAPIError = types.NewError(err, types.ErrorCodeInvalidRequest)
+		}
 		return
 	}
 
@@ -114,9 +120,17 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 		return
 	}
 
-	meta := request.GetTokenCountMeta()
+	needSensitiveCheck := setting.ShouldCheckPromptSensitive()
+	needCountToken := constant.CountToken
+	// Avoid building huge CombineText (strings.Join) when token counting and sensitive check are both disabled.
+	var meta *types.TokenCountMeta
+	if needSensitiveCheck || needCountToken {
+		meta = request.GetTokenCountMeta()
+	} else {
+		meta = fastTokenCountMetaForPricing(request)
+	}
 
-	if setting.ShouldCheckPromptSensitive() {
+	if needSensitiveCheck && meta != nil {
 		contains, words := service.CheckSensitiveText(meta.CombineText)
 		if contains {
 			logger.LogWarn(c, fmt.Sprintf("user sensitive words detected: %s", strings.Join(words, ", ")))
@@ -157,16 +171,32 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 		}
 	}()
 
-	for i := 0; i <= common.RetryTimes; i++ {
-		channel, err := getChannel(c, group, originalModel, i)
-		if err != nil {
-			logger.LogError(c, err.Error())
-			newAPIError = err
+	retryParam := &service.RetryParam{
+		Ctx:        c,
+		TokenGroup: relayInfo.TokenGroup,
+		ModelName:  relayInfo.OriginModelName,
+		Retry:      common.GetPointer(0),
+	}
+
+	for ; retryParam.GetRetry() <= common.RetryTimes; retryParam.IncreaseRetry() {
+		channel, channelErr := getChannel(c, relayInfo, retryParam)
+		if channelErr != nil {
+			logger.LogError(c, channelErr.Error())
+			newAPIError = channelErr
 			break
 		}
 
 		addUsedChannel(c, channel.Id)
-		requestBody, _ := common.GetRequestBody(c)
+		requestBody, bodyErr := common.GetRequestBody(c)
+		if bodyErr != nil {
+			// Ensure consistent 413 for oversized bodies even when error occurs later (e.g., retry path)
+			if common.IsRequestBodyTooLargeError(bodyErr) || errors.Is(bodyErr, common.ErrRequestBodyTooLarge) {
+				newAPIError = types.NewErrorWithStatusCode(bodyErr, types.ErrorCodeReadRequestBodyFailed, http.StatusRequestEntityTooLarge, types.ErrOptionWithSkipRetry())
+			} else {
+				newAPIError = types.NewErrorWithStatusCode(bodyErr, types.ErrorCodeReadRequestBodyFailed, http.StatusBadRequest, types.ErrOptionWithSkipRetry())
+			}
+			break
+		}
 		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 
 		switch relayFormat {
@@ -186,7 +216,7 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 		processChannelError(c, *types.NewChannelError(channel.Id, channel.Type, channel.Name, channel.ChannelInfo.IsMultiKey, common.GetContextKeyString(c, constant.ContextKeyChannelKey), channel.GetAutoBan()), newAPIError)
 
-		if !shouldRetry(c, newAPIError, common.RetryTimes-i) {
+		if !shouldRetry(c, newAPIError, common.RetryTimes-retryParam.GetRetry()) {
 			break
 		}
 	}
@@ -211,8 +241,35 @@ func addUsedChannel(c *gin.Context, channelId int) {
 	c.Set("use_channel", useChannel)
 }
 
-func getChannel(c *gin.Context, group, originalModel string, retryCount int) (*model.Channel, *types.NewAPIError) {
-	if retryCount == 0 {
+func fastTokenCountMetaForPricing(request dto.Request) *types.TokenCountMeta {
+	if request == nil {
+		return &types.TokenCountMeta{}
+	}
+	meta := &types.TokenCountMeta{
+		TokenType: types.TokenTypeTokenizer,
+	}
+	switch r := request.(type) {
+	case *dto.GeneralOpenAIRequest:
+		if r.MaxCompletionTokens > r.MaxTokens {
+			meta.MaxTokens = int(r.MaxCompletionTokens)
+		} else {
+			meta.MaxTokens = int(r.MaxTokens)
+		}
+	case *dto.OpenAIResponsesRequest:
+		meta.MaxTokens = int(r.MaxOutputTokens)
+	case *dto.ClaudeRequest:
+		meta.MaxTokens = int(r.MaxTokens)
+	case *dto.ImageRequest:
+		// Pricing for image requests depends on ImagePriceRatio; safe to compute even when CountToken is disabled.
+		return r.GetTokenCountMeta()
+	default:
+		// Best-effort: leave CombineText empty to avoid large allocations.
+	}
+	return meta
+}
+
+func getChannel(c *gin.Context, info *relaycommon.RelayInfo, retryParam *service.RetryParam) (*model.Channel, *types.NewAPIError) {
+	if info.ChannelMeta == nil {
 		autoBan := c.GetBool("auto_ban")
 		autoBanInt := 1
 		if !autoBan {
@@ -225,14 +282,18 @@ func getChannel(c *gin.Context, group, originalModel string, retryCount int) (*m
 			AutoBan: &autoBanInt,
 		}, nil
 	}
-	channel, selectGroup, err := service.CacheGetRandomSatisfiedChannel(c, group, originalModel, retryCount)
+	channel, selectGroup, err := service.CacheGetRandomSatisfiedChannel(retryParam)
+
+	info.PriceData.GroupRatioInfo = helper.HandleGroupRatio(c, info)
+
 	if err != nil {
-		return nil, types.NewError(fmt.Errorf("获取分组 %s 下模型 %s 的可用渠道失败（retry）: %s", selectGroup, originalModel, err.Error()), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
+		return nil, types.NewError(fmt.Errorf("获取分组 %s 下模型 %s 的可用渠道失败（retry）: %s", selectGroup, info.OriginModelName, err.Error()), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
 	}
 	if channel == nil {
-		return nil, types.NewError(fmt.Errorf("分组 %s 下模型 %s 的可用渠道不存在（retry）", selectGroup, originalModel), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
+		return nil, types.NewError(fmt.Errorf("分组 %s 下模型 %s 的可用渠道不存在（retry）", selectGroup, info.OriginModelName), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
 	}
-	newAPIError := middleware.SetupContextForSelectedChannel(c, channel, originalModel)
+
+	newAPIError := middleware.SetupContextForSelectedChannel(c, channel, info.OriginModelName)
 	if newAPIError != nil {
 		return nil, newAPIError
 	}
@@ -366,7 +427,7 @@ func RelayMidjourney(c *gin.Context) {
 }
 
 func RelayNotImplemented(c *gin.Context) {
-	err := dto.OpenAIError{
+	err := types.OpenAIError{
 		Message: "API not implemented",
 		Type:    "new_api_error",
 		Param:   "",
@@ -378,7 +439,7 @@ func RelayNotImplemented(c *gin.Context) {
 }
 
 func RelayNotFound(c *gin.Context) {
-	err := dto.OpenAIError{
+	err := types.OpenAIError{
 		Message: fmt.Sprintf("Invalid URL (%s %s)", c.Request.Method, c.Request.URL.Path),
 		Type:    "invalid_request_error",
 		Param:   "",
@@ -392,8 +453,6 @@ func RelayNotFound(c *gin.Context) {
 func RelayTask(c *gin.Context) {
 	retryTimes := common.RetryTimes
 	channelId := c.GetInt("channel_id")
-	group := c.GetString("group")
-	originalModel := c.GetString("original_model")
 	c.Set("use_channel", []string{fmt.Sprintf("%d", channelId)})
 	relayInfo, err := relaycommon.GenRelayInfo(c, types.RelayFormatTask, nil, nil)
 	if err != nil {
@@ -403,8 +462,14 @@ func RelayTask(c *gin.Context) {
 	if taskErr == nil {
 		retryTimes = 0
 	}
-	for i := 0; shouldRetryTaskRelay(c, channelId, taskErr, retryTimes) && i < retryTimes; i++ {
-		channel, newAPIError := getChannel(c, group, originalModel, i)
+	retryParam := &service.RetryParam{
+		Ctx:        c,
+		TokenGroup: relayInfo.TokenGroup,
+		ModelName:  relayInfo.OriginModelName,
+		Retry:      common.GetPointer(0),
+	}
+	for ; shouldRetryTaskRelay(c, channelId, taskErr, retryTimes) && retryParam.GetRetry() < retryTimes; retryParam.IncreaseRetry() {
+		channel, newAPIError := getChannel(c, relayInfo, retryParam)
 		if newAPIError != nil {
 			logger.LogError(c, fmt.Sprintf("CacheGetRandomSatisfiedChannel failed: %s", newAPIError.Error()))
 			taskErr = service.TaskErrorWrapperLocal(newAPIError.Err, "get_channel_failed", http.StatusInternalServerError)
@@ -414,10 +479,18 @@ func RelayTask(c *gin.Context) {
 		useChannel := c.GetStringSlice("use_channel")
 		useChannel = append(useChannel, fmt.Sprintf("%d", channelId))
 		c.Set("use_channel", useChannel)
-		logger.LogInfo(c, fmt.Sprintf("using channel #%d to retry (remain times %d)", channel.Id, i))
+		logger.LogInfo(c, fmt.Sprintf("using channel #%d to retry (remain times %d)", channel.Id, retryParam.GetRetry()))
 		//middleware.SetupContextForSelectedChannel(c, channel, originalModel)
 
-		requestBody, _ := common.GetRequestBody(c)
+		requestBody, err := common.GetRequestBody(c)
+		if err != nil {
+			if common.IsRequestBodyTooLargeError(err) || errors.Is(err, common.ErrRequestBodyTooLarge) {
+				taskErr = service.TaskErrorWrapperLocal(err, "read_request_body_failed", http.StatusRequestEntityTooLarge)
+			} else {
+				taskErr = service.TaskErrorWrapperLocal(err, "read_request_body_failed", http.StatusBadRequest)
+			}
+			break
+		}
 		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 		taskErr = taskRelayHandler(c, relayInfo)
 	}

controller/task.go

@@ -88,7 +88,7 @@ func UpdateSunoTaskAll(ctx context.Context, taskChannelM map[int][]string, taskM
 	for channelId, taskIds := range taskChannelM {
 		err := updateSunoTaskAll(ctx, channelId, taskIds, taskM)
 		if err != nil {
-			logger.LogError(ctx, fmt.Sprintf("渠道 #%d 更新异步任务失败: %d", channelId, err.Error()))
+			logger.LogError(ctx, fmt.Sprintf("渠道 #%d 更新异步任务失败: %s", channelId, err.Error()))
 		}
 	}
 	return nil
@@ -141,7 +141,7 @@ func updateSunoTaskAll(ctx context.Context, channelId int, taskIds []string, tas
 		return err
 	}
 	if !responseItems.IsSuccess() {
-		common.SysLog(fmt.Sprintf("渠道 #%d 未完成的任务有: %d, 成功获取到任务数: %d", channelId, len(taskIds), string(responseBody)))
+		common.SysLog(fmt.Sprintf("渠道 #%d 未完成的任务有: %d, 成功获取到任务数: %s", channelId, len(taskIds), string(responseBody)))
 		return err
 	}
 

controller/token.go

@@ -171,6 +171,7 @@ func AddToken(c *gin.Context) {
 		ModelLimits:        token.ModelLimits,
 		AllowIps:           token.AllowIps,
 		Group:              token.Group,
+		CrossGroupRetry:    token.CrossGroupRetry,
 	}
 	err = cleanToken.Insert()
 	if err != nil {
@@ -248,6 +249,7 @@ func UpdateToken(c *gin.Context) {
 		cleanToken.ModelLimits = token.ModelLimits
 		cleanToken.AllowIps = token.AllowIps
 		cleanToken.Group = token.Group
+		cleanToken.CrossGroupRetry = token.CrossGroupRetry
 	}
 	err = cleanToken.Update()
 	if err != nil {

controller/topup_creem.go

@@ -7,12 +7,12 @@ import (
 	"encoding/hex"
 	"encoding/json"
 	"fmt"
-	"io"
-	"log"
-	"net/http"
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/setting"
+	"io"
+	"log"
+	"net/http"
 	"time"
 
 	"github.com/gin-gonic/gin"

controller/user.go

@@ -110,18 +110,17 @@ func setupLogin(user *model.User, c *gin.Context) {
 		})
 		return
 	}
-	cleanUser := model.User{
-		Id:          user.Id,
-		Username:    user.Username,
-		DisplayName: user.DisplayName,
-		Role:        user.Role,
-		Status:      user.Status,
-		Group:       user.Group,
-	}
 	c.JSON(http.StatusOK, gin.H{
 		"message": "",
 		"success": true,
-		"data":    cleanUser,
+		"data": map[string]any{
+			"id":           user.Id,
+			"username":     user.Username,
+			"display_name": user.DisplayName,
+			"role":         user.Role,
+			"status":       user.Status,
+			"group":        user.Group,
+		},
 	})
 }
 
@@ -288,11 +287,37 @@ func GetAllUsers(c *gin.Context) {
 	return
 }
 
+// SearchUsers handles a request to find users by keyword, group, or external identifier filters and returns paginated results.
+// It requires at least one search parameter (keyword, group, or any of the external ID/email filters) and responds with an error if none are provided.
+// On success it populates the page query with matching users and total count and returns that page info.
 func SearchUsers(c *gin.Context) {
 	keyword := c.Query("keyword")
 	group := c.Query("group")
+	filters := map[string]string{
+		"github_id":   c.Query("github_id"),
+		"discord_id":  c.Query("discord_id"),
+		"oidc_id":     c.Query("oidc_id"),
+		"wechat_id":   c.Query("wechat_id"),
+		"email":       c.Query("email"),
+		"telegram_id": c.Query("telegram_id"),
+		"linux_do_id": c.Query("linux_do_id"),
+	}
+
+	// 检查是否至少有一个搜索条件
+	hasFilter := keyword != "" || group != ""
+	for _, v := range filters {
+		if v != "" {
+			hasFilter = true
+			break
+		}
+	}
+	if !hasFilter {
+		common.ApiErrorMsg(c, "at least one search parameter is required")
+		return
+	}
+
 	pageInfo := common.GetPageQuery(c)
-	users, total, err := model.SearchUsers(keyword, group, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
+	users, total, err := model.SearchUsers(keyword, group, filters, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
 	if err != nil {
 		common.ApiError(c, err)
 		return
@@ -764,7 +789,10 @@ func checkUpdatePassword(originalPassword string, newPassword string, userId int
 	if err != nil {
 		return
 	}
-	if !common.ValidatePasswordAndHash(originalPassword, currentUser.Password) {
+
+	// 密码不为空,需要验证原密码
+	// 支持第一次账号绑定时原密码为空的情况
+	if !common.ValidatePasswordAndHash(originalPassword, currentUser.Password) && currentUser.Password != "" {
 		err = fmt.Errorf("原密码错误")
 		return
 	}
@@ -1291,4 +1319,4 @@ func UpdateUserSetting(c *gin.Context) {
 		"success": true,
 		"message": "设置已更新",
 	})
-}
+}

dto/audio.go

@@ -2,6 +2,7 @@ package dto
 
 import (
 	"encoding/json"
+	"strings"
 
 	"github.com/QuantumNous/new-api/types"
 
@@ -24,11 +25,14 @@ func (r *AudioRequest) GetTokenCountMeta() *types.TokenCountMeta {
 		CombineText: r.Input,
 		TokenType:   types.TokenTypeTextNumber,
 	}
+	if strings.Contains(r.Model, "gpt") {
+		meta.TokenType = types.TokenTypeTokenizer
+	}
 	return meta
 }
 
 func (r *AudioRequest) IsStream(c *gin.Context) bool {
-	return false
+	return r.StreamFormat == "sse"
 }
 
 func (r *AudioRequest) SetModelName(modelName string) {

dto/error.go

@@ -1,26 +1,32 @@
 package dto
 
-import "github.com/QuantumNous/new-api/types"
+import (
+	"encoding/json"
 
-type OpenAIError struct {
-	Message string `json:"message"`
-	Type    string `json:"type"`
-	Param   string `json:"param"`
-	Code    any    `json:"code"`
-}
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/types"
+)
+
+//type OpenAIError struct {
+//	Message string `json:"message"`
+//	Type    string `json:"type"`
+//	Param   string `json:"param"`
+//	Code    any    `json:"code"`
+//}
 
 type OpenAIErrorWithStatusCode struct {
-	Error      OpenAIError `json:"error"`
-	StatusCode int         `json:"status_code"`
+	Error      types.OpenAIError `json:"error"`
+	StatusCode int               `json:"status_code"`
 	LocalError bool
 }
 
 type GeneralErrorResponse struct {
-	Error    types.OpenAIError `json:"error"`
-	Message  string            `json:"message"`
-	Msg      string            `json:"msg"`
-	Err      string            `json:"err"`
-	ErrorMsg string            `json:"error_msg"`
+	Error    json.RawMessage `json:"error"`
+	Message  string          `json:"message"`
+	Msg      string          `json:"msg"`
+	Err      string          `json:"err"`
+	ErrorMsg string          `json:"error_msg"`
+	Metadata json.RawMessage   `json:"metadata,omitempty"`
 	Header   struct {
 		Message string `json:"message"`
 	} `json:"header"`
@@ -31,9 +37,35 @@ type GeneralErrorResponse struct {
 	} `json:"response"`
 }
 
+func (e GeneralErrorResponse) TryToOpenAIError() *types.OpenAIError {
+	var openAIError types.OpenAIError
+	if len(e.Error) > 0 {
+		err := common.Unmarshal(e.Error, &openAIError)
+		if err == nil && openAIError.Message != "" {
+			return &openAIError
+		}
+	}
+	return nil
+}
+
 func (e GeneralErrorResponse) ToMessage() string {
-	if e.Error.Message != "" {
-		return e.Error.Message
+	if len(e.Error) > 0 {
+		switch common.GetJsonType(e.Error) {
+		case "object":
+			var openAIError types.OpenAIError
+			err := common.Unmarshal(e.Error, &openAIError)
+			if err == nil && openAIError.Message != "" {
+				return openAIError.Message
+			}
+		case "string":
+			var msg string
+			err := common.Unmarshal(e.Error, &msg)
+			if err == nil && msg != "" {
+				return msg
+			}
+		default:
+			return string(e.Error)
+		}
 	}
 	if e.Message != "" {
 		return e.Message

dto/gemini.go

@@ -22,6 +22,27 @@ type GeminiChatRequest struct {
 	CachedContent      string                     `json:"cachedContent,omitempty"`
 }
 
+// UnmarshalJSON allows GeminiChatRequest to accept both snake_case and camelCase fields.
+func (r *GeminiChatRequest) UnmarshalJSON(data []byte) error {
+	type Alias GeminiChatRequest
+	var aux struct {
+		Alias
+		SystemInstructionSnake *GeminiChatContent `json:"system_instruction,omitempty"`
+	}
+
+	if err := common.Unmarshal(data, &aux); err != nil {
+		return err
+	}
+
+	*r = GeminiChatRequest(aux.Alias)
+
+	if aux.SystemInstructionSnake != nil {
+		r.SystemInstructions = aux.SystemInstructionSnake
+	}
+
+	return nil
+}
+
 type ToolConfig struct {
 	FunctionCallingConfig *FunctionCallingConfig `json:"functionCallingConfig,omitempty"`
 	RetrievalConfig       *RetrievalConfig       `json:"retrievalConfig,omitempty"`

go.mod

@@ -27,6 +27,7 @@ require (
 	github.com/golang-jwt/jwt/v5 v5.3.0
 	github.com/google/uuid v1.6.0
 	github.com/gorilla/websocket v1.5.0
+	github.com/grafana/pyroscope-go v1.2.7
 	github.com/jfreymuth/oggvorbis v1.0.5
 	github.com/jinzhu/copier v0.4.0
 	github.com/joho/godotenv v1.5.1
@@ -77,11 +78,11 @@ require (
 	github.com/go-sql-driver/mysql v1.7.0 // indirect
 	github.com/go-webauthn/x v0.1.25 // indirect
 	github.com/goccy/go-json v0.10.2 // indirect
-	github.com/google/go-cmp v0.6.0 // indirect
 	github.com/google/go-tpm v0.9.5 // indirect
 	github.com/gorilla/context v1.1.1 // indirect
 	github.com/gorilla/securecookie v1.1.1 // indirect
 	github.com/gorilla/sessions v1.2.1 // indirect
+	github.com/grafana/pyroscope-go/godeltaprof v0.1.9 // indirect
 	github.com/icza/bitio v1.1.0 // indirect
 	github.com/jackc/pgpassfile v1.0.0 // indirect
 	github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
@@ -91,6 +92,7 @@ require (
 	github.com/jinzhu/inflection v1.0.0 // indirect
 	github.com/jinzhu/now v1.1.5 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
+	github.com/klauspost/compress v1.17.8 // indirect
 	github.com/klauspost/cpuid/v2 v2.3.0 // indirect
 	github.com/leodido/go-urn v1.4.0 // indirect
 	github.com/mattn/go-isatty v0.0.20 // indirect

go.sum

@@ -118,9 +118,8 @@ github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeN
 github.com/google/go-tpm v0.9.5 h1:ocUmnDebX54dnW+MQWGQRbdaAcJELsa6PqZhJ48KwVU=
 github.com/google/go-tpm v0.9.5/go.mod h1:h9jEsEECg7gtLis0upRBQU+GhYVH6jMjrFxI8u6bVUY=
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
-github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26 h1:Xim43kblpZXfIBQsbuBVKCudVG457BR2GZFIz3uw3hQ=
-github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26/go.mod h1:dDKJzRmX4S37WGHujM7tX//fmj1uioxKzKxz3lo4HJo=
 github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e h1:ijClszYn+mADRFY17kjQEVQ1XRhq2/JR1M3sGqeJoxs=
+github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e/go.mod h1:boTsfXsheKC2y+lKOCMpSfarhxDeIzfZG1jqGcPl3cA=
 github.com/google/uuid v1.1.2/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
 github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
@@ -132,6 +131,10 @@ github.com/gorilla/sessions v1.2.1 h1:DHd3rPN5lE3Ts3D8rKkQ8x/0kqfeNmBAaiSi+o7Fsg
 github.com/gorilla/sessions v1.2.1/go.mod h1:dk2InVEVJ0sfLlnXv9EAgkf6ecYs/i80K/zI+bUmuGM=
 github.com/gorilla/websocket v1.5.0 h1:PPwGk2jz7EePpoHN/+ClbZu8SPxiqlu12wZP/3sWmnc=
 github.com/gorilla/websocket v1.5.0/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE=
+github.com/grafana/pyroscope-go v1.2.7 h1:VWBBlqxjyR0Cwk2W6UrE8CdcdD80GOFNutj0Kb1T8ac=
+github.com/grafana/pyroscope-go v1.2.7/go.mod h1:o/bpSLiJYYP6HQtvcoVKiE9s5RiNgjYTj1DhiddP2Pc=
+github.com/grafana/pyroscope-go/godeltaprof v0.1.9 h1:c1Us8i6eSmkW+Ez05d3co8kasnuOY813tbMN8i/a3Og=
+github.com/grafana/pyroscope-go/godeltaprof v0.1.9/go.mod h1:2+l7K7twW49Ct4wFluZD3tZ6e0SjanjcUUBPVD/UuGU=
 github.com/icza/bitio v1.1.0 h1:ysX4vtldjdi3Ygai5m1cWy4oLkhWTAi+SyO6HC8L9T0=
 github.com/icza/bitio v1.1.0/go.mod h1:0jGnlLAx8MKMr9VGnn/4YrvZiprkvBelsVIbA9Jjr9A=
 github.com/icza/mighty v0.0.0-20180919140131-cfd07d671de6 h1:8UsGZ2rr2ksmEru6lToqnXgA8Mz1DP11X4zSJ159C3k=
@@ -160,12 +163,15 @@ github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwA
 github.com/json-iterator/go v1.1.9/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
 github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM=
 github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo=
+github.com/klauspost/compress v1.17.8 h1:YcnTYrq7MikUT7k0Yb5eceMmALQPYBW/Xltxn0NAMnU=
+github.com/klauspost/compress v1.17.8/go.mod h1:Di0epgTjJY877eYKx5yC51cX2A2Vl2ibi7bDH9ttBbw=
 github.com/klauspost/cpuid/v2 v2.3.0 h1:S4CRMLnYUhGeDFDqkGriYKdfoFlDnMtqTiI/sFzhA9Y=
 github.com/klauspost/cpuid/v2 v2.3.0/go.mod h1:hqwkgyIinND0mEev00jJYCxPNVRVXFQeu1XKlok6oO0=
 github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
 github.com/kr/pretty v0.2.1/go.mod h1:ipq/a2n7PKx3OHsz4KJII5eveXtPO4qwEXGdVfWzfnI=
-github.com/kr/pretty v0.3.0 h1:WgNl7dwNpEZ6jJ9k1snq4pZsg7DOEN8hP9Xw0Tsjwk0=
 github.com/kr/pretty v0.3.0/go.mod h1:640gp4NfQd8pI5XOwp5fnNeVWj67G7CFk/SaSQn7NBk=
+github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
+github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
 github.com/kr/pty v1.1.1/go.mod h1:pFQYn66WHrOpPYNljwOMqo10TkYh1fy3cYio2l3bCsQ=
 github.com/kr/pty v1.1.8/go.mod h1:O1sed60cT9XZ5uDucP5qwvh+TE3NnUj51EiZO/lmSfw=
 github.com/kr/text v0.1.0/go.mod h1:4Jbv+DJW3UT/LiOwJeYQe1efqtUx/iVham/4vfdArNI=
@@ -214,14 +220,11 @@ github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZb
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/pquerna/otp v1.5.0 h1:NMMR+WrmaqXU4EzdGJEE1aUUI0AMRzsp96fFFWNPwxs=
 github.com/pquerna/otp v1.5.0/go.mod h1:dkJfzwRKNiegxyNb54X/3fLwhCynbMspSyWKnvi1AEg=
-github.com/remyoudompheng/bigfft v0.0.0-20200410134404-eec4a21b6bb0/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
 github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec h1:W09IVJc94icq4NjY3clb7Lk8O1qJ8BdBEF8z0ibU0rE=
 github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
 github.com/rogpeppe/go-internal v1.6.1/go.mod h1:xXDCJY+GAPziupqXw64V24skbSoqbTEfhy4qGm1nDQc=
 github.com/rogpeppe/go-internal v1.8.0 h1:FCbCCtXNOY3UtUuHUYaghJg4y7Fd14rXifAYUAtL9R8=
 github.com/rogpeppe/go-internal v1.8.0/go.mod h1:WmiCO8CzOY8rg0OYDC4/i/2WRWAB6poM+XZ2dLUbcbE=
-github.com/samber/lo v1.39.0 h1:4gTz1wUhNYLhFSKl6O+8peW0v2F4BCY034GRpU9WnuA=
-github.com/samber/lo v1.39.0/go.mod h1:+m/ZKRl6ClXCE2Lgf3MsQlWfh4bn1bz6CXEOxnEXnEA=
 github.com/samber/lo v1.52.0 h1:Rvi+3BFHES3A8meP33VPAxiBZX/Aws5RxrschYGjomw=
 github.com/samber/lo v1.52.0/go.mod h1:4+MXEGsJzbKGaUEQFKBq2xtfuznW9oz/WrgyzMzRoM0=
 github.com/shirou/gopsutil v3.21.11+incompatible h1:+1+c1VGhc88SSonWP6foOcLhvnKlUeu/erjjvaPEYiI=
@@ -231,6 +234,7 @@ github.com/shopspring/decimal v1.4.0/go.mod h1:gawqmDU56v4yIKSwfBSFip1HdCCXN8/+D
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
+github.com/stretchr/objx v0.5.2 h1:xuMeJ0Sdp5ZMRXx/aWO6RZxdr3beISkG5/G/aIRr3pY=
 github.com/stretchr/objx v0.5.2/go.mod h1:FRsXN1f5AsAjCGJKqEizvkpNtU+EGNCLh3NxZ/8L+MA=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
@@ -288,12 +292,12 @@ golang.org/x/arch v0.21.0/go.mod h1:dNHoOeKiyja7GTvF9NJS1l3Z2yntpQNzgrjh1cU103A=
 golang.org/x/crypto v0.0.0-20210711020723-a769d52b0f97/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
 golang.org/x/crypto v0.45.0 h1:jMBrvKuj23MTlT0bQEOBcAE0mjg8mK9RXFhRH6nyF3Q=
 golang.org/x/crypto v0.45.0/go.mod h1:XTGrrkGJve7CYK7J8PEww4aY7gM3qMCElcJQ8n8JdX4=
-golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 h1:985EYyeCOxTpcgOTJpflJUwOeEz0CQOdPt73OzpE9F8=
-golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0/go.mod h1:/lliqkxwWAhPjf5oSOIJup2XcqJaw8RGS6k3TGEc7GI=
 golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b h1:M2rDM6z3Fhozi9O7NWsxAkg/yqS/lQJ6PmkyIV3YP+o=
 golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b/go.mod h1:3//PLf8L/X+8b4vuAfHzxeRUl04Adcb341+IGKfnqS8=
 golang.org/x/image v0.23.0 h1:HseQ7c2OpPKTPVzNjG5fwJsOTCiiwS4QdsYi5XU6H68=
 golang.org/x/image v0.23.0/go.mod h1:wJJBTdLfCCf3tiHa1fNxpZmUI4mmoZvwMCPP0ddoNKY=
+golang.org/x/mod v0.29.0 h1:HV8lRxZC4l2cr3Zq1LvtOsi/ThTgWnUk/y64QSs8GwA=
+golang.org/x/mod v0.29.0/go.mod h1:NyhrlYXJ2H4eJiRy/WDBO6HMqZQ6q9nk4JzS3NuCK+w=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
 golang.org/x/net v0.0.0-20210520170846-37e1c6afe023/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.47.0 h1:Mx+4dIFzqraBXUugkia1OOvlD6LemFo1ALMHjrXDOhY=
@@ -321,6 +325,8 @@ golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.31.0 h1:aC8ghyu4JhP8VojJ2lEHBnochRno1sgL6nEi9WGFGMM=
 golang.org/x/text v0.31.0/go.mod h1:tKRAlv61yKIjGGHX/4tP1LTbc13YSec1pxVEWXzfoeM=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.38.0 h1:Hx2Xv8hISq8Lm16jvBZ2VQf+RLmbd7wVUsALibYI/IQ=
+golang.org/x/tools v0.38.0/go.mod h1:yEsQ/d/YK8cjh0L6rZlY8tgtlKiBNTL14pGDJPJpYQs=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
 google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
@@ -350,19 +356,29 @@ gorm.io/driver/postgres v1.5.2/go.mod h1:fmpX0m2I1PKuR7mKZiEluwrP3hbs+ps7JIGMUBp
 gorm.io/gorm v1.23.8/go.mod h1:l2lP/RyAtc1ynaTjFksBde/O8v9oOGIApu2/xRitmZk=
 gorm.io/gorm v1.25.2 h1:gs1o6Vsa+oVKG/a9ElL3XgyGfghFfkKA2SInQaCyMho=
 gorm.io/gorm v1.25.2/go.mod h1:L4uxeKpfBml98NYqVqwAdmV1a2nBtAec/cf3fpucW/k=
-modernc.org/libc v1.22.5 h1:91BNch/e5B0uPbJFgqbxXuOnxBQjlS//icfQEGmvyjE=
-modernc.org/libc v1.22.5/go.mod h1:jj+Z7dTNX8fBScMVNRAYZ/jF91K8fdT2hYMThc3YjBY=
+modernc.org/cc/v4 v4.26.5 h1:xM3bX7Mve6G8K8b+T11ReenJOT+BmVqQj0FY5T4+5Y4=
+modernc.org/cc/v4 v4.26.5/go.mod h1:uVtb5OGqUKpoLWhqwNQo/8LwvoiEBLvZXIQ/SmO6mL0=
+modernc.org/ccgo/v4 v4.28.1 h1:wPKYn5EC/mYTqBO373jKjvX2n+3+aK7+sICCv4Fjy1A=
+modernc.org/ccgo/v4 v4.28.1/go.mod h1:uD+4RnfrVgE6ec9NGguUNdhqzNIeeomeXf6CL0GTE5Q=
+modernc.org/fileutil v1.3.40 h1:ZGMswMNc9JOCrcrakF1HrvmergNLAmxOPjizirpfqBA=
+modernc.org/fileutil v1.3.40/go.mod h1:HxmghZSZVAz/LXcMNwZPA/DRrQZEVP9VX0V4LQGQFOc=
+modernc.org/gc/v2 v2.6.5 h1:nyqdV8q46KvTpZlsw66kWqwXRHdjIlJOhG6kxiV/9xI=
+modernc.org/gc/v2 v2.6.5/go.mod h1:YgIahr1ypgfe7chRuJi2gD7DBQiKSLMPgBQe9oIiito=
+modernc.org/goabi0 v0.2.0 h1:HvEowk7LxcPd0eq6mVOAEMai46V+i7Jrj13t4AzuNks=
+modernc.org/goabi0 v0.2.0/go.mod h1:CEFRnnJhKvWT1c1JTI3Avm+tgOWbkOu5oPA8eH8LnMI=
 modernc.org/libc v1.66.10 h1:yZkb3YeLx4oynyR+iUsXsybsX4Ubx7MQlSYEw4yj59A=
 modernc.org/libc v1.66.10/go.mod h1:8vGSEwvoUoltr4dlywvHqjtAqHBaw0j1jI7iFBTAr2I=
-modernc.org/mathutil v1.5.0 h1:rV0Ko/6SfM+8G+yKiyI830l3Wuz1zRutdslNoQ0kfiQ=
-modernc.org/mathutil v1.5.0/go.mod h1:mZW8CKdRPY1v87qxC/wUdX5O1qDzXMP5TH3wjfpga6E=
 modernc.org/mathutil v1.7.1 h1:GCZVGXdaN8gTqB1Mf/usp1Y/hSqgI2vAGGP4jZMCxOU=
 modernc.org/mathutil v1.7.1/go.mod h1:4p5IwJITfppl0G4sUEDtCr4DthTaT47/N3aT6MhfgJg=
-modernc.org/memory v1.5.0 h1:N+/8c5rE6EqugZwHii4IFsaJ7MUhoWX07J5tC/iI5Ds=
-modernc.org/memory v1.5.0/go.mod h1:PkUhL0Mugw21sHPeskwZW4D6VscE/GQJOnIpCnW6pSU=
 modernc.org/memory v1.11.0 h1:o4QC8aMQzmcwCK3t3Ux/ZHmwFPzE6hf2Y5LbkRs+hbI=
 modernc.org/memory v1.11.0/go.mod h1:/JP4VbVC+K5sU2wZi9bHoq2MAkCnrt2r98UGeSK7Mjw=
-modernc.org/sqlite v1.23.1 h1:nrSBg4aRQQwq59JpvGEQ15tNxoO5pX/kUjcRNwSAGQM=
-modernc.org/sqlite v1.23.1/go.mod h1:OrDj17Mggn6MhE+iPbBNf7RGKODDE9NFT0f3EwDzJqk=
+modernc.org/opt v0.1.4 h1:2kNGMRiUjrp4LcaPuLY2PzUfqM/w9N23quVwhKt5Qm8=
+modernc.org/opt v0.1.4/go.mod h1:03fq9lsNfvkYSfxrfUhZCWPk1lm4cq4N+Bh//bEtgns=
+modernc.org/sortutil v1.2.1 h1:+xyoGf15mM3NMlPDnFqrteY07klSFxLElE2PVuWIJ7w=
+modernc.org/sortutil v1.2.1/go.mod h1:7ZI3a3REbai7gzCLcotuw9AC4VZVpYMjDzETGsSMqJE=
 modernc.org/sqlite v1.40.1 h1:VfuXcxcUWWKRBuP8+BR9L7VnmusMgBNNnBYGEe9w/iY=
 modernc.org/sqlite v1.40.1/go.mod h1:9fjQZ0mB1LLP0GYrp39oOJXx/I2sxEnZtzCmEQIKvGE=
+modernc.org/strutil v1.2.1 h1:UneZBkQA+DX2Rp35KcM69cSsNES9ly8mQWD71HKlOA0=
+modernc.org/strutil v1.2.1/go.mod h1:EHkiggD70koQxjVdSBM3JKM7k6L0FbGE5eymy9i3B9A=
+modernc.org/token v1.1.0 h1:Xl7Ap9dKaEs5kLoOQeQmPWevfnk/DM5qcLcYlA8ys6Y=
+modernc.org/token v1.1.0/go.mod h1:UGzOrNV1mAFSEB63lOFHIpNRUVMvYTc6yu1SMY/XTDM=

main.go

@@ -124,6 +124,11 @@ func main() {
 		common.SysLog("pprof enabled")
 	}
 
+	err = common.StartPyroScope()
+	if err != nil {
+		common.SysError(fmt.Sprintf("start pyroscope error : %v", err))
+	}
+
 	// Initialize HTTP server
 	server := gin.New()
 	server.Use(gin.CustomRecovery(func(c *gin.Context, err any) {

middleware/auth.go

@@ -2,12 +2,14 @@ package middleware
 
 import (
 	"fmt"
+	"net"
 	"net/http"
 	"strconv"
 	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
@@ -216,10 +218,14 @@ func TokenAuth() func(c *gin.Context) {
 		}
 		key := c.Request.Header.Get("Authorization")
 		parts := make([]string, 0)
-		key = strings.TrimPrefix(key, "Bearer ")
+		if strings.HasPrefix(key, "Bearer ") || strings.HasPrefix(key, "bearer ") {
+			key = strings.TrimSpace(key[7:])
+		}
 		if key == "" || key == "midjourney-proxy" {
 			key = c.Request.Header.Get("mj-api-secret")
-			key = strings.TrimPrefix(key, "Bearer ")
+			if strings.HasPrefix(key, "Bearer ") || strings.HasPrefix(key, "bearer ") {
+				key = strings.TrimSpace(key[7:])
+			}
 			key = strings.TrimPrefix(key, "sk-")
 			parts = strings.Split(key, "-")
 			key = parts[0]
@@ -240,13 +246,20 @@ func TokenAuth() func(c *gin.Context) {
 			return
 		}
 
-		allowIpsMap := token.GetIpLimitsMap()
-		if len(allowIpsMap) != 0 {
+		allowIps := token.GetIpLimits()
+		if len(allowIps) > 0 {
 			clientIp := c.ClientIP()
-			if _, ok := allowIpsMap[clientIp]; !ok {
+			logger.LogDebug(c, "Token has IP restrictions, checking client IP %s", clientIp)
+			ip := net.ParseIP(clientIp)
+			if ip == nil {
+				abortWithOpenAiMessage(c, http.StatusForbidden, "无法解析客户端 IP 地址")
+				return
+			}
+			if common.IsIpInCIDRList(ip, allowIps) == false {
 				abortWithOpenAiMessage(c, http.StatusForbidden, "您的 IP 不在令牌允许访问的列表中")
 				return
 			}
+			logger.LogDebug(c, "Client IP %s passed the token IP restrictions check", clientIp)
 		}
 
 		userCache, err := model.GetUserCache(token.UserId)
@@ -307,7 +320,8 @@ func SetupContextForToken(c *gin.Context, token *model.Token, parts ...string) e
 	} else {
 		c.Set("token_model_limit_enabled", false)
 	}
-	c.Set("token_group", token.Group)
+	common.SetContextKey(c, constant.ContextKeyTokenGroup, token.Group)
+	common.SetContextKey(c, constant.ContextKeyTokenCrossGroupRetry, token.CrossGroupRetry)
 	if len(parts) > 1 {
 		if model.IsAdmin(token.UserId) {
 			c.Set("specific_channel_id", parts[1])

middleware/distributor.go

@@ -97,7 +97,12 @@ func Distribute() func(c *gin.Context) {
 						common.SetContextKey(c, constant.ContextKeyUsingGroup, usingGroup)
 					}
 				}
-				channel, selectGroup, err = service.CacheGetRandomSatisfiedChannel(c, usingGroup, modelRequest.Model, 0)
+				channel, selectGroup, err = service.CacheGetRandomSatisfiedChannel(&service.RetryParam{
+					Ctx:        c,
+					ModelName:  modelRequest.Model,
+					TokenGroup: usingGroup,
+					Retry:      common.GetPointer(0),
+				})
 				if err != nil {
 					showGroup := usingGroup
 					if usingGroup == "auto" {
@@ -157,7 +162,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 			}
 			midjourneyModel, mjErr, success := service.GetMjRequestModel(relayMode, &midjourneyRequest)
 			if mjErr != nil {
-				return nil, false, fmt.Errorf(mjErr.Description)
+				return nil, false, fmt.Errorf("%s", mjErr.Description)
 			}
 			if midjourneyModel == "" {
 				if !success {

middleware/gzip.go

@@ -5,32 +5,69 @@ import (
 	"io"
 	"net/http"
 
+	"github.com/QuantumNous/new-api/constant"
 	"github.com/andybalholm/brotli"
 	"github.com/gin-gonic/gin"
 )
 
+type readCloser struct {
+	io.Reader
+	closeFn func() error
+}
+
+func (rc *readCloser) Close() error {
+	if rc.closeFn != nil {
+		return rc.closeFn()
+	}
+	return nil
+}
+
 func DecompressRequestMiddleware() gin.HandlerFunc {
 	return func(c *gin.Context) {
 		if c.Request.Body == nil || c.Request.Method == http.MethodGet {
 			c.Next()
 			return
 		}
+		maxMB := constant.MaxRequestBodyMB
+		if maxMB <= 0 {
+			maxMB = 32
+		}
+		maxBytes := int64(maxMB) << 20
+
+		origBody := c.Request.Body
+		wrapMaxBytes := func(body io.ReadCloser) io.ReadCloser {
+			return http.MaxBytesReader(c.Writer, body, maxBytes)
+		}
+
 		switch c.GetHeader("Content-Encoding") {
 		case "gzip":
-			gzipReader, err := gzip.NewReader(c.Request.Body)
+			gzipReader, err := gzip.NewReader(origBody)
 			if err != nil {
+				_ = origBody.Close()
 				c.AbortWithStatus(http.StatusBadRequest)
 				return
 			}
-			defer gzipReader.Close()
-
-			// Replace the request body with the decompressed data
-			c.Request.Body = io.NopCloser(gzipReader)
+			// Replace the request body with the decompressed data, and enforce a max size (post-decompression).
+			c.Request.Body = wrapMaxBytes(&readCloser{
+				Reader: gzipReader,
+				closeFn: func() error {
+					_ = gzipReader.Close()
+					return origBody.Close()
+				},
+			})
 			c.Request.Header.Del("Content-Encoding")
 		case "br":
-			reader := brotli.NewReader(c.Request.Body)
-			c.Request.Body = io.NopCloser(reader)
+			reader := brotli.NewReader(origBody)
+			c.Request.Body = wrapMaxBytes(&readCloser{
+				Reader: reader,
+				closeFn: func() error {
+					return origBody.Close()
+				},
+			})
 			c.Request.Header.Del("Content-Encoding")
+		default:
+			// Even for uncompressed bodies, enforce a max size to avoid huge request allocations.
+			c.Request.Body = wrapMaxBytes(origBody)
 		}
 
 		// Continue processing the request

model/channel.go

@@ -254,6 +254,9 @@ func (channel *Channel) Save() error {
 }
 
 func (channel *Channel) SaveWithoutKey() error {
+	if channel.Id == 0 {
+		return errors.New("channel ID is 0")
+	}
 	return DB.Omit("key").Save(channel).Error
 }
 

model/token.go

@@ -6,7 +6,6 @@ import (
 	"strings"
 
 	"github.com/QuantumNous/new-api/common"
-
 	"github.com/bytedance/gopkg/util/gopool"
 	"gorm.io/gorm"
 )
@@ -27,6 +26,7 @@ type Token struct {
 	AllowIps           *string        `json:"allow_ips" gorm:"default:''"`
 	UsedQuota          int            `json:"used_quota" gorm:"default:0"` // used quota
 	Group              string         `json:"group" gorm:"default:''"`
+	CrossGroupRetry    bool           `json:"cross_group_retry" gorm:"default:false"` // 跨分组重试，仅auto分组有效
 	DeletedAt          gorm.DeletedAt `gorm:"index"`
 }
 
@@ -34,26 +34,26 @@ func (token *Token) Clean() {
 	token.Key = ""
 }
 
-func (token *Token) GetIpLimitsMap() map[string]any {
+func (token *Token) GetIpLimits() []string {
 	// delete empty spaces
 	//split with \n
-	ipLimitsMap := make(map[string]any)
+	ipLimits := make([]string, 0)
 	if token.AllowIps == nil {
-		return ipLimitsMap
+		return ipLimits
 	}
 	cleanIps := strings.ReplaceAll(*token.AllowIps, " ", "")
 	if cleanIps == "" {
-		return ipLimitsMap
+		return ipLimits
 	}
 	ips := strings.Split(cleanIps, "\n")
 	for _, ip := range ips {
 		ip = strings.TrimSpace(ip)
 		ip = strings.ReplaceAll(ip, ",", "")
-		if common.IsIP(ip) {
-			ipLimitsMap[ip] = true
+		if ip != "" {
+			ipLimits = append(ipLimits, ip)
 		}
 	}
-	return ipLimitsMap
+	return ipLimits
 }
 
 func GetAllUserTokens(userId int, startIdx int, num int) ([]*Token, error) {
@@ -112,7 +112,12 @@ func ValidateUserToken(key string) (token *Token, err error) {
 		}
 		return token, nil
 	}
-	return nil, errors.New("无效的令牌")
+	common.SysLog("ValidateUserToken: failed to get token: " + err.Error())
+	if errors.Is(err, gorm.ErrRecordNotFound) {
+		return nil, errors.New("无效的令牌")
+	} else {
+		return nil, errors.New("无效的令牌，数据库查询出错，请联系管理员")
+	}
 }
 
 func GetTokenByIds(id int, userId int) (*Token, error) {
@@ -185,7 +190,7 @@ func (token *Token) Update() (err error) {
 		}
 	}()
 	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota",
-		"model_limits_enabled", "model_limits", "allow_ips", "group").Updates(token).Error
+		"model_limits_enabled", "model_limits", "allow_ips", "group", "cross_group_retry").Updates(token).Error
 	return err
 }
 

model/user.go

@@ -185,6 +185,11 @@ func GetMaxUserId() int {
 	return user.Id
 }
 
+// GetAllUsers retrieves a paginated list of users and the total number of users.
+// It returns results that include soft-deleted records, ordered by id descending,
+// and omits the password field from the returned user objects.
+// The pageInfo parameter provides the page size and start index for pagination.
+// Returns the slice of users, the total user count, and any error encountered.
 func GetAllUsers(pageInfo *common.PageInfo) (users []*User, total int64, err error) {
 	// Start transaction
 	tx := DB.Begin()
@@ -219,7 +224,10 @@ func GetAllUsers(pageInfo *common.PageInfo) (users []*User, total int64, err err
 	return users, total, nil
 }
 
-func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User, int64, error) {
+// SearchUsers searches for users matching the provided keyword, group, and exact-match filters.
+// The function accepts a keyword (performs LIKE match on username, email, and display_name; if the keyword is numeric it also matches id), a group to scope results, a map of exact-match filters (allowed keys: "github_id", "discord_id", "oidc_id", "wechat_id", "email", "telegram_id", "linux_do_id"), and pagination parameters startIdx and num.
+// It returns the matched users, the total number of records matching the query (ignoring pagination), and an error if the operation fails.
+func SearchUsers(keyword string, group string, filters map[string]string, startIdx int, num int) ([]*User, int64, error) {
 	var users []*User
 	var total int64
 	var err error
@@ -238,32 +246,46 @@ func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User,
 	// 构建基础查询
 	query := tx.Unscoped().Model(&User{})
 
-	// 构建搜索条件
-	likeCondition := "username LIKE ? OR email LIKE ? OR display_name LIKE ?"
+	// 允许的过滤字段白名单
+	allowedFields := map[string]bool{
+		"github_id":   true,
+		"discord_id":  true,
+		"oidc_id":     true,
+		"wechat_id":   true,
+		"email":       true,
+		"telegram_id": true,
+		"linux_do_id": true,
+	}
 
-	// 尝试将关键字转换为整数ID
-	keywordInt, err := strconv.Atoi(keyword)
-	if err == nil {
-		// 如果是数字，同时搜索ID和其他字段
-		likeCondition = "id = ? OR " + likeCondition
-		if group != "" {
-			query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
-				keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
-		} else {
+	// 应用精确匹配过滤器
+	for field, value := range filters {
+		if value != "" && allowedFields[field] {
+			query = query.Where(field+" = ?", value)
+		}
+	}
+
+	// 构建搜索条件
+	if keyword != "" {
+		likeCondition := "username LIKE ? OR email LIKE ? OR display_name LIKE ?"
+
+		// 尝试将关键字转换为整数ID
+		keywordInt, err := strconv.Atoi(keyword)
+		if err == nil {
+			// 如果是数字，同时搜索ID和其他字段
+			likeCondition = "id = ? OR " + likeCondition
 			query = query.Where(likeCondition,
 				keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
-		}
-	} else {
-		// 非数字关键字，只搜索字符串字段
-		if group != "" {
-			query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
-				"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
 		} else {
+			// 非数字关键字，只搜索字符串字段
 			query = query.Where(likeCondition,
 				"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
 		}
 	}
 
+	if group != "" {
+		query = query.Where(commonGroupCol+" = ?", group)
+	}
+
 	// 获取总数
 	err = query.Count(&total).Error
 	if err != nil {
@@ -928,4 +950,4 @@ func RootUserExists() bool {
 		return false
 	}
 	return true
-}
+}

relay/audio_handler.go

@@ -67,8 +67,11 @@ func AudioHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 		service.ResetStatusCode(newAPIError, statusCodeMappingStr)
 		return newAPIError
 	}
-
-	postConsumeQuota(c, info, usage.(*dto.Usage), "")
+	if usage.(*dto.Usage).CompletionTokenDetails.AudioTokens > 0 || usage.(*dto.Usage).PromptTokensDetails.AudioTokens > 0 {
+		service.PostAudioConsumeQuota(c, info, usage.(*dto.Usage), "")
+	} else {
+		postConsumeQuota(c, info, usage.(*dto.Usage), "")
+	}
 
 	return nil
 }

relay/channel/aws/constants.go

@@ -18,7 +18,7 @@ var awsModelIDMap = map[string]string{
 	"claude-opus-4-1-20250805":   "anthropic.claude-opus-4-1-20250805-v1:0",
 	"claude-sonnet-4-5-20250929": "anthropic.claude-sonnet-4-5-20250929-v1:0",
 	"claude-haiku-4-5-20251001":  "anthropic.claude-haiku-4-5-20251001-v1:0",
-	"claude-opus-4-5-20251101":  "anthropic.claude-opus-4-5-20251101-v1:0",
+	"claude-opus-4-5-20251101":   "anthropic.claude-opus-4-5-20251101-v1:0",
 	// Nova models
 	"nova-micro-v1:0":   "amazon.nova-micro-v1:0",
 	"nova-lite-v1:0":    "amazon.nova-lite-v1:0",

relay/channel/aws/relay-aws.go

@@ -18,6 +18,7 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/pkg/errors"
 
+	"github.com/QuantumNous/new-api/setting/model_setting"
 	"github.com/aws/aws-sdk-go-v2/aws"
 	"github.com/aws/aws-sdk-go-v2/credentials"
 	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
@@ -129,7 +130,7 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 				Accept:      aws.String("application/json"),
 				ContentType: aws.String("application/json"),
 			}
-			awsReq.Body, err = common.Marshal(awsClaudeReq)
+			awsReq.Body, err = buildAwsRequestBody(c, info, awsClaudeReq)
 			if err != nil {
 				return nil, types.NewError(errors.Wrap(err, "marshal aws request fail"), types.ErrorCodeBadRequestBody)
 			}
@@ -141,7 +142,7 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 				Accept:      aws.String("application/json"),
 				ContentType: aws.String("application/json"),
 			}
-			awsReq.Body, err = common.Marshal(awsClaudeReq)
+			awsReq.Body, err = buildAwsRequestBody(c, info, awsClaudeReq)
 			if err != nil {
 				return nil, types.NewError(errors.Wrap(err, "marshal aws request fail"), types.ErrorCodeBadRequestBody)
 			}
@@ -151,6 +152,24 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 	}
 }
 
+// buildAwsRequestBody prepares the payload for AWS requests, applying passthrough rules when enabled.
+func buildAwsRequestBody(c *gin.Context, info *relaycommon.RelayInfo, awsClaudeReq any) ([]byte, error) {
+	if model_setting.GetGlobalSettings().PassThroughRequestEnabled || info.ChannelSetting.PassThroughBodyEnabled {
+		body, err := common.GetRequestBody(c)
+		if err != nil {
+			return nil, errors.Wrap(err, "get request body for pass-through fail")
+		}
+		var data map[string]interface{}
+		if err := common.Unmarshal(body, &data); err != nil {
+			return nil, errors.Wrap(err, "pass-through unmarshal request body fail")
+		}
+		delete(data, "model")
+		delete(data, "stream")
+		return common.Marshal(data)
+	}
+	return common.Marshal(awsClaudeReq)
+}
+
 func getAwsRegionPrefix(awsRegionId string) string {
 	parts := strings.Split(awsRegionId, "-")
 	regionPrefix := ""

relay/channel/baidu/relay-baidu.go

@@ -150,7 +150,7 @@ func baiduHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respon
 		return types.NewError(err, types.ErrorCodeBadResponseBody), nil
 	}
 	if baiduResponse.ErrorMsg != "" {
-		return types.NewError(fmt.Errorf(baiduResponse.ErrorMsg), types.ErrorCodeBadResponseBody), nil
+		return types.NewError(fmt.Errorf("%s", baiduResponse.ErrorMsg), types.ErrorCodeBadResponseBody), nil
 	}
 	fullTextResponse := responseBaidu2OpenAI(&baiduResponse)
 	jsonResponse, err := json.Marshal(fullTextResponse)
@@ -175,7 +175,7 @@ func baiduEmbeddingHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *ht
 		return types.NewError(err, types.ErrorCodeBadResponseBody), nil
 	}
 	if baiduResponse.ErrorMsg != "" {
-		return types.NewError(fmt.Errorf(baiduResponse.ErrorMsg), types.ErrorCodeBadResponseBody), nil
+		return types.NewError(fmt.Errorf("%s", baiduResponse.ErrorMsg), types.ErrorCodeBadResponseBody), nil
 	}
 	fullTextResponse := embeddingResponseBaidu2OpenAI(&baiduResponse)
 	jsonResponse, err := json.Marshal(fullTextResponse)

relay/channel/claude/relay-claude.go

@@ -483,9 +483,11 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *dto.ClaudeResponse
 				}
 			}
 		} else if claudeResponse.Type == "message_delta" {
-			finishReason := stopReasonClaude2OpenAI(*claudeResponse.Delta.StopReason)
-			if finishReason != "null" {
-				choice.FinishReason = &finishReason
+			if claudeResponse.Delta != nil && claudeResponse.Delta.StopReason != nil {
+				finishReason := stopReasonClaude2OpenAI(*claudeResponse.Delta.StopReason)
+				if finishReason != "null" {
+					choice.FinishReason = &finishReason
+				}
 			}
 			//claudeUsage = &claudeResponse.Usage
 		} else if claudeResponse.Type == "message_stop" {

relay/channel/coze/relay-coze.go

@@ -208,7 +208,7 @@ func handleCozeEvent(c *gin.Context, event string, data string, responseText *st
 			return
 		}
 
-		common.SysLog(fmt.Sprintf("stream event error: ", errorData.Code, errorData.Message))
+		common.SysLog(fmt.Sprintf("stream event error: %v %v", errorData.Code, errorData.Message))
 	}
 }
 

relay/channel/gemini/adaptor.go

@@ -13,6 +13,7 @@ import (
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/relay/constant"
 	"github.com/QuantumNous/new-api/setting/model_setting"
+	"github.com/QuantumNous/new-api/setting/reasoning"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -137,7 +138,7 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			info.UpstreamModelName = strings.TrimSuffix(info.UpstreamModelName, "-thinking")
 		} else if strings.HasSuffix(info.UpstreamModelName, "-nothinking") {
 			info.UpstreamModelName = strings.TrimSuffix(info.UpstreamModelName, "-nothinking")
-		} else if baseModel, level := parseThinkingLevelSuffix(info.UpstreamModelName); level != "" {
+		} else if baseModel, level, ok := reasoning.TrimEffortSuffix(info.UpstreamModelName); ok && level != "" {
 			info.UpstreamModelName = baseModel
 		}
 	}

relay/channel/gemini/relay-gemini-native.go

@@ -94,10 +94,10 @@ func GeminiTextGenerationStreamHandler(c *gin.Context, info *relaycommon.RelayIn
 	helper.SetEventStreamHeaders(c)
 
 	return geminiStreamHandler(c, info, resp, func(data string, geminiResponse *dto.GeminiChatResponse) bool {
-		// 直接发送 GeminiChatResponse 响应
 		err := helper.StringData(c, data)
 		if err != nil {
-			logger.LogError(c, err.Error())
+			logger.LogError(c, "failed to write stream data: "+err.Error())
+			return false
 		}
 		info.SendResponseCount++
 		return true

relay/channel/gemini/relay-gemini.go

@@ -98,6 +98,7 @@ func clampThinkingBudget(modelName string, budget int) int {
 // "effort": "high" - Allocates a large portion of tokens for reasoning (approximately 80% of max_tokens)
 // "effort": "medium" - Allocates a moderate portion of tokens (approximately 50% of max_tokens)
 // "effort": "low" - Allocates a smaller portion of tokens (approximately 20% of max_tokens)
+// "effort": "minimal" - Allocates a minimal portion of tokens (approximately 5% of max_tokens)
 func clampThinkingBudgetByEffort(modelName string, effort string) int {
 	isNew25Pro := isNew25ProModel(modelName)
 	is25FlashLite := is25FlashLiteModel(modelName)
@@ -118,18 +119,12 @@ func clampThinkingBudgetByEffort(modelName string, effort string) int {
 		maxBudget = maxBudget * 50 / 100
 	case "low":
 		maxBudget = maxBudget * 20 / 100
+	case "minimal":
+		maxBudget = maxBudget * 5 / 100
 	}
 	return clampThinkingBudget(modelName, maxBudget)
 }
 
-func parseThinkingLevelSuffix(modelName string) (string, string) {
-	base, level, ok := reasoning.TrimEffortSuffix(modelName)
-	if !ok {
-		return modelName, ""
-	}
-	return base, level
-}
-
 func ThinkingAdaptor(geminiRequest *dto.GeminiChatRequest, info *relaycommon.RelayInfo, oaiRequest ...dto.GeneralOpenAIRequest) {
 	if model_setting.GetGeminiSettings().ThinkingAdapterEnabled {
 		modelName := info.UpstreamModelName
@@ -186,7 +181,7 @@ func ThinkingAdaptor(geminiRequest *dto.GeminiChatRequest, info *relaycommon.Rel
 					ThinkingBudget: common.GetPointer(0),
 				}
 			}
-		} else if _, level := parseThinkingLevelSuffix(modelName); level != "" {
+		} else if _, level, ok := reasoning.TrimEffortSuffix(info.UpstreamModelName); ok && level != "" {
 			geminiRequest.GenerationConfig.ThinkingConfig = &dto.GeminiThinkingConfig{
 				IncludeThoughts: true,
 				ThinkingLevel:   level,

relay/channel/openai/adaptor.go

@@ -42,7 +42,7 @@ type Adaptor struct {
 // support OAI models: o1-mini/o3-mini/o4-mini/o1/o3 etc...
 // minimal effort only available in gpt-5
 func parseReasoningEffortFromModelSuffix(model string) (string, string) {
-	effortSuffixes := []string{"-high", "-minimal", "-low", "-medium", "-none"}
+	effortSuffixes := []string{"-high", "-minimal", "-low", "-medium", "-none", "-xhigh"}
 	for _, suffix := range effortSuffixes {
 		if strings.HasSuffix(model, suffix) {
 			effort := strings.TrimPrefix(suffix, "-")

relay/channel/openai/audio.go

@@ -0,0 +1,145 @@
+package openai
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"math"
+	"net/http"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/QuantumNous/new-api/relay/helper"
+	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/types"
+	"github.com/gin-gonic/gin"
+)
+
+func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) *dto.Usage {
+	// the status code has been judged before, if there is a body reading failure,
+	// it should be regarded as a non-recoverable error, so it should not return err for external retry.
+	// Analogous to nginx's load balancing, it will only retry if it can't be requested or
+	// if the upstream returns a specific status code, once the upstream has already written the header,
+	// the subsequent failure of the response body should be regarded as a non-recoverable error,
+	// and can be terminated directly.
+	defer service.CloseResponseBodyGracefully(resp)
+	usage := &dto.Usage{}
+	usage.PromptTokens = info.GetEstimatePromptTokens()
+	usage.TotalTokens = info.GetEstimatePromptTokens()
+	for k, v := range resp.Header {
+		c.Writer.Header().Set(k, v[0])
+	}
+	c.Writer.WriteHeader(resp.StatusCode)
+
+	if info.IsStream {
+		helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+			if service.SundaySearch(data, "usage") {
+				var simpleResponse dto.SimpleResponse
+				err := common.Unmarshal([]byte(data), &simpleResponse)
+				if err != nil {
+					logger.LogError(c, err.Error())
+				}
+				if simpleResponse.Usage.TotalTokens != 0 {
+					usage.PromptTokens = simpleResponse.Usage.InputTokens
+					usage.CompletionTokens = simpleResponse.OutputTokens
+					usage.TotalTokens = simpleResponse.TotalTokens
+				}
+			}
+			_ = helper.StringData(c, data)
+			return true
+		})
+	} else {
+		common.SetContextKey(c, constant.ContextKeyLocalCountTokens, true)
+		// 读取响应体到缓冲区
+		bodyBytes, err := io.ReadAll(resp.Body)
+		if err != nil {
+			logger.LogError(c, fmt.Sprintf("failed to read TTS response body: %v", err))
+			c.Writer.WriteHeaderNow()
+			return usage
+		}
+
+		// 写入响应到客户端
+		c.Writer.WriteHeaderNow()
+		_, err = c.Writer.Write(bodyBytes)
+		if err != nil {
+			logger.LogError(c, fmt.Sprintf("failed to write TTS response: %v", err))
+		}
+
+		// 计算音频时长并更新 usage
+		audioFormat := "mp3" // 默认格式
+		if audioReq, ok := info.Request.(*dto.AudioRequest); ok && audioReq.ResponseFormat != "" {
+			audioFormat = audioReq.ResponseFormat
+		}
+
+		var duration float64
+		var durationErr error
+
+		if audioFormat == "pcm" {
+			// PCM 格式没有文件头，根据 OpenAI TTS 的 PCM 参数计算时长
+			// 采样率: 24000 Hz, 位深度: 16-bit (2 bytes), 声道数: 1
+			const sampleRate = 24000
+			const bytesPerSample = 2
+			const channels = 1
+			duration = float64(len(bodyBytes)) / float64(sampleRate*bytesPerSample*channels)
+		} else {
+			ext := "." + audioFormat
+			reader := bytes.NewReader(bodyBytes)
+			duration, durationErr = common.GetAudioDuration(c.Request.Context(), reader, ext)
+		}
+
+		usage.PromptTokensDetails.TextTokens = usage.PromptTokens
+
+		if durationErr != nil {
+			logger.LogWarn(c, fmt.Sprintf("failed to get audio duration: %v", durationErr))
+			// 如果无法获取时长，则设置保底的 CompletionTokens，根据body大小计算
+			sizeInKB := float64(len(bodyBytes)) / 1000.0
+			estimatedTokens := int(math.Ceil(sizeInKB)) // 粗略估算每KB约等于1 token
+			usage.CompletionTokens = estimatedTokens
+			usage.CompletionTokenDetails.AudioTokens = estimatedTokens
+		} else if duration > 0 {
+			// 计算 token: ceil(duration) / 60.0 * 1000，即每分钟 1000 tokens
+			completionTokens := int(math.Round(math.Ceil(duration) / 60.0 * 1000))
+			usage.CompletionTokens = completionTokens
+			usage.CompletionTokenDetails.AudioTokens = completionTokens
+		}
+		usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
+	}
+
+	return usage
+}
+
+func OpenaiSTTHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, responseFormat string) (*types.NewAPIError, *dto.Usage) {
+	defer service.CloseResponseBodyGracefully(resp)
+
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError), nil
+	}
+	// 写入新的 response body
+	service.IOCopyBytesGracefully(c, resp, responseBody)
+
+	var responseData struct {
+		Usage *dto.Usage `json:"usage"`
+	}
+	if err := common.Unmarshal(responseBody, &responseData); err == nil && responseData.Usage != nil {
+		if responseData.Usage.TotalTokens > 0 {
+			usage := responseData.Usage
+			if usage.PromptTokens == 0 {
+				usage.PromptTokens = usage.InputTokens
+			}
+			if usage.CompletionTokens == 0 {
+				usage.CompletionTokens = usage.OutputTokens
+			}
+			return nil, usage
+		}
+	}
+
+	usage := &dto.Usage{}
+	usage.PromptTokens = info.GetEstimatePromptTokens()
+	usage.CompletionTokens = 0
+	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
+	return nil, usage
+}

relay/channel/openai/helper.go

@@ -172,7 +172,7 @@ func handleLastResponse(lastStreamData string, responseId *string, createAt *int
 	shouldSendLastResp *bool) error {
 
 	var lastStreamResponse dto.ChatCompletionsStreamResponse
-	if err := json.Unmarshal(common.StringToByteSlice(lastStreamData), &lastStreamResponse); err != nil {
+	if err := common.Unmarshal(common.StringToByteSlice(lastStreamData), &lastStreamResponse); err != nil {
 		return err
 	}
 

relay/channel/openai/relay-openai.go

@@ -1,7 +1,6 @@
 package openai
 
 import (
-	"encoding/json"
 	"fmt"
 	"io"
 	"net/http"
@@ -151,7 +150,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 		var streamResp struct {
 			Usage *dto.Usage `json:"usage"`
 		}
-		err := json.Unmarshal([]byte(secondLastStreamData), &streamResp)
+		err := common.Unmarshal([]byte(secondLastStreamData), &streamResp)
 		if err == nil && streamResp.Usage != nil && service.ValidUsage(streamResp.Usage) {
 			usage = streamResp.Usage
 			containStreamUsage = true
@@ -327,68 +326,6 @@ func streamTTSResponse(c *gin.Context, resp *http.Response) {
 	}
 }
 
-func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) *dto.Usage {
-	// the status code has been judged before, if there is a body reading failure,
-	// it should be regarded as a non-recoverable error, so it should not return err for external retry.
-	// Analogous to nginx's load balancing, it will only retry if it can't be requested or
-	// if the upstream returns a specific status code, once the upstream has already written the header,
-	// the subsequent failure of the response body should be regarded as a non-recoverable error,
-	// and can be terminated directly.
-	defer service.CloseResponseBodyGracefully(resp)
-	usage := &dto.Usage{}
-	usage.PromptTokens = info.GetEstimatePromptTokens()
-	usage.TotalTokens = info.GetEstimatePromptTokens()
-	for k, v := range resp.Header {
-		c.Writer.Header().Set(k, v[0])
-	}
-	c.Writer.WriteHeader(resp.StatusCode)
-
-	isStreaming := resp.ContentLength == -1 || resp.Header.Get("Content-Length") == ""
-	if isStreaming {
-		streamTTSResponse(c, resp)
-	} else {
-		c.Writer.WriteHeaderNow()
-		_, err := io.Copy(c.Writer, resp.Body)
-		if err != nil {
-			logger.LogError(c, err.Error())
-		}
-	}
-	return usage
-}
-
-func OpenaiSTTHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, responseFormat string) (*types.NewAPIError, *dto.Usage) {
-	defer service.CloseResponseBodyGracefully(resp)
-
-	responseBody, err := io.ReadAll(resp.Body)
-	if err != nil {
-		return types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError), nil
-	}
-	// 写入新的 response body
-	service.IOCopyBytesGracefully(c, resp, responseBody)
-
-	var responseData struct {
-		Usage *dto.Usage `json:"usage"`
-	}
-	if err := json.Unmarshal(responseBody, &responseData); err == nil && responseData.Usage != nil {
-		if responseData.Usage.TotalTokens > 0 {
-			usage := responseData.Usage
-			if usage.PromptTokens == 0 {
-				usage.PromptTokens = usage.InputTokens
-			}
-			if usage.CompletionTokens == 0 {
-				usage.CompletionTokens = usage.OutputTokens
-			}
-			return nil, usage
-		}
-	}
-
-	usage := &dto.Usage{}
-	usage.PromptTokens = info.GetEstimatePromptTokens()
-	usage.CompletionTokens = 0
-	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
-	return nil, usage
-}
-
 func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.NewAPIError, *dto.RealtimeUsage) {
 	if info == nil || info.ClientWs == nil || info.TargetWs == nil {
 		return types.NewError(fmt.Errorf("invalid websocket connection"), types.ErrorCodeBadResponse), nil
@@ -659,7 +596,7 @@ func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, res
 		if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
 			usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
 		}
-	case constant.ChannelTypeZhipu_v4:
+	case constant.ChannelTypeZhipu_v4, constant.ChannelTypeMoonshot:
 		if usage.PromptTokensDetails.CachedTokens == 0 {
 			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
 				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
@@ -687,7 +624,7 @@ func extractCachedTokensFromBody(body []byte) (int, bool) {
 		} `json:"usage"`
 	}
 
-	if err := json.Unmarshal(body, &payload); err != nil {
+	if err := common.Unmarshal(body, &payload); err != nil {
 		return 0, false
 	}
 

relay/channel/task/jimeng/adaptor.go

@@ -196,7 +196,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 	}
 
 	if jResp.Code != 10000 {
-		taskErr = service.TaskErrorWrapper(fmt.Errorf(jResp.Message), fmt.Sprintf("%d", jResp.Code), http.StatusInternalServerError)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", jResp.Message), fmt.Sprintf("%d", jResp.Code), http.StatusInternalServerError)
 		return
 	}
 

relay/channel/task/kling/adaptor.go

@@ -186,7 +186,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 		return
 	}
 	if kResp.Code != 0 {
-		taskErr = service.TaskErrorWrapperLocal(fmt.Errorf(kResp.Message), "task_failed", http.StatusBadRequest)
+		taskErr = service.TaskErrorWrapperLocal(fmt.Errorf("%s", kResp.Message), "task_failed", http.StatusBadRequest)
 		return
 	}
 	ov := dto.NewOpenAIVideo()

relay/channel/task/suno/adaptor.go

@@ -105,7 +105,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 		return
 	}
 	if !sunoResponse.IsSuccess() {
-		taskErr = service.TaskErrorWrapper(fmt.Errorf(sunoResponse.Message), sunoResponse.Code, http.StatusInternalServerError)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", sunoResponse.Message), sunoResponse.Code, http.StatusInternalServerError)
 		return
 	}
 

relay/channel/vertex/adaptor.go

@@ -51,10 +51,43 @@ type Adaptor struct {
 }
 
 func (a *Adaptor) ConvertGeminiRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeminiChatRequest) (any, error) {
+	// Vertex AI does not support functionResponse.id; keep it stripped here for consistency.
+	if model_setting.GetGeminiSettings().RemoveFunctionResponseIdEnabled {
+		removeFunctionResponseID(request)
+	}
 	geminiAdaptor := gemini.Adaptor{}
 	return geminiAdaptor.ConvertGeminiRequest(c, info, request)
 }
 
+func removeFunctionResponseID(request *dto.GeminiChatRequest) {
+	if request == nil {
+		return
+	}
+
+	if len(request.Contents) > 0 {
+		for i := range request.Contents {
+			if len(request.Contents[i].Parts) == 0 {
+				continue
+			}
+			for j := range request.Contents[i].Parts {
+				part := &request.Contents[i].Parts[j]
+				if part.FunctionResponse == nil {
+					continue
+				}
+				if len(part.FunctionResponse.ID) > 0 {
+					part.FunctionResponse.ID = nil
+				}
+			}
+		}
+	}
+
+	if len(request.Requests) > 0 {
+		for i := range request.Requests {
+			removeFunctionResponseID(&request.Requests[i])
+		}
+	}
+}
+
 func (a *Adaptor) ConvertClaudeRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.ClaudeRequest) (any, error) {
 	if v, ok := claudeModelMap[info.UpstreamModelName]; ok {
 		c.Set("request_model", v)

relay/channel/zhipu_4v/dto.go

@@ -4,6 +4,7 @@ import (
 	"time"
 
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/types"
 )
 
 //	type ZhipuMessage struct {
@@ -37,7 +38,7 @@ type ZhipuV4Response struct {
 	Model               string                         `json:"model"`
 	TextResponseChoices []dto.OpenAITextResponseChoice `json:"choices"`
 	Usage               dto.Usage                      `json:"usage"`
-	Error               dto.OpenAIError                `json:"error"`
+	Error               types.OpenAIError              `json:"error"`
 }
 
 //

relay/common/relay_info.go

@@ -11,6 +11,7 @@ import (
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	relayconstant "github.com/QuantumNous/new-api/relay/constant"
+	"github.com/QuantumNous/new-api/setting/model_setting"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -81,8 +82,9 @@ type TokenCountMeta struct {
 type RelayInfo struct {
 	TokenId           int
 	TokenKey          string
+	TokenGroup        string
 	UserId            int
-	UsingGroup        string // 使用的分组
+	UsingGroup        string // 使用的分组，当auto跨分组重试时，会变动
 	UserGroup         string // 用户所在分组
 	TokenUnlimited    bool
 	StartTime         time.Time
@@ -373,6 +375,12 @@ func genBaseRelayInfo(c *gin.Context, request dto.Request) *RelayInfo {
 	//channelId := common.GetContextKeyInt(c, constant.ContextKeyChannelId)
 	//paramOverride := common.GetContextKeyStringMap(c, constant.ContextKeyChannelParamOverride)
 
+	tokenGroup := common.GetContextKeyString(c, constant.ContextKeyTokenGroup)
+	// 当令牌分组为空时，表示使用用户分组
+	if tokenGroup == "" {
+		tokenGroup = common.GetContextKeyString(c, constant.ContextKeyUserGroup)
+	}
+
 	startTime := common.GetContextKeyTime(c, constant.ContextKeyRequestStartTime)
 	if startTime.IsZero() {
 		startTime = time.Now()
@@ -400,6 +408,7 @@ func genBaseRelayInfo(c *gin.Context, request dto.Request) *RelayInfo {
 		TokenId:        common.GetContextKeyInt(c, constant.ContextKeyTokenId),
 		TokenKey:       common.GetContextKeyString(c, constant.ContextKeyTokenKey),
 		TokenUnlimited: common.GetContextKeyBool(c, constant.ContextKeyTokenUnlimited),
+		TokenGroup:     tokenGroup,
 
 		isFirstResponse: true,
 		RelayMode:       relayconstant.Path2RelayMode(c.Request.URL.Path),
@@ -626,3 +635,47 @@ func RemoveDisabledFields(jsonData []byte, channelOtherSettings dto.ChannelOther
 	}
 	return jsonDataAfter, nil
 }
+
+// RemoveGeminiDisabledFields removes disabled fields from Gemini request JSON data
+// Currently supports removing functionResponse.id field which Vertex AI does not support
+func RemoveGeminiDisabledFields(jsonData []byte) ([]byte, error) {
+	if !model_setting.GetGeminiSettings().RemoveFunctionResponseIdEnabled {
+		return jsonData, nil
+	}
+
+	var data map[string]interface{}
+	if err := common.Unmarshal(jsonData, &data); err != nil {
+		common.SysError("RemoveGeminiDisabledFields Unmarshal error: " + err.Error())
+		return jsonData, nil
+	}
+
+	// Process contents array
+	// Handle both camelCase (functionResponse) and snake_case (function_response)
+	if contents, ok := data["contents"].([]interface{}); ok {
+		for _, content := range contents {
+			if contentMap, ok := content.(map[string]interface{}); ok {
+				if parts, ok := contentMap["parts"].([]interface{}); ok {
+					for _, part := range parts {
+						if partMap, ok := part.(map[string]interface{}); ok {
+							// Check functionResponse (camelCase)
+							if funcResp, ok := partMap["functionResponse"].(map[string]interface{}); ok {
+								delete(funcResp, "id")
+							}
+							// Check function_response (snake_case)
+							if funcResp, ok := partMap["function_response"].(map[string]interface{}); ok {
+								delete(funcResp, "id")
+							}
+						}
+					}
+				}
+			}
+		}
+	}
+
+	jsonDataAfter, err := common.Marshal(data)
+	if err != nil {
+		common.SysError("RemoveGeminiDisabledFields Marshal error: " + err.Error())
+		return jsonData, nil
+	}
+	return jsonDataAfter, nil
+}

relay/compatible_handler.go

@@ -181,7 +181,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 		return newApiErr
 	}
 
-	if strings.HasPrefix(info.OriginModelName, "gpt-4o-audio") {
+	if usage.(*dto.Usage).CompletionTokenDetails.AudioTokens > 0 || usage.(*dto.Usage).PromptTokensDetails.AudioTokens > 0 {
 		service.PostAudioConsumeQuota(c, info, usage.(*dto.Usage), "")
 	} else {
 		postConsumeQuota(c, info, usage.(*dto.Usage), "")
@@ -300,14 +300,20 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage
 	if !relayInfo.PriceData.UsePrice {
 		baseTokens := dPromptTokens
 		// 减去 cached tokens
+		// Anthropic API 的 input_tokens 已经不包含缓存 tokens，不需要减去
+		// OpenAI/OpenRouter 等 API 的 prompt_tokens 包含缓存 tokens，需要减去
 		var cachedTokensWithRatio decimal.Decimal
 		if !dCacheTokens.IsZero() {
-			baseTokens = baseTokens.Sub(dCacheTokens)
+			if relayInfo.ChannelType != constant.ChannelTypeAnthropic {
+				baseTokens = baseTokens.Sub(dCacheTokens)
+			}
 			cachedTokensWithRatio = dCacheTokens.Mul(dCacheRatio)
 		}
 		var dCachedCreationTokensWithRatio decimal.Decimal
 		if !dCachedCreationTokens.IsZero() {
-			baseTokens = baseTokens.Sub(dCachedCreationTokens)
+			if relayInfo.ChannelType != constant.ChannelTypeAnthropic {
+				baseTokens = baseTokens.Sub(dCachedCreationTokens)
+			}
 			dCachedCreationTokensWithRatio = dCachedCreationTokens.Mul(dCachedCreationRatio)
 		}
 

relay/helper/common.go

@@ -14,15 +14,28 @@ import (
 	"github.com/gorilla/websocket"
 )
 
-func FlushWriter(c *gin.Context) error {
-	if c.Writer == nil {
+func FlushWriter(c *gin.Context) (err error) {
+	defer func() {
+		if r := recover(); r != nil {
+			err = fmt.Errorf("flush panic recovered: %v", r)
+		}
+	}()
+
+	if c == nil || c.Writer == nil {
 		return nil
 	}
-	if flusher, ok := c.Writer.(http.Flusher); ok {
-		flusher.Flush()
-		return nil
+
+	if c.Request != nil && c.Request.Context().Err() != nil {
+		return fmt.Errorf("request context done: %w", c.Request.Context().Err())
 	}
-	return errors.New("streaming error: flusher not found")
+
+	flusher, ok := c.Writer.(http.Flusher)
+	if !ok {
+		return errors.New("streaming error: flusher not found")
+	}
+
+	flusher.Flush()
+	return nil
 }
 
 func SetEventStreamHeaders(c *gin.Context) {
@@ -66,17 +79,31 @@ func ResponseChunkData(c *gin.Context, resp dto.ResponsesStreamResponse, data st
 }
 
 func StringData(c *gin.Context, str string) error {
-	//str = strings.TrimPrefix(str, "data: ")
-	//str = strings.TrimSuffix(str, "\r")
+	if c == nil || c.Writer == nil {
+		return errors.New("context or writer is nil")
+	}
+
+	if c.Request != nil && c.Request.Context().Err() != nil {
+		return fmt.Errorf("request context done: %w", c.Request.Context().Err())
+	}
+
 	c.Render(-1, common.CustomEvent{Data: "data: " + str})
-	_ = FlushWriter(c)
-	return nil
+	return FlushWriter(c)
 }
 
 func PingData(c *gin.Context) error {
-	c.Writer.Write([]byte(": PING\n\n"))
-	_ = FlushWriter(c)
-	return nil
+	if c == nil || c.Writer == nil {
+		return errors.New("context or writer is nil")
+	}
+
+	if c.Request != nil && c.Request.Context().Err() != nil {
+		return fmt.Errorf("request context done: %w", c.Request.Context().Err())
+	}
+
+	if _, err := c.Writer.Write([]byte(": PING\n\n")); err != nil {
+		return fmt.Errorf("write ping data failed: %w", err)
+	}
+	return FlushWriter(c)
 }
 
 func ObjectData(c *gin.Context, object interface{}) error {

relay/relay_task.go

@@ -196,7 +196,7 @@ func RelayTaskSubmit(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.
 	// handle response
 	if resp != nil && resp.StatusCode != http.StatusOK {
 		responseBody, _ := io.ReadAll(resp.Body)
-		taskErr = service.TaskErrorWrapper(fmt.Errorf(string(responseBody)), "fail_to_fetch_task", resp.StatusCode)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", string(responseBody)), "fail_to_fetch_task", resp.StatusCode)
 		return
 	}
 

service/channel_select.go

@@ -11,31 +11,151 @@ import (
 	"github.com/gin-gonic/gin"
 )
 
-func CacheGetRandomSatisfiedChannel(c *gin.Context, group string, modelName string, retry int) (*model.Channel, string, error) {
+type RetryParam struct {
+	Ctx          *gin.Context
+	TokenGroup   string
+	ModelName    string
+	Retry        *int
+	resetNextTry bool
+}
+
+func (p *RetryParam) GetRetry() int {
+	if p.Retry == nil {
+		return 0
+	}
+	return *p.Retry
+}
+
+func (p *RetryParam) SetRetry(retry int) {
+	p.Retry = &retry
+}
+
+func (p *RetryParam) IncreaseRetry() {
+	if p.resetNextTry {
+		p.resetNextTry = false
+		return
+	}
+	if p.Retry == nil {
+		p.Retry = new(int)
+	}
+	*p.Retry++
+}
+
+func (p *RetryParam) ResetRetryNextTry() {
+	p.resetNextTry = true
+}
+
+// CacheGetRandomSatisfiedChannel tries to get a random channel that satisfies the requirements.
+// 尝试获取一个满足要求的随机渠道。
+//
+// For "auto" tokenGroup with cross-group Retry enabled:
+// 对于启用了跨分组重试的 "auto" tokenGroup：
+//
+//   - Each group will exhaust all its priorities before moving to the next group.
+//     每个分组会用完所有优先级后才会切换到下一个分组。
+//
+//   - Uses ContextKeyAutoGroupIndex to track current group index.
+//     使用 ContextKeyAutoGroupIndex 跟踪当前分组索引。
+//
+//   - Uses ContextKeyAutoGroupRetryIndex to track the global Retry count when current group started.
+//     使用 ContextKeyAutoGroupRetryIndex 跟踪当前分组开始时的全局重试次数。
+//
+//   - priorityRetry = Retry - startRetryIndex, represents the priority level within current group.
+//     priorityRetry = Retry - startRetryIndex，表示当前分组内的优先级级别。
+//
+//   - When GetRandomSatisfiedChannel returns nil (priorities exhausted), moves to next group.
+//     当 GetRandomSatisfiedChannel 返回 nil（优先级用完）时，切换到下一个分组。
+//
+// Example flow (2 groups, each with 2 priorities, RetryTimes=3):
+// 示例流程（2个分组，每个有2个优先级，RetryTimes=3）：
+//
+//	Retry=0: GroupA, priority0 (startRetryIndex=0, priorityRetry=0)
+//	         分组A, 优先级0
+//
+//	Retry=1: GroupA, priority1 (startRetryIndex=0, priorityRetry=1)
+//	         分组A, 优先级1
+//
+//	Retry=2: GroupA exhausted → GroupB, priority0 (startRetryIndex=2, priorityRetry=0)
+//	         分组A用完 → 分组B, 优先级0
+//
+//	Retry=3: GroupB, priority1 (startRetryIndex=2, priorityRetry=1)
+//	         分组B, 优先级1
+func CacheGetRandomSatisfiedChannel(param *RetryParam) (*model.Channel, string, error) {
 	var channel *model.Channel
 	var err error
-	selectGroup := group
-	userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
-	if group == "auto" {
+	selectGroup := param.TokenGroup
+	userGroup := common.GetContextKeyString(param.Ctx, constant.ContextKeyUserGroup)
+
+	if param.TokenGroup == "auto" {
 		if len(setting.GetAutoGroups()) == 0 {
 			return nil, selectGroup, errors.New("auto groups is not enabled")
 		}
-		for _, autoGroup := range GetUserAutoGroup(userGroup) {
-			logger.LogDebug(c, "Auto selecting group:", autoGroup)
-			channel, _ = model.GetRandomSatisfiedChannel(autoGroup, modelName, retry)
-			if channel == nil {
-				continue
-			} else {
-				c.Set("auto_group", autoGroup)
-				selectGroup = autoGroup
-				logger.LogDebug(c, "Auto selected group:", autoGroup)
-				break
+		autoGroups := GetUserAutoGroup(userGroup)
+
+		// startGroupIndex: the group index to start searching from
+		// startGroupIndex: 开始搜索的分组索引
+		startGroupIndex := 0
+		crossGroupRetry := common.GetContextKeyBool(param.Ctx, constant.ContextKeyTokenCrossGroupRetry)
+
+		if lastGroupIndex, exists := common.GetContextKey(param.Ctx, constant.ContextKeyAutoGroupIndex); exists {
+			if idx, ok := lastGroupIndex.(int); ok {
+				startGroupIndex = idx
 			}
 		}
+
+		for i := startGroupIndex; i < len(autoGroups); i++ {
+			autoGroup := autoGroups[i]
+			// Calculate priorityRetry for current group
+			// 计算当前分组的 priorityRetry
+			priorityRetry := param.GetRetry()
+			// If moved to a new group, reset priorityRetry and update startRetryIndex
+			// 如果切换到新分组，重置 priorityRetry 并更新 startRetryIndex
+			if i > startGroupIndex {
+				priorityRetry = 0
+			}
+			logger.LogDebug(param.Ctx, "Auto selecting group: %s, priorityRetry: %d", autoGroup, priorityRetry)
+
+			channel, _ = model.GetRandomSatisfiedChannel(autoGroup, param.ModelName, priorityRetry)
+			if channel == nil {
+				// Current group has no available channel for this model, try next group
+				// 当前分组没有该模型的可用渠道，尝试下一个分组
+				logger.LogDebug(param.Ctx, "No available channel in group %s for model %s at priorityRetry %d, trying next group", autoGroup, param.ModelName, priorityRetry)
+				// 重置状态以尝试下一个分组
+				common.SetContextKey(param.Ctx, constant.ContextKeyAutoGroupIndex, i+1)
+				common.SetContextKey(param.Ctx, constant.ContextKeyAutoGroupRetryIndex, 0)
+				// Reset retry counter so outer loop can continue for next group
+				// 重置重试计数器，以便外层循环可以为下一个分组继续
+				param.SetRetry(0)
+				continue
+			}
+			common.SetContextKey(param.Ctx, constant.ContextKeyAutoGroup, autoGroup)
+			selectGroup = autoGroup
+			logger.LogDebug(param.Ctx, "Auto selected group: %s", autoGroup)
+
+			// Prepare state for next retry
+			// 为下一次重试准备状态
+			if crossGroupRetry && priorityRetry >= common.RetryTimes {
+				// Current group has exhausted all retries, prepare to switch to next group
+				// This request still uses current group, but next retry will use next group
+				// 当前分组已用完所有重试次数，准备切换到下一个分组
+				// 本次请求仍使用当前分组，但下次重试将使用下一个分组
+				logger.LogDebug(param.Ctx, "Current group %s retries exhausted (priorityRetry=%d >= RetryTimes=%d), preparing switch to next group for next retry", autoGroup, priorityRetry, common.RetryTimes)
+				common.SetContextKey(param.Ctx, constant.ContextKeyAutoGroupIndex, i+1)
+				// Reset retry counter so outer loop can continue for next group
+				// 重置重试计数器，以便外层循环可以为下一个分组继续
+				param.SetRetry(0)
+				param.ResetRetryNextTry()
+			} else {
+				// Stay in current group, save current state
+				// 保持在当前分组，保存当前状态
+				common.SetContextKey(param.Ctx, constant.ContextKeyAutoGroupIndex, i)
+			}
+			break
+		}
 	} else {
-		channel, err = model.GetRandomSatisfiedChannel(group, modelName, retry)
+		channel, err = model.GetRandomSatisfiedChannel(param.TokenGroup, param.ModelName, param.GetRetry())
 		if err != nil {
-			return nil, group, err
+			return nil, param.TokenGroup, err
 		}
 	}
 	return channel, selectGroup, nil

service/convert.go

@@ -389,25 +389,29 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 				}
 
 				idx := blockIndex
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_start",
-					ContentBlock: &dto.ClaudeMediaMessage{
-						Id:    toolCall.ID,
-						Type:  "tool_use",
-						Name:  toolCall.Function.Name,
-						Input: map[string]interface{}{},
-					},
-				})
+				if toolCall.Function.Name != "" {
+					claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+						Index: &idx,
+						Type:  "content_block_start",
+						ContentBlock: &dto.ClaudeMediaMessage{
+							Id:    toolCall.ID,
+							Type:  "tool_use",
+							Name:  toolCall.Function.Name,
+							Input: map[string]interface{}{},
+						},
+					})
+				}
 
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type:        "input_json_delta",
-						PartialJson: &toolCall.Function.Arguments,
-					},
-				})
+				if len(toolCall.Function.Arguments) > 0 {
+					claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+						Index: &idx,
+						Type:  "content_block_delta",
+						Delta: &dto.ClaudeMediaMessage{
+							Type:        "input_json_delta",
+							PartialJson: &toolCall.Function.Arguments,
+						},
+					})
+				}
 
 				info.ClaudeConvertInfo.Index = blockIndex
 			}

service/error.go

@@ -90,24 +90,38 @@ func RelayErrorHandler(ctx context.Context, resp *http.Response, showBodyWhenFai
 	}
 	CloseResponseBodyGracefully(resp)
 	var errResponse dto.GeneralErrorResponse
+	buildErrWithBody := func(message string) error {
+		if message == "" {
+			return fmt.Errorf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody))
+		}
+		return fmt.Errorf("bad response status code %d, message: %s, body: %s", resp.StatusCode, message, string(responseBody))
+	}
 
 	err = common.Unmarshal(responseBody, &errResponse)
 	if err != nil {
 		if showBodyWhenFail {
-			newApiErr.Err = fmt.Errorf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody))
+			newApiErr.Err = buildErrWithBody("")
 		} else {
-			if common.DebugEnabled {
-				logger.LogInfo(ctx, fmt.Sprintf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody)))
-			}
+			logger.LogError(ctx, fmt.Sprintf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody)))
 			newApiErr.Err = fmt.Errorf("bad response status code %d", resp.StatusCode)
 		}
 		return
 	}
-	if errResponse.Error.Message != "" {
+
+	if common.GetJsonType(errResponse.Error) == "object" {
 		// General format error (OpenAI, Anthropic, Gemini, etc.)
-		newApiErr = types.WithOpenAIError(errResponse.Error, resp.StatusCode)
-	} else {
-		newApiErr = types.NewOpenAIError(errors.New(errResponse.ToMessage()), types.ErrorCodeBadResponseStatusCode, resp.StatusCode)
+		oaiError := errResponse.TryToOpenAIError()
+		if oaiError != nil {
+			newApiErr = types.WithOpenAIError(*oaiError, resp.StatusCode)
+			if showBodyWhenFail {
+				newApiErr.Err = buildErrWithBody(newApiErr.Error())
+			}
+			return
+		}
+	}
+	newApiErr = types.NewOpenAIError(errors.New(errResponse.ToMessage()), types.ErrorCodeBadResponseStatusCode, resp.StatusCode)
+	if showBodyWhenFail {
+		newApiErr.Err = buildErrWithBody(newApiErr.Error())
 	}
 	return
 }

service/quota.go

@@ -95,7 +95,7 @@ func PreWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usag
 		return err
 	}
 
-	token, err := model.GetTokenByKey(strings.TrimLeft(relayInfo.TokenKey, "sk-"), false)
+	token, err := model.GetTokenByKey(strings.TrimPrefix(relayInfo.TokenKey, "sk-"), false)
 	if err != nil {
 		return err
 	}
@@ -108,7 +108,7 @@ func PreWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usag
 	groupRatio := ratio_setting.GetGroupRatio(relayInfo.UsingGroup)
 	modelRatio, _, _ := ratio_setting.GetModelRatio(modelName)
 
-	autoGroup, exists := ctx.Get("auto_group")
+	autoGroup, exists := common.GetContextKey(ctx, constant.ContextKeyAutoGroup)
 	if exists {
 		groupRatio = ratio_setting.GetGroupRatio(autoGroup.(string))
 		log.Printf("final group ratio: %f", groupRatio)

service/token_counter.go

@@ -317,7 +317,7 @@ func EstimateRequestToken(c *gin.Context, meta *types.TokenCountMeta, info *rela
 	for i, file := range meta.Files {
 		switch file.FileType {
 		case types.FileTypeImage:
-			if common.IsOpenAITextModel(info.OriginModelName) {
+			if common.IsOpenAITextModel(model) {
 				token, err := getImageToken(file, model, info.IsStream)
 				if err != nil {
 					return 0, fmt.Errorf("error counting image token, media index[%d], original data[%s], err: %v", i, file.OriginData, err)

setting/model_setting/gemini.go

@@ -4,7 +4,7 @@ import (
 	"github.com/QuantumNous/new-api/setting/config"
 )
 
-// GeminiSettings 定义Gemini模型的配置
+// GeminiSettings defines Gemini model configuration. 注意bool要以enabled结尾才可以生效编辑
 type GeminiSettings struct {
 	SafetySettings                        map[string]string `json:"safety_settings"`
 	VersionSettings                       map[string]string `json:"version_settings"`
@@ -12,6 +12,7 @@ type GeminiSettings struct {
 	ThinkingAdapterEnabled                bool              `json:"thinking_adapter_enabled"`
 	ThinkingAdapterBudgetTokensPercentage float64           `json:"thinking_adapter_budget_tokens_percentage"`
 	FunctionCallThoughtSignatureEnabled   bool              `json:"function_call_thought_signature_enabled"`
+	RemoveFunctionResponseIdEnabled       bool              `json:"remove_function_response_id_enabled"`
 }
 
 // 默认配置
@@ -30,6 +31,7 @@ var defaultGeminiSettings = GeminiSettings{
 	ThinkingAdapterEnabled:                false,
 	ThinkingAdapterBudgetTokensPercentage: 0.6,
 	FunctionCallThoughtSignatureEnabled:   true,
+	RemoveFunctionResponseIdEnabled:       true,
 }
 
 // 全局实例

setting/ratio_setting/model_ratio.go

@@ -7,7 +7,6 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/setting/operation_setting"
-	"github.com/QuantumNous/new-api/setting/reasoning"
 )
 
 // from songquanpeng/one-api
@@ -297,6 +296,7 @@ var defaultModelPrice = map[string]float64{
 	"mj_upload":                      0.05,
 	"sora-2":                         0.3,
 	"sora-2-pro":                     0.5,
+	"gpt-4o-mini-tts":                0.3,
 }
 
 var defaultAudioRatio = map[string]float64{
@@ -304,11 +304,13 @@ var defaultAudioRatio = map[string]float64{
 	"gpt-4o-mini-audio-preview":    66.67,
 	"gpt-4o-realtime-preview":      8,
 	"gpt-4o-mini-realtime-preview": 16.67,
+	"gpt-4o-mini-tts":              25,
 }
 
 var defaultAudioCompletionRatio = map[string]float64{
 	"gpt-4o-realtime":      2,
 	"gpt-4o-mini-realtime": 2,
+	"gpt-4o-mini-tts":      1,
 }
 
 var (
@@ -536,7 +538,10 @@ func getHardcodedCompletionModelRatio(name string) (float64, bool) {
 			if name == "gpt-4o-2024-05-13" {
 				return 3, true
 			}
-			return 4, true
+			if strings.HasPrefix(name, "gpt-4o-mini-tts") {
+				return 20, false
+			}
+			return 4, false
 		}
 		// gpt-5 匹配
 		if strings.HasPrefix(name, "gpt-5") {
@@ -823,10 +828,6 @@ func FormatMatchingModelName(name string) string {
 		name = handleThinkingBudgetModel(name, "gemini-2.5-pro", "gemini-2.5-pro-thinking-*")
 	}
 
-	if base, _, ok := reasoning.TrimEffortSuffix(name); ok {
-		name = base
-	}
-
 	if strings.HasPrefix(name, "gpt-4-gizmo") {
 		name = "gpt-4-gizmo-*"
 	}

setting/reasoning/suffix.go

@@ -6,7 +6,7 @@ import (
 	"github.com/samber/lo"
 )
 
-var EffortSuffixes = []string{"-high", "-medium", "-low"}
+var EffortSuffixes = []string{"-high", "-medium", "-low", "-minimal"}
 
 // TrimEffortSuffix -> modelName level(low) exists
 func TrimEffortSuffix(modelName string) (string, string, bool) {

setting/system_setting/discord.go

@@ -3,9 +3,9 @@ package system_setting
 import "github.com/QuantumNous/new-api/setting/config"
 
 type DiscordSettings struct {
-	Enabled               bool   `json:"enabled"`
-	ClientId              string `json:"client_id"`
-	ClientSecret          string `json:"client_secret"`
+	Enabled      bool   `json:"enabled"`
+	ClientId     string `json:"client_id"`
+	ClientSecret string `json:"client_secret"`
 }
 
 // 默认配置

types/error.go

@@ -1,6 +1,7 @@
 package types
 
 import (
+	"encoding/json"
 	"errors"
 	"fmt"
 	"net/http"
@@ -10,10 +11,11 @@ import (
 )
 
 type OpenAIError struct {
-	Message string `json:"message"`
-	Type    string `json:"type"`
-	Param   string `json:"param"`
-	Code    any    `json:"code"`
+	Message  string          `json:"message"`
+	Type     string          `json:"type"`
+	Param    string          `json:"param"`
+	Code     any             `json:"code"`
+	Metadata json.RawMessage `json:"metadata,omitempty"`
 }
 
 type ClaudeError struct {
@@ -92,6 +94,15 @@ type NewAPIError struct {
 	errorType      ErrorType
 	errorCode      ErrorCode
 	StatusCode     int
+	Metadata       json.RawMessage
+}
+
+// Unwrap enables errors.Is / errors.As to work with NewAPIError by exposing the underlying error.
+func (e *NewAPIError) Unwrap() error {
+	if e == nil {
+		return nil
+	}
+	return e.Err
 }
 
 func (e *NewAPIError) GetErrorCode() ErrorCode {
@@ -293,6 +304,13 @@ func WithOpenAIError(openAIError OpenAIError, statusCode int, ops ...NewAPIError
 		Err:        errors.New(openAIError.Message),
 		errorCode:  ErrorCode(code),
 	}
+	// OpenRouter
+	if len(openAIError.Metadata) > 0 {
+		openAIError.Message = fmt.Sprintf("%s (%s)", openAIError.Message, openAIError.Metadata)
+		e.Metadata = openAIError.Metadata
+		e.RelayError = openAIError
+		e.Err = errors.New(openAIError.Message)
+	}
 	for _, op := range ops {
 		op(e)
 	}

web/bun.lock

@@ -48,6 +48,7 @@
         "@so1ve/prettier-config": "^3.1.0",
         "@vitejs/plugin-react": "^4.2.1",
         "autoprefixer": "^10.4.21",
+        "code-inspector-plugin": "^1.3.3",
         "eslint": "8.57.0",
         "eslint-plugin-header": "^3.1.1",
         "eslint-plugin-react-hooks": "^5.2.0",
@@ -139,6 +140,18 @@
 
     "@chevrotain/utils": ["@chevrotain/utils@11.0.3", "", {}, "sha512-YslZMgtJUyuMbZ+aKvfF3x1f5liK4mWNxghFRv7jqRR9C3R3fAOGTTKvxXDa2Y1s9zSbcpuO0cAxDYsc9SrXoQ=="],
 
+    "@code-inspector/core": ["@code-inspector/core@1.3.3", "", { "dependencies": { "@vue/compiler-dom": "^3.5.13", "chalk": "^4.1.1", "dotenv": "^16.1.4", "launch-ide": "1.3.0", "portfinder": "^1.0.28" } }, "sha512-1SUCY/XiJ3LuA9TPfS9i7/cUcmdLsgB0chuDcP96ixB2tvYojzgCrglP7CHUGZa1dtWuRLuCiDzkclLetpV4ew=="],
+
+    "@code-inspector/esbuild": ["@code-inspector/esbuild@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3" } }, "sha512-GzX5LQbvh9DXINSUyWymG8Y7u5Tq4oJAnnrCoRiYxQvKBUuu2qVMzpZHIA2iDGxvazgZvr2OK+Sh/We4LutViA=="],
+
+    "@code-inspector/mako": ["@code-inspector/mako@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3" } }, "sha512-YPTHwpDtz9zn1vimMcJFCM6ELdBoivY7t2GzgY/iCTfgm6pu1H+oWZiBC35edqYAB7+xE8frspnNsmBhsrA36A=="],
+
+    "@code-inspector/turbopack": ["@code-inspector/turbopack@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3", "@code-inspector/webpack": "1.3.3" } }, "sha512-XhqsMtts/Int64LkpO00b4rlg1bw0otlRebX8dSVgZfsujj+Jdv2ngKmQ6RBN3vgj/zV7BfgBLeGgJn7D1kT3A=="],
+
+    "@code-inspector/vite": ["@code-inspector/vite@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3", "chalk": "4.1.1" } }, "sha512-phsHVYBsxAhfi6jJ+vpmxuF6jYMuVbozs5e8pkEJL2hQyGVkzP77vfCh1wzmQHcmKUKb2tlrFcvAsRb7oA1W7w=="],
+
+    "@code-inspector/webpack": ["@code-inspector/webpack@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3" } }, "sha512-qYih7syRXgM45KaWFNNk5Ed4WitVQHCI/2s/DZMFaF1Y2FA9qd1wPGiggNeqdcUsjf9TvVBQw/89gPQZIGwSqQ=="],
+
     "@dnd-kit/accessibility": ["@dnd-kit/accessibility@3.1.1", "", { "dependencies": { "tslib": "^2.0.0" }, "peerDependencies": { "react": ">=16.8.0" } }, "sha512-2P+YgaXF+gRsIihwwY1gCsQSYnu9Zyj2py8kY5fFvUM1qm2WA2u639R6YNVfU4GWr+ZM5mqEsfHZZLoRONbemw=="],
 
     "@dnd-kit/core": ["@dnd-kit/core@6.3.1", "", { "dependencies": { "@dnd-kit/accessibility": "^3.1.1", "@dnd-kit/utilities": "^3.2.2", "tslib": "^2.0.0" }, "peerDependencies": { "react": ">=16.8.0", "react-dom": ">=16.8.0" } }, "sha512-xkGBRQQab4RLwgXxoqETICr6S5JlogafbhNsidmrkVv2YRs5MLwpjoF2qpiGjQt8S9AoxtIV603s0GIUpY5eYQ=="],
@@ -713,6 +726,12 @@
 
     "@vitejs/plugin-react": ["@vitejs/plugin-react@4.3.4", "", { "dependencies": { "@babel/core": "^7.26.0", "@babel/plugin-transform-react-jsx-self": "^7.25.9", "@babel/plugin-transform-react-jsx-source": "^7.25.9", "@types/babel__core": "^7.20.5", "react-refresh": "^0.14.2" }, "peerDependencies": { "vite": "^4.2.0 || ^5.0.0 || ^6.0.0" } }, "sha512-SCCPBJtYLdE8PX/7ZQAs1QAZ8Jqwih+0VBLum1EGqmCCQal+MIUqLCzj3ZUy8ufbC0cAM4LRlSTm7IQJwWT4ug=="],
 
+    "@vue/compiler-core": ["@vue/compiler-core@3.5.26", "", { "dependencies": { "@babel/parser": "^7.28.5", "@vue/shared": "3.5.26", "entities": "^7.0.0", "estree-walker": "^2.0.2", "source-map-js": "^1.2.1" } }, "sha512-vXyI5GMfuoBCnv5ucIT7jhHKl55Y477yxP6fc4eUswjP8FG3FFVFd41eNDArR+Uk3QKn2Z85NavjaxLxOC19/w=="],
+
+    "@vue/compiler-dom": ["@vue/compiler-dom@3.5.26", "", { "dependencies": { "@vue/compiler-core": "3.5.26", "@vue/shared": "3.5.26" } }, "sha512-y1Tcd3eXs834QjswshSilCBnKGeQjQXB6PqFn/1nxcQw4pmG42G8lwz+FZPAZAby6gZeHSt/8LMPfZ4Rb+Bd/A=="],
+
+    "@vue/shared": ["@vue/shared@3.5.26", "", {}, "sha512-7Z6/y3uFI5PRoKeorTOSXKcDj0MSasfNNltcslbFrPpcw6aXRUALq4IfJlaTRspiWIUOEZbrpM+iQGmCOiWe4A=="],
+
     "abs-svg-path": ["abs-svg-path@0.1.1", "", {}, "sha512-d8XPSGjfyzlXC3Xx891DJRyZfqk5JU0BJrDQcsWomFIV1/BIzPW5HDH5iDdWpqWaav0YVIEzT1RHTwWr0FFshA=="],
 
     "acorn": ["acorn@8.15.0", "", { "bin": { "acorn": "bin/acorn" } }, "sha512-NZyJarBfL7nWwIq+FDL6Zp/yHEhePMNnnJ0y3qfieCrmNvYct8uvtiV41UvlSe6apAfk0fY1FbWx+NwfmpvtTg=="],
@@ -747,6 +766,8 @@
 
     "astring": ["astring@1.9.0", "", { "bin": { "astring": "bin/astring" } }, "sha512-LElXdjswlqjWrPpJFg1Fx4wpkOCxj1TDHlSV4PlaRxHGWko024xICaa97ZkMfs6DRKlCguiAI+rbXv5GWwXIkg=="],
 
+    "async": ["async@3.2.6", "", {}, "sha512-htCUDlxyyCLMgaM3xXg0C0LW2xqfuQ6p05pCEIsXuyQ+a1koYKTuBMzRNwmybfLgvJDMd0r1LTn4+E0Ti6C2AA=="],
+
     "async-validator": ["async-validator@3.5.2", "", {}, "sha512-8eLCg00W9pIRZSB781UUX/H6Oskmm8xloZfr09lz5bikRpBVDlJ3hRVuxxP1SxcwsEYfJ4IU8Q19Y8/893r3rQ=="],
 
     "asynckit": ["asynckit@0.4.0", "", {}, "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q=="],
@@ -793,7 +814,7 @@
 
     "ccount": ["ccount@2.0.1", "", {}, "sha512-eyrF0jiFpY+3drT6383f1qhkbGsLSifNAjA61IUjZjmLCWjItY6LB9ft9YhoDgwfmclB2zhu51Lc7+95b8NRAg=="],
 
-    "chalk": ["chalk@4.1.2", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA=="],
+    "chalk": ["chalk@4.1.1", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-diHzdDKxcU+bAsUboHLPEDQiw0qEe0qd7SYUn3HgcFlWgbDcfLGswOHYeGrHKzG9z6UYf01d9VFMfZxPM1xZSg=="],
 
     "character-entities": ["character-entities@2.0.2", "", {}, "sha512-shx7oQ0Awen/BRIdkjkvz54PnEEI/EjwXDSIZp86/KKdbafHh1Df/RYGBhn4hbe2+uKC9FnT5UCEdyPz3ai9hQ=="],
 
@@ -825,6 +846,8 @@
 
     "clsx": ["clsx@2.1.1", "", {}, "sha512-eYm0QWBtUrBWZWG0d386OGAw16Z995PiOVo2B7bjWSbHedGl5e0ZWaq65kOGgUSNesEIDkB9ISbTg/JK9dhCZA=="],
 
+    "code-inspector-plugin": ["code-inspector-plugin@1.3.3", "", { "dependencies": { "@code-inspector/core": "1.3.3", "@code-inspector/esbuild": "1.3.3", "@code-inspector/mako": "1.3.3", "@code-inspector/turbopack": "1.3.3", "@code-inspector/vite": "1.3.3", "@code-inspector/webpack": "1.3.3", "chalk": "4.1.1" } }, "sha512-yDi84v5tgXFSZLLXqHl/Mc2qy9d2CxcYhIaP192NhqTG1zA5uVtiNIzvDAXh5Vaqy8QGYkvBfbG/i55b/sXaSQ=="],
+
     "collapse-white-space": ["collapse-white-space@2.1.0", "", {}, "sha512-loKTxY1zCOuG4j9f6EPnuyyYkf58RnhhWTvRoZEokgB+WbdXehfjFviyOVYkqzEWz1Q5kRiZdBYS5SwxbQYwzw=="],
 
     "color-convert": ["color-convert@2.0.1", "", { "dependencies": { "color-name": "~1.1.4" } }, "sha512-RRECPsj7iu/xb5oKYcsFHSppFNnsj/52OVTRKb4zP5onXwVF3zVmmToNcOfGC+CRDpfK/U584fMg38ZHCaElKQ=="],
@@ -975,6 +998,8 @@
 
     "dompurify": ["dompurify@3.2.6", "", { "optionalDependencies": { "@types/trusted-types": "^2.0.7" } }, "sha512-/2GogDQlohXPZe6D6NOgQvXLPSYBqIWMnZ8zzOhn09REE4eyAzb+Hed3jhoM9OkuaJ8P6ZGTTVWQKAi8ieIzfQ=="],
 
+    "dotenv": ["dotenv@16.6.1", "", {}, "sha512-uBq4egWHTcTt33a72vpSG0z3HnPuIl6NqYcTrKEg2azoEyl2hpW0zqlxysq2pK9HlDIHyHyakeYaYnSAwd8bow=="],
+
     "dunder-proto": ["dunder-proto@1.0.1", "", { "dependencies": { "call-bind-apply-helpers": "^1.0.1", "es-errors": "^1.3.0", "gopd": "^1.2.0" } }, "sha512-KIN/nDJBQRcXw0MLVhZE9iQHmG68qAVIBg9CqmUYjmQIhgij9U5MFvrqkUL5FbtyyzZuOeOt0zdeRe4UY7ct+A=="],
 
     "eastasianwidth": ["eastasianwidth@0.2.0", "", {}, "sha512-I88TYZWc9XiYHRQ4/3c5rjjfgkjhLyW2luGIheGERbNQ6OY7yTybanSpDXZa8y7VUP9YmDcYa+eyq4ca7iLqWA=="],
@@ -985,7 +1010,7 @@
 
     "emoji-regex": ["emoji-regex@10.4.0", "", {}, "sha512-EC+0oUMY1Rqm4O6LLrgjtYDvcVYTy7chDnM4Q7030tP4Kwj3u/pR6gP9ygnp2CJMK5Gq+9Q2oqmrFJAz01DXjw=="],
 
-    "entities": ["entities@6.0.0", "", {}, "sha512-aKstq2TDOndCn4diEyp9Uq/Flu2i1GlLkc6XIDQSDMuaFE3OPW5OphLCyQ5SpSJZTb4reN+kTcYru5yIfXoRPw=="],
+    "entities": ["entities@7.0.0", "", {}, "sha512-FDWG5cmEYf2Z00IkYRhbFrwIwvdFKH07uV8dvNy0omp/Qb1xcyCWp2UDtcwJF4QZZvk0sLudP6/hAu42TaqVhQ=="],
 
     "error-ex": ["error-ex@1.3.2", "", { "dependencies": { "is-arrayish": "^0.2.1" } }, "sha512-7dFHNmqeFSEt2ZBsCriorKnn3Z2pj+fd9kmI6QoWw4//DL+icEBfc0U7qJCisqrTsKTjw4fNFy2pW9OqStD84g=="],
 
@@ -1305,6 +1330,8 @@
 
     "langium": ["langium@3.3.1", "", { "dependencies": { "chevrotain": "~11.0.3", "chevrotain-allstar": "~0.3.0", "vscode-languageserver": "~9.0.1", "vscode-languageserver-textdocument": "~1.0.11", "vscode-uri": "~3.0.8" } }, "sha512-QJv/h939gDpvT+9SiLVlY7tZC3xB2qK57v0J04Sh9wpMb6MP1q8gB21L3WIo8T5P1MSMg3Ep14L7KkDCFG3y4w=="],
 
+    "launch-ide": ["launch-ide@1.3.0", "", { "dependencies": { "chalk": "^4.1.1", "dotenv": "^16.1.4" } }, "sha512-pxiF+HVNMV0dDc6Z0q89RDmzMF9XmSGaOn4ueTegjMy3cUkezc3zrki5PCiz68zZIqAuhW7iwoWX7JO4Kn6B0A=="],
+
     "layout-base": ["layout-base@1.0.2", "", {}, "sha512-8h2oVEZNktL4BH2JCOI90iD1yXwL6iNW7KcCKT2QZgQJR2vbqDsldCTPRU9NifTCqHZci57XvQQ15YTu+sTYPg=="],
 
     "leva": ["leva@0.10.0", "", { "dependencies": { "@radix-ui/react-portal": "1.0.2", "@radix-ui/react-tooltip": "1.0.5", "@stitches/react": "^1.2.8", "@use-gesture/react": "^10.2.5", "colord": "^2.9.2", "dequal": "^2.0.2", "merge-value": "^1.0.0", "react-colorful": "^5.5.1", "react-dropzone": "^12.0.0", "v8n": "^1.3.3", "zustand": "^3.6.9" }, "peerDependencies": { "react": "^18.0.0 || ^19.0.0", "react-dom": "^18.0.0 || ^19.0.0" } }, "sha512-RiNJWmeqQdKIeHuVXgshmxIHu144a2AMYtLxKf8Nm1j93pisDPexuQDHKNdQlbo37wdyDQibLjY9JKGIiD7gaw=="],
@@ -1595,6 +1622,8 @@
 
     "polished": ["polished@4.3.1", "", { "dependencies": { "@babel/runtime": "^7.17.8" } }, "sha512-OBatVyC/N7SCW/FaDHrSd+vn0o5cS855TOmYi4OkdWUMSJCET/xip//ch8xGUvtr3i44X9LVyWwQlRMTN3pwSA=="],
 
+    "portfinder": ["portfinder@1.0.38", "", { "dependencies": { "async": "^3.2.6", "debug": "^4.3.6" } }, "sha512-rEwq/ZHlJIKw++XtLAO8PPuOQA/zaPJOZJ37BVuN97nLpMJeuDVLVGRwbFoBgLudgdTMP2hdRJP++H+8QOA3vg=="],
+
     "postcss": ["postcss@8.5.3", "", { "dependencies": { "nanoid": "^3.3.8", "picocolors": "^1.1.1", "source-map-js": "^1.2.1" } }, "sha512-dle9A3yYxlBSrt8Fu+IpjGT8SY8hN0mlaA6GY8t0P5PjIOZemULz/E2Bnm/2dcUOena75OTNkHI76uZBNUUq3A=="],
 
     "postcss-import": ["postcss-import@15.1.0", "", { "dependencies": { "postcss-value-parser": "^4.0.0", "read-cache": "^1.0.0", "resolve": "^1.1.7" }, "peerDependencies": { "postcss": "^8.0.0" } }, "sha512-hpr+J05B2FVYUAXHeK1YyI267J/dDDhMU6B6civm8hSY1jYJnBXxzKDKDswzJmtLHryrjhnDjqqp/49t8FALew=="],
@@ -2081,6 +2110,8 @@
 
     "@babel/traverse/globals": ["globals@11.12.0", "", {}, "sha512-WOBp/EEGUiIsJSp7wcv/y6MO+lV9UoncWqxuFfm8eBwzWNgyfBd6Gz+IeKQ9jCmyhoH99g15M3T+QaVHFjizVA=="],
 
+    "@code-inspector/core/chalk": ["chalk@4.1.2", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA=="],
+
     "@douyinfe/semi-foundation/remark-gfm": ["remark-gfm@4.0.0", "", { "dependencies": { "@types/mdast": "^4.0.0", "mdast-util-gfm": "^3.0.0", "micromark-extension-gfm": "^3.0.0", "remark-parse": "^11.0.0", "remark-stringify": "^11.0.0", "unified": "^11.0.0" } }, "sha512-U92vJgBPkbw4Zfu/IiW2oTZLSL3Zpv+uI7My2eq8JxKgqraFdU8YUGicEJCEgSbeaG+QDFqIcwwfMTOEelPxuA=="],
 
     "@emotion/babel-plugin/@emotion/hash": ["@emotion/hash@0.9.2", "", {}, "sha512-MyqliTZGuOm3+5ZRSaaBGP3USLw6+EGykkwZns2EPC5g8jJ4z9OrdZY9apkl3+UP9+sdz76YYkwCKP5gh8iY3g=="],
@@ -2131,6 +2162,10 @@
 
     "@visactor/vrender-kits/roughjs": ["roughjs@4.5.2", "", { "dependencies": { "path-data-parser": "^0.1.0", "points-on-curve": "^0.2.0", "points-on-path": "^0.2.1" } }, "sha512-2xSlLDKdsWyFxrveYWk9YQ/Y9UfK38EAMRNkYkMqYBJvPX8abCa9PN0x3w02H8Oa6/0bcZICJU+U95VumPqseg=="],
 
+    "@vue/compiler-core/@babel/parser": ["@babel/parser@7.28.5", "", { "dependencies": { "@babel/types": "^7.28.5" }, "bin": "./bin/babel-parser.js" }, "sha512-KKBU1VGYR7ORr3At5HAtUQ+TV3SzRCXmA/8OdDZiLDBIZxVyzXuztPjfLd3BV1PRAQGCMWWSHYhL0F8d5uHBDQ=="],
+
+    "@vue/compiler-core/estree-walker": ["estree-walker@2.0.2", "", {}, "sha512-Rfkk/Mp/DL7JVje3u18FxFujQlTNR2q6QfMSMB7AvCBx91NGj/ba3kCfza0f6dVDbw7YlRf/nDrn7pQrCCyQ/w=="],
+
     "antd/rc-collapse": ["rc-collapse@3.9.0", "", { "dependencies": { "@babel/runtime": "^7.10.1", "classnames": "2.x", "rc-motion": "^2.3.4", "rc-util": "^5.27.0" }, "peerDependencies": { "react": ">=16.9.0", "react-dom": ">=16.9.0" } }, "sha512-swDdz4QZ4dFTo4RAUMLL50qP0EY62N2kvmk2We5xYdRwcRn8WcYtuetCJpwpaCbUfUt5+huLpVxhvmnK+PHrkA=="],
 
     "antd/scroll-into-view-if-needed": ["scroll-into-view-if-needed@3.1.0", "", { "dependencies": { "compute-scroll-into-view": "^3.0.2" } }, "sha512-49oNpRjWRvnU8NyGVmUaYG4jtTkNonFZI86MmGRDqBphEK2EXT9gdEUoQPZhuBM8yWHxCWbobltqYO5M4XrUvQ=="],
@@ -2155,6 +2190,8 @@
 
     "esast-util-from-js/acorn": ["acorn@8.14.0", "", { "bin": { "acorn": "bin/acorn" } }, "sha512-cl669nCJTZBsL97OF4kUQm5g5hC2uihk0NxY3WENAC0TYdILVkAyHymAntgxGkl7K+t0cXIrH5siy5S4XkFycA=="],
 
+    "eslint/chalk": ["chalk@4.1.2", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA=="],
+
     "extend-shallow/is-extendable": ["is-extendable@0.1.1", "", {}, "sha512-5BMULNob1vgFX6EjQw5izWDxrecWK9AM72rugNr0TFldMOi0fj6Jk+zeKIt0xGj4cEfQIJth4w3OKWOJ4f+AFw=="],
 
     "fast-glob/glob-parent": ["glob-parent@5.1.2", "", { "dependencies": { "is-glob": "^4.0.1" } }, "sha512-AOIgSQCepiJYwP3ARnGx+5VnTu2HBYdzbGP45eLw1vr3zB3vZLeyed1sC9hnbcOc9/SrMyM5RPQrkGz4aS9Zow=="],
@@ -2181,6 +2218,8 @@
 
     "katex/commander": ["commander@8.3.0", "", {}, "sha512-OkTL9umf+He2DZkUq8f8J9of7yL6RJKI24dVITBmNfZBmri9zYZQrKkuXiKhyfPSu8tUhnVBB1iKXevvnlR4Ww=="],
 
+    "launch-ide/chalk": ["chalk@4.1.2", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA=="],
+
     "leva/react-dropzone": ["react-dropzone@12.1.0", "", { "dependencies": { "attr-accept": "^2.2.2", "file-selector": "^0.5.0", "prop-types": "^15.8.1" }, "peerDependencies": { "react": ">= 16.8" } }, "sha512-iBYHA1rbopIvtzokEX4QubO6qk5IF/x3BtKGu74rF2JkQDXnwC4uO/lHKpaw4PJIV6iIAYOlwLv2FpiGyqHNog=="],
 
     "mdast-util-find-and-replace/escape-string-regexp": ["escape-string-regexp@5.0.0", "", {}, "sha512-/veY75JbMK4j1yjvuUxuVsiS/hr/4iHs9FTT6cgTexxdE0Ly/glccBAkloH/DofkjRbZU3bnoj38mOmhkZ0lHw=="],
@@ -2201,6 +2240,8 @@
 
     "parse-entities/@types/unist": ["@types/unist@2.0.11", "", {}, "sha512-CmBKiL6NNo/OqgmMn95Fk9Whlp2mtvIv+KNpQKN2F4SjvrEesubTRWGYSg+BnWZOnlCaSTU1sMpsBOzgbYhnsA=="],
 
+    "parse5/entities": ["entities@6.0.0", "", {}, "sha512-aKstq2TDOndCn4diEyp9Uq/Flu2i1GlLkc6XIDQSDMuaFE3OPW5OphLCyQ5SpSJZTb4reN+kTcYru5yIfXoRPw=="],
+
     "path-scurry/lru-cache": ["lru-cache@11.2.2", "", {}, "sha512-F9ODfyqML2coTIsQpSkRHnLSZMtkU8Q+mSfcaIyKwy58u+8k5nvAYeiNhsyMARvzNcXJ9QfWVrcPsC9e9rAxtg=="],
 
     "prettier-package-json/commander": ["commander@4.1.1", "", {}, "sha512-NOKm8xhkzAjzFx8B2v5OAHT+u5pRQc2UCa2Vq9jYL/31o2wi9mxBA7LIFs3sV5VSC49z6pEhfbMULvShKj26WA=="],
@@ -2269,6 +2310,8 @@
 
     "@radix-ui/react-primitive/@radix-ui/react-slot/@radix-ui/react-compose-refs": ["@radix-ui/react-compose-refs@1.0.0", "", { "dependencies": { "@babel/runtime": "^7.13.10" }, "peerDependencies": { "react": "^16.8 || ^17.0 || ^18.0" } }, "sha512-0KaSv6sx787/hK3eF53iOkiSLwAGlFMx5lotrqD2pTjB18KbybKoEIgkNZTKC60YECDQTKGTRcDBILwZVqVKvA=="],
 
+    "@vue/compiler-core/@babel/parser/@babel/types": ["@babel/types@7.28.5", "", { "dependencies": { "@babel/helper-string-parser": "^7.27.1", "@babel/helper-validator-identifier": "^7.28.5" } }, "sha512-qQ5m48eI/MFLQ5PxQj4PFaprjyCTLI37ElWMmNs0K8Lk3dVeOdNpB3ks8jc7yM5CDmVC73eMVk/trk3fgmrUpA=="],
+
     "antd/scroll-into-view-if-needed/compute-scroll-into-view": ["compute-scroll-into-view@3.1.1", "", {}, "sha512-VRhuHOLoKYOy4UbilLbUzbYg93XLjv2PncJC50EuTWPA3gaja1UjBsUP/D/9/juV3vQFr6XBEzn9KCAHdUvOHw=="],
 
     "cytoscape-fcose/cose-base/layout-base": ["layout-base@2.0.1", "", {}, "sha512-dp3s92+uNI1hWIpPGH3jK2kxE2lMjdXdr+DH8ynZHpd6PUlH6x6cbuXnoMmiNumznqaNO31xu9e79F0uuZ0JFg=="],
@@ -2325,6 +2368,10 @@
 
     "@radix-ui/react-popper/@floating-ui/react-dom/@floating-ui/dom/@floating-ui/core": ["@floating-ui/core@0.7.3", "", {}, "sha512-buc8BXHmG9l82+OQXOFU3Kr2XQx9ys01U/Q9HMIrZ300iLc8HLMgh7dcCqgYzAzf4BkoQvDcXf5Y+CuEZ5JBYg=="],
 
+    "@vue/compiler-core/@babel/parser/@babel/types/@babel/helper-string-parser": ["@babel/helper-string-parser@7.27.1", "", {}, "sha512-qMlSxKbpRlAridDExk92nSobyDdpPijUq2DW6oDnUqd0iOGxmQjyqhMIihI9+zv4LPyZdRje2cavWPbCbWm3eA=="],
+
+    "@vue/compiler-core/@babel/parser/@babel/types/@babel/helper-validator-identifier": ["@babel/helper-validator-identifier@7.28.5", "", {}, "sha512-qSs4ifwzKJSV39ucNjsvc6WVHs6b7S03sOh2OcHF9UHfVPqWWALUsNUVzhSBiItjRZoLHx7nIarVjqKVusUZ1Q=="],
+
     "simplify-geojson/concat-stream/readable-stream/string_decoder": ["string_decoder@0.10.31", "", {}, "sha512-ev2QzSzWPYmy9GuqfIVildA4OdcGLeFZQrq5ys6RtiuF+RQQiZWr8TZNyAcuVXyQRYfEO+MsoB/1BuQVhOJuoQ=="],
 
     "sucrase/glob/minimatch/brace-expansion": ["brace-expansion@2.0.1", "", { "dependencies": { "balanced-match": "^1.0.0" } }, "sha512-XnAIvQ8eM+kC6aULx6wuQiwVsnzsi9d3WxzV3FpWTGA19F621kwdbsAcFKXgKUHZWsy+mY6iL1sHTxWEFCytDA=="],

web/package.json

@@ -78,15 +78,16 @@
     "@so1ve/prettier-config": "^3.1.0",
     "@vitejs/plugin-react": "^4.2.1",
     "autoprefixer": "^10.4.21",
+    "code-inspector-plugin": "^1.3.3",
     "eslint": "8.57.0",
     "eslint-plugin-header": "^3.1.1",
     "eslint-plugin-react-hooks": "^5.2.0",
+    "i18next-cli": "^1.10.3",
     "postcss": "^8.5.3",
     "prettier": "^3.0.0",
     "tailwindcss": "^3",
     "typescript": "4.4.2",
-    "vite": "^5.2.0",
-    "i18next-cli": "^1.10.3"
+    "vite": "^5.2.0"
   },
   "prettier": {
     "singleQuote": true,

web/src/components/settings/ModelSetting.jsx

@@ -32,6 +32,7 @@ const ModelSetting = () => {
     'gemini.safety_settings': '',
     'gemini.version_settings': '',
     'gemini.supported_imagine_models': '',
+    'gemini.remove_function_response_id_enabled': true,
     'claude.model_headers_settings': '',
     'claude.thinking_adapter_enabled': true,
     'claude.default_max_tokens': '',
@@ -64,6 +65,7 @@ const ModelSetting = () => {
             item.value = JSON.stringify(JSON.parse(item.value), null, 2);
           }
         }
+        // Keep boolean config keys ending with enabled/Enabled so UI parses correctly.
         if (item.key.endsWith('Enabled') || item.key.endsWith('enabled')) {
           newInputs[item.key] = toBoolean(item.value);
         } else {

web/src/components/settings/PersonalSetting.jsx

@@ -314,10 +314,10 @@ const PersonalSetting = () => {
   };
 
   const changePassword = async () => {
-    if (inputs.original_password === '') {
-      showError(t('请输入原密码！'));
-      return;
-    }
+    // if (inputs.original_password === '') {
+    //   showError(t('请输入原密码！'));
+    //   return;
+    // }
     if (inputs.set_new_password === '') {
       showError(t('请输入新密码！'));
       return;

web/src/components/table/channels/ChannelsColumnDefs.jsx

@@ -39,7 +39,11 @@ import {
   showError,
 } from '../../../helpers';
 import { CHANNEL_OPTIONS } from '../../../constants';
-import { IconTreeTriangleDown, IconMore } from '@douyinfe/semi-icons';
+import {
+  IconTreeTriangleDown,
+  IconMore,
+  IconAlertTriangle,
+} from '@douyinfe/semi-icons';
 import { FaRandom } from 'react-icons/fa';
 
 // Render functions
@@ -187,6 +191,28 @@ const renderResponseTime = (responseTime, t) => {
   }
 };
 
+const isRequestPassThroughEnabled = (record) => {
+  if (!record || record.children !== undefined) {
+    return false;
+  }
+  const settingValue = record.setting;
+  if (!settingValue) {
+    return false;
+  }
+  if (typeof settingValue === 'object') {
+    return settingValue.pass_through_body_enabled === true;
+  }
+  if (typeof settingValue !== 'string') {
+    return false;
+  }
+  try {
+    const parsed = JSON.parse(settingValue);
+    return parsed?.pass_through_body_enabled === true;
+  } catch (error) {
+    return false;
+  }
+};
+
 export const getChannelsColumns = ({
   t,
   COLUMN_KEYS,
@@ -219,8 +245,9 @@ export const getChannelsColumns = ({
       title: t('名称'),
       dataIndex: 'name',
       render: (text, record, index) => {
-        if (record.remark && record.remark.trim() !== '') {
-          return (
+        const passThroughEnabled = isRequestPassThroughEnabled(record);
+        const nameNode =
+          record.remark && record.remark.trim() !== '' ? (
             <Tooltip
               content={
                 <div className='flex flex-col gap-2 max-w-xs'>
@@ -250,9 +277,32 @@ export const getChannelsColumns = ({
             >
               <span>{text}</span>
             </Tooltip>
+          ) : (
+            <span>{text}</span>
           );
+
+        if (!passThroughEnabled) {
+          return nameNode;
         }
-        return text;
+
+        return (
+          <Space spacing={6} align='center'>
+            {nameNode}
+            <Tooltip
+              content={t(
+                '该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。',
+              )}
+              trigger='hover'
+              position='topLeft'
+            >
+              <span className='inline-flex items-center'>
+                <IconAlertTriangle
+                  style={{ color: 'var(--semi-color-warning)' }}
+                />
+              </span>
+            </Tooltip>
+          </Space>
+        );
       },
     },
     {

web/src/components/table/channels/index.jsx

@@ -18,6 +18,8 @@ For commercial licensing, please contact support@quantumnous.com
 */
 
 import React from 'react';
+import { Banner } from '@douyinfe/semi-ui';
+import { IconAlertTriangle } from '@douyinfe/semi-icons';
 import CardPro from '../../common/ui/CardPro';
 import ChannelsTable from './ChannelsTable';
 import ChannelsActions from './ChannelsActions';
@@ -63,6 +65,22 @@ const ChannelsPage = () => {
       />
 
       {/* Main Content */}
+      {channelsData.globalPassThroughEnabled ? (
+        <Banner
+          type='warning'
+          closeIcon={null}
+          icon={
+            <IconAlertTriangle
+              size='large'
+              style={{ color: 'var(--semi-color-warning)' }}
+            />
+          }
+          description={channelsData.t(
+            '已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。',
+          )}
+          style={{ marginBottom: 12 }}
+        />
+      ) : null}
       <CardPro
         type='type3'
         tabsArea={<ChannelsTabs {...channelsData} />}

web/src/components/table/channels/modals/EditChannelModal.jsx

@@ -1604,7 +1604,7 @@ const EditChannelModal = (props) => {
         >
           {() => (
             <Spin spinning={loading}>
-              <div className='p-2' ref={formContainerRef}>
+              <div className='p-2 space-y-3' ref={formContainerRef}>
                 <div ref={(el) => (formSectionRefs.current.basicInfo = el)}>
                   <Card className='!rounded-2xl shadow-sm border-0 mb-6'>
                     {/* Header: Basic Info */}

web/src/components/table/tokens/TokensColumnDefs.jsx

@@ -88,7 +88,7 @@ const renderStatus = (text, record, t) => {
 };
 
 // Render group column
-const renderGroupColumn = (text, t) => {
+const renderGroupColumn = (text, record, t) => {
   if (text === 'auto') {
     return (
       <Tooltip
@@ -98,8 +98,8 @@ const renderGroupColumn = (text, t) => {
         position='top'
       >
         <Tag color='white' shape='circle'>
-          {' '}
-          {t('智能熔断')}{' '}
+          {t('智能熔断')}
+          {record && record.cross_group_retry ? `(${t('跨分组')})` : ''}
         </Tag>
       </Tooltip>
     );
@@ -455,7 +455,7 @@ export const getTokensColumns = ({
       title: t('分组'),
       dataIndex: 'group',
       key: 'group',
-      render: (text) => renderGroupColumn(text, t),
+      render: (text, record) => renderGroupColumn(text, record, t),
     },
     {
       title: t('密钥'),

web/src/components/table/tokens/modals/EditTokenModal.jsx

@@ -73,6 +73,7 @@ const EditTokenModal = (props) => {
     model_limits: [],
     allow_ips: '',
     group: '',
+    cross_group_retry: false,
     tokenCount: 1,
   });
 
@@ -377,6 +378,16 @@ const EditTokenModal = (props) => {
                       />
                     )}
                   </Col>
+                  <Col span={24} style={{ display: values.group === 'auto' ? 'block' : 'none' }}>
+                    <Form.Switch
+                      field='cross_group_retry'
+                      label={t('跨分组重试')}
+                      size='default'
+                      extraText={t(
+                        '开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道',
+                      )}
+                    />
+                  </Col>
                   <Col xs={24} sm={24} md={24} lg={10} xl={10}>
                     <Form.DatePicker
                       field='expired_time'
@@ -499,7 +510,7 @@ const EditTokenModal = (props) => {
                     <Form.Switch
                       field='unlimited_quota'
                       label={t('无限额度')}
-                      size='large'
+                      size='default'
                       extraText={t(
                         '令牌的额度仅用于限制令牌本身的最大额度使用量，实际的使用受到账户的剩余额度限制',
                       )}
@@ -546,11 +557,11 @@ const EditTokenModal = (props) => {
                   <Col span={24}>
                     <Form.TextArea
                       field='allow_ips'
-                      label={t('IP白名单')}
+                      label={t('IP白名单（支持CIDR表达式）')}
                       placeholder={t('允许的IP，一行一个，不填写则不限制')}
                       autosize
                       rows={1}
-                      extraText={t('请勿过度信任此功能，IP可能被伪造')}
+                      extraText={t('请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用')}
                       showClear
                       style={{ width: '100%' }}
                     />

web/src/hooks/channels/useChannelsData.jsx

@@ -26,6 +26,7 @@ import {
   showSuccess,
   loadChannelModels,
   copy,
+  toBoolean,
 } from '../../helpers';
 import {
   CHANNEL_OPTIONS,
@@ -85,6 +86,26 @@ export const useChannelsData = () => {
   const [isBatchTesting, setIsBatchTesting] = useState(false);
   const [modelTablePage, setModelTablePage] = useState(1);
   const [selectedEndpointType, setSelectedEndpointType] = useState('');
+  const [globalPassThroughEnabled, setGlobalPassThroughEnabled] =
+    useState(false);
+
+  const fetchGlobalPassThroughEnabled = async () => {
+    try {
+      const res = await API.get('/api/option/');
+      const { success, data } = res?.data || {};
+      if (!success || !Array.isArray(data)) {
+        return;
+      }
+      const option = data.find(
+        (item) => item?.key === 'global.pass_through_request_enabled',
+      );
+      if (option) {
+        setGlobalPassThroughEnabled(toBoolean(option.value));
+      }
+    } catch (error) {
+      setGlobalPassThroughEnabled(false);
+    }
+  };
 
   // 使用 ref 来避免闭包问题，类似旧版实现
   const shouldStopBatchTestingRef = useRef(false);
@@ -140,6 +161,7 @@ export const useChannelsData = () => {
       });
     fetchGroups().then();
     loadChannelModels().then();
+    fetchGlobalPassThroughEnabled().then();
   }, []);
 
   // Column visibility management
@@ -1026,6 +1048,7 @@ export const useChannelsData = () => {
     enableBatchDelete,
     statusFilter,
     compactMode,
+    globalPassThroughEnabled,
 
     // UI states
     showEdit,

web/src/i18n/i18n.js

@@ -42,6 +42,7 @@ i18n
       vi: viTranslation,
     },
     fallbackLng: 'zh',
+    nsSeparator: false,
     interpolation: {
       escapeValue: false,
     },

web/src/i18n/locales/en.json

@@ -97,7 +97,7 @@
     "Homepage URL 填": "Fill in the Homepage URL",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "IP whitelist",
+    "IP白名单（支持CIDR表达式）": "IP whitelist (supports CIDR expressions)",
     "IP限制": "IP restrictions",
     "IP黑名单": "IP blacklist",
     "JSON": "JSON",
@@ -153,6 +153,7 @@
     "URL链接": "URL Link",
     "USD (美元)": "USD (US Dollar)",
     "User Info Endpoint": "User Info Endpoint",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI does not support the functionResponse.id field. When enabled, this field will be automatically removed",
     "Webhook 密钥": "Webhook Secret",
     "Webhook 签名密钥": "Webhook Signature Key",
     "Webhook地址": "Webhook URL",
@@ -839,6 +840,9 @@
     "开启后，对免费模型（倍率为0，或者价格为0）的模型也会预消耗额度": "After enabling, free models (ratio 0 or price 0) will also pre-consume quota",
     "开启后，将定期发送ping数据保持连接活跃": "After enabling, ping data will be sent periodically to keep the connection active",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "When enabled, all requests will be directly forwarded to the upstream without any processing (redirects and channel adaptation will also be disabled). Please enable with caution.",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Request pass-through is enabled for this channel. Built-in NewAPI features such as parameter overrides, model redirection, and channel adaptation will be disabled. This is not a best practice. If this causes issues, please do not submit an issue.",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Global request pass-through is enabled. Built-in NewAPI features such as parameter overrides, model redirection, and channel adaptation will be disabled. This is not a best practice. If this causes issues, please do not submit an issue.",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "Request pass-through is enabled for this channel; built-in NewAPI features such as parameter overrides and model redirection will be disabled. This is not a best practice.",
     "开启后不限制：必须设置模型倍率": "After enabling, no limit: must set model ratio",
     "开启后未登录用户无法访问模型广场": "When enabled, unauthenticated users cannot access the model marketplace",
     "开启批量操作": "Enable batch selection",
@@ -1510,6 +1514,7 @@
     "私有IP访问详细说明": "⚠️ Security Warning: Enabling this allows access to internal network resources (localhost, private networks). Only enable if you need to access internal services and understand the security implications.",
     "私有部署地址": "Private Deployment Address",
     "秒": "Second",
+    "移除 functionResponse.id 字段": "Remove functionResponse.id Field",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Removal of One API copyright mark must first be authorized. Project maintenance requires a lot of effort. If this project is meaningful to you, please actively support it.",
     "窗口处理": "window handling",
     "窗口等待": "window wait",
@@ -1752,7 +1757,7 @@
     "请先阅读并同意用户协议和隐私政策": "Please read and agree to the user agreement and privacy policy first",
     "请再次输入新密码": "Please enter the new password again",
     "请前往个人设置 → 安全设置进行配置。": "Please go to Personal Settings → Security Settings to configure.",
-    "请勿过度信任此功能，IP可能被伪造": "Do not over-trust this feature, IP can be spoofed",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Do not over-trust this feature, IP can be spoofed, please use it in conjunction with gateways such as nginx and CDN",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Please edit Group ratios in system settings to add new groups:",
     "请填写完整的产品信息": "Please fill in complete product information",
     "请填写完整的管理员账号信息": "Please fill in the complete administrator account information",
@@ -2177,6 +2182,9 @@
     "默认区域，如: us-central1": "Default region, e.g.: us-central1",
     "默认折叠侧边栏": "Default collapse sidebar",
     "默认测试模型": "Default Test Model",
-    "默认补全倍率": "Default completion ratio"
+    "默认补全倍率": "Default completion ratio",
+    "跨分组重试": "Cross-group retry",
+    "跨分组": "Cross-group",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "After enabling, when the current group channel fails, it will try the next group's channel in order"
   }
 }

web/src/i18n/locales/fr.json

@@ -47,12 +47,12 @@
     "API Key 模式下不支持批量创建": "Création en lot non prise en charge en mode clé API",
     "API 地址和相关配置": "URL de l'API et configuration associée",
     "API 密钥": "Clé API",
-    "API 文档": "Documentation de l'API",
-    "API 配置": "Configuration de l'API",
-    "API令牌管理": "Gestion des jetons d'API",
-    "API使用记录": "Enregistrements d'utilisation de l'API",
+    "API 文档": "Docs API",
+    "API 配置": "Config. API",
+    "API令牌管理": "Jetons API",
+    "API使用记录": "Journaux d'API",
     "API信息": "Informations sur l'API",
-    "API信息管理，可以配置多个API地址用于状态展示和负载均衡（最多50个）": "Gestion des informations de l'API, vous pouvez configurer plusieurs adresses d'API pour l'affichage de l'état et l'équilibrage de charge (maximum 50)",
+    "API信息管理，可以配置多个API地址用于状态展示和负载均衡（最多50个）": "Infos API, vous pouvez configurer plusieurs adresses d'API pour l'affichage de l'état et l'équilibrage de charge (maximum 50)",
     "API地址": "URL de base",
     "API渠道配置": "Configuration du canal de l'API",
     "API端点": "Points de terminaison de l'API",
@@ -99,7 +99,7 @@
     "Homepage URL 填": "Remplir l'URL de la page d'accueil",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "Liste blanche d'adresses IP",
+    "IP白名单（支持CIDR表达式）": "Liste blanche d'adresses IP (prise en charge des expressions CIDR)",
     "IP限制": "Restrictions d'IP",
     "IP黑名单": "Liste noire d'adresses IP",
     "JSON": "JSON",
@@ -112,7 +112,7 @@
     "LinuxDO": "LinuxDO",
     "LinuxDO ID": "ID LinuxDO",
     "Logo 图片地址": "Adresse de l'image du logo",
-    "Midjourney 任务记录": "Enregistrements de tâches Midjourney",
+    "Midjourney 任务记录": "Tâches Midjourney",
     "MIT许可证": "Licence MIT",
     "New API项目仓库地址：": "Adresse du référentiel du projet New API : ",
     "OIDC": "OIDC",
@@ -136,7 +136,7 @@
     "SMTP 访问凭证": "Informations d'identification d'accès SMTP",
     "SMTP 账户": "Compte SMTP",
     "SSRF防护开关详细说明": "L'interrupteur principal contrôle si la protection SSRF est activée. Lorsqu'elle est désactivée, toutes les vérifications SSRF sont contournées, autorisant l'accès à n'importe quelle URL. ⚠️ Ne désactivez cette fonctionnalité que dans des environnements entièrement fiables.",
-    "SSRF防护设置": "Paramètres de protection SSRF",
+    "SSRF防护设置": "Protection SSRF",
     "SSRF防护详细说明": "La protection SSRF empêche les utilisateurs malveillants d'utiliser votre serveur pour accéder aux ressources du réseau interne. Configurez des listes blanches pour les domaines/IP de confiance et limitez les ports autorisés. S'applique aux téléchargements de fichiers, aux webhooks et aux notifications.",
     "store 字段用于授权 OpenAI 存储请求数据以评估和优化产品。默认关闭，开启后可能导致 Codex 无法正常使用": "Le champ store autorise OpenAI à stocker les données de requête pour l'évaluation et l'optimisation du produit. Désactivé par défaut. L'activation peut causer un dysfonctionnement de Codex",
     "Stripe 设置": "Paramètres Stripe",
@@ -150,10 +150,11 @@
     "Turnstile Site Key": "Clé du site Turnstile",
     "Unix时间戳": "Horodatage Unix",
     "Uptime Kuma地址": "Adresse Uptime Kuma",
-    "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Gestion des catégories de surveillance Uptime Kuma, vous pouvez configurer plusieurs catégories de surveillance pour l'affichage de l'état du service (maximum 20)",
+    "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Catégories de surveillance Uptime Kuma, vous pouvez configurer plusieurs catégories de surveillance pour l'affichage de l'état du service (maximum 20)",
     "URL链接": "Lien URL",
     "USD (美元)": "USD (Dollar US)",
     "User Info Endpoint": "Point de terminaison des informations utilisateur",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI ne prend pas en charge le champ functionResponse.id. Lorsqu'il est activé, ce champ sera automatiquement supprimé",
     "Webhook 密钥": "Clé Webhook",
     "Webhook 签名密钥": "Clé de signature Webhook",
     "Webhook地址": "URL du Webhook",
@@ -203,9 +204,9 @@
     "个": " individuel",
     "个人中心": "Centre personnel",
     "个人中心区域": "Zone du centre personnel",
-    "个人信息设置": "Paramètres des informations personnelles",
-    "个人设置": "Paramètres personnels",
-    "个性化设置": "Paramètres de personnalisation",
+    "个人信息设置": "Infos personnelles",
+    "个人设置": "Profil",
+    "个性化设置": "Personnalisation",
     "个性化设置左侧边栏的显示内容": "Personnaliser le contenu affiché dans la barre latérale gauche",
     "个未配置模型": "modèles non configurés",
     "个模型": "modèles",
@@ -263,26 +264,26 @@
     "令牌已重置并已复制到剪贴板": "Le jeton a été réinitialisé et copié dans le presse-papiers",
     "令牌更新成功！": "Jeton mis à jour avec succès !",
     "令牌的额度仅用于限制令牌本身的最大额度使用量，实际的使用受到账户的剩余额度限制": "Le quota du jeton est uniquement utilisé pour limiter l'utilisation maximale du quota du jeton lui-même, et l'utilisation réelle est limitée par le quota restant du compte",
-    "令牌管理": "Gestion des jetons",
+    "令牌管理": "Jetons",
     "以下上游数据可能不可信：": "Les données en amont suivantes peuvent ne pas être fiables : ",
     "以下文件解析失败，已忽略：{{list}}": "L'analyse des fichiers suivants a échoué, ignorés : {{list}}",
     "以及": "et",
-    "仪表盘设置": "Paramètres du tableau de bord",
+    "仪表盘设置": "Tableau de bord",
     "价格": "Tarifs",
     "价格：${{price}} * {{ratioType}}：{{ratio}}": "Prix : ${{price}} * {{ratioType}} : {{ratio}}",
-    "价格设置": "Paramètres de prix",
+    "价格设置": "Prix",
     "价格设置方式": "Méthode de configuration des prix",
     "任务 ID": "ID de la tâche",
     "任务ID": "ID de la tâche",
-    "任务日志": "Journaux de tâches",
+    "任务日志": "Tâches",
     "任务状态": "Statut de la tâche",
-    "任务记录": "Enregistrements de tâches",
+    "任务记录": "Tâches",
     "企业账户为特殊返回格式，需要特殊处理，如果非企业账户，请勿勾选": "Les comptes d'entreprise ont un format de retour spécial et nécessitent un traitement particulier. Si ce n'est pas un compte d'entreprise, veuillez ne pas cocher cette case.",
     "优先级": "Priorité",
     "优惠": "Remise",
     "低于此额度时将发送邮件提醒用户": "Un rappel par e-mail sera envoyé lorsque le quota tombera en dessous de ce seuil",
     "余额": "Solde",
-    "余额充值管理": "Gestion de la recharge du solde",
+    "余额充值管理": "Recharge du solde",
     "你似乎并没有修改什么": "Vous ne semblez rien avoir modifié",
     "你可以在“自定义模型名称”处手动添加它们，然后点击填入后再提交，或者直接使用下方操作自动处理。": "Vous pouvez les ajouter manuellement dans « Noms de modèles personnalisés », cliquer sur Remplir puis soumettre, ou utiliser directement les actions ci-dessous pour les traiter automatiquement.",
     "使用 Discord 继续": "Continuer avec Discord",
@@ -297,7 +298,7 @@
     "使用 用户名 注册": "S'inscrire avec un nom d'utilisateur",
     "使用 邮箱或用户名 登录": "Connectez-vous avec votre e-mail ou votre nom d'utilisateur",
     "使用ID排序": "Trier par ID",
-    "使用日志": "Journaux d'utilisation",
+    "使用日志": "Journaux",
     "使用模式": "Mode d'utilisation",
     "使用统计": "Statistiques d'utilisation",
     "使用认证器应用（如 Google Authenticator、Microsoft Authenticator）扫描下方二维码：": "Utilisez une application d'authentification (telle que Google Authenticator, Microsoft Authenticator) pour scanner le code QR ci-dessous :",
@@ -327,7 +328,7 @@
     "供应商名称": "Nom du fournisseur",
     "供应商图标": "Icône du fournisseur",
     "供应商更新成功！": "Fournisseur mis à jour avec succès !",
-    "侧边栏管理（全局控制）": "Gestion de la barre latérale (contrôle global)",
+    "侧边栏管理（全局控制）": "Barre latérale (Global)",
     "侧边栏设置保存成功": "Paramètres de la barre latérale enregistrés avec succès",
     "保存": "Enregistrer",
     "保存 Discord OAuth 设置": "Enregistrer les paramètres OAuth Discord",
@@ -401,7 +402,7 @@
     "充值数量": "Quantité de recharge",
     "充值数量，最低 ": "Quantité de recharge, minimum ",
     "充值数量不能小于": "Le montant de la recharge ne peut pas être inférieur à",
-    "充值方式设置": "Paramètres de la méthode de recharge",
+    "充值方式设置": "Méthodes recharge",
     "充值方式设置不是合法的 JSON 字符串": "Les paramètres de la méthode de recharge ne sont pas une chaîne JSON valide",
     "充值确认": "Confirmation de la recharge",
     "充值账单": "Factures de recharge",
@@ -417,8 +418,8 @@
     "兑换码创建成功！": "Code d'échange créé avec succès !",
     "兑换码将以文本文件的形式下载，文件名为兑换码的名称。": "Le code d'échange sera téléchargé sous forme de fichier texte, le nom de fichier étant le nom du code d'échange.",
     "兑换码更新成功！": "Code d'échange mis à jour avec succès !",
-    "兑换码生成管理": "Gestion de la génération de codes d'échange",
-    "兑换码管理": "Gestion des codes d'échange",
+    "兑换码生成管理": "Génération de codes",
+    "兑换码管理": "Codes d'échange",
     "兑换额度": "Utiliser",
     "全局控制侧边栏区域和功能显示，管理员隐藏的功能用户无法启用": "Contrôle global des zones et des fonctions de la barre latérale, les utilisateurs ne peuvent pas activer les fonctions masquées par les administrateurs",
     "全局设置": "Paramètres globaux",
@@ -447,7 +448,7 @@
     "共 {{total}} 项，当前显示 {{start}}-{{end}} 项": "Total {{total}} éléments, affichage actuel {{start}}-{{end}} éléments",
     "关": "Fermer",
     "关于": "À propos",
-    "关于我们": "À propos de nous",
+    "关于我们": "Nous",
     "关于系统的详细信息": "Informations détaillées sur le système",
     "关于项目": "À propos du projet",
     "关键字(id或者名称)": "Mot-clé (id ou nom)",
@@ -459,7 +460,7 @@
     "其他": "Autre",
     "其他注册选项": "Autres options d'inscription",
     "其他登录选项": "Autres options de connexion",
-    "其他设置": "Autres paramètres",
+    "其他设置": "Autres",
     "其他详情": "Autres détails",
     "内容": "Contenu",
     "内容较大，已启用性能优化模式": "Le contenu est volumineux, le mode d'optimisation des performances a été activé",
@@ -471,14 +472,14 @@
     "准备完成初始化": "Prêt à terminer l'initialisation",
     "分类名称": "Nom de la catégorie",
     "分组": "Groupe",
-    "分组与模型定价设置": "Paramètres de groupe et de tarification du modèle",
+    "分组与模型定价设置": "Groupe et tarification",
     "分组价格": "Prix de groupe",
     "分组倍率": "Ratio",
-    "分组倍率设置": "Paramètres de ratio de groupe",
+    "分组倍率设置": "Ratio de groupe",
     "分组倍率设置，可以在此处新增分组或修改现有分组的倍率，格式为 JSON 字符串，例如：{\"vip\": 0.5, \"test\": 1}，表示 vip 分组的倍率为 0.5，test 分组的倍率为 1": "Paramètres de ratio de groupe, vous pouvez ajouter de nouveaux groupes ou modifier le ratio des groupes existants ici, au format de chaîne JSON, par exemple : {\"vip\": 0,5, \"test\": 1}, ce qui signifie que le ratio du groupe vip est 0,5 et celui du groupe test est 1",
     "分组特殊倍率": "Ratio spécial de groupe",
     "分组特殊可用分组": "Groupes spéciaux disponibles",
-    "分组设置": "Paramètres de groupe",
+    "分组设置": "Groupe",
     "分组速率配置优先级高于全局速率限制。": "La priorité de configuration du taux de groupe est supérieure à la limite de taux globale.",
     "分组速率限制": "Limitation du taux de groupe",
     "分钟": "minutes",
@@ -491,7 +492,7 @@
     "划转金额最低为": "Le montant minimum du virement est de",
     "划转额度": "Montant du virement",
     "列出的模型将不会自动添加或移除-thinking/-nothinking 后缀": "Les modèles listés ici n'ajouteront ni ne retireront automatiquement le suffixe -thinking/-nothinking.",
-    "列设置": "Paramètres de colonne",
+    "列设置": "Colonnes",
     "创建令牌默认选择auto分组，初始令牌也将设为auto（否则留空，为用户默认分组）": "Lors de la création d'un jeton, le groupe auto est sélectionné par défaut, et le jeton initial sera également défini sur auto (sinon laisser vide, pour le groupe par défaut de l'utilisateur)",
     "创建失败": "Échec de la création",
     "创建成功": "Création réussie",
@@ -570,7 +571,7 @@
     "可用端点类型": "Types de points de terminaison pris en charge",
     "可用邀请额度": "Quota d'invitation disponible",
     "可视化": "Visualisation",
-    "可视化倍率设置": "Paramètres de ratio de modèle visuel",
+    "可视化倍率设置": "Ratio visuel",
     "可视化编辑": "Édition visuelle",
     "可选，公告的补充说明": "Facultatif, informations supplémentaires pour l'avis",
     "可选值": "Valeur facultative",
@@ -696,7 +697,7 @@
     "字段透传控制": "Contrôle du passage des champs",
     "存在重复的键名：": "Il existe des noms de clés en double :",
     "安全提醒": "Rappel de sécurité",
-    "安全设置": "Paramètres de sécurité",
+    "安全设置": "Sécurité",
     "安全验证": "Vérification de sécurité",
     "安全验证级别": "Niveau de vérification de la sécurité",
     "安装指南": "Guide d'installation",
@@ -719,7 +720,7 @@
     "密码修改成功！": "Mot de passe changé avec succès !",
     "密码已复制到剪贴板：": "Le mot de passe a été copié dans le presse-papiers : ",
     "密码已重置并已复制到剪贴板：": "Le mot de passe a été réinitialisé et copié dans le presse-papiers : ",
-    "密码管理": "Gestion des mots de passe",
+    "密码管理": "Mots de passe",
     "密码重置": "Réinitialisation du mot de passe",
     "密码重置完成": "Réinitialisation du mot de passe terminée",
     "密码重置确认": "Confirmation de la réinitialisation du mot de passe",
@@ -761,8 +762,8 @@
     "小时": "Heure",
     "尚未使用": "Pas encore utilisé",
     "局部重绘-提交": "Varier la région",
-    "屏蔽词列表": "Liste des mots sensibles",
-    "屏蔽词过滤设置": "Paramètres de filtrage des mots sensibles",
+    "屏蔽词列表": "Mots sensibles",
+    "屏蔽词过滤设置": "Filtrage mots sensibles",
     "展开": "Développer",
     "展开更多": "Développer plus",
     "展示价格": "Prix affiché",
@@ -847,6 +848,9 @@
     "开启后，对免费模型（倍率为0，或者价格为0）的模型也会预消耗额度": "Après activation, les modèles gratuits (ratio 0 ou prix 0) préconsommeront également du quota",
     "开启后，将定期发送ping数据保持连接活跃": "Après activation, des données ping seront envoyées périodiquement pour maintenir la connexion active",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "Après activation, toutes les requêtes seront directement transmises en amont sans aucun traitement (la redirection et l'adaptation de canal seront également désactivées), veuillez activer avec prudence",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "La transmission des requêtes est activée pour ce canal. Les fonctionnalités intégrées de NewAPI (surcharge des paramètres, redirection de modèle, adaptation du canal, etc.) seront désactivées. Ce n'est pas une bonne pratique. Si cela cause des problèmes, merci de ne pas ouvrir d'issue.",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "La transmission globale des requêtes est activée. Les fonctionnalités intégrées de NewAPI (surcharge des paramètres, redirection de modèle, adaptation du canal, etc.) seront désactivées. Ce n'est pas une bonne pratique. Si cela cause des problèmes, merci de ne pas ouvrir d'issue.",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "La transmission des requêtes est activée pour ce canal ; les fonctionnalités intégrées de NewAPI (comme la surcharge des paramètres et la redirection de modèle) seront désactivées. Ce n'est pas une bonne pratique.",
     "开启后不限制：必须设置模型倍率": "Après l'activation, aucune limite : le ratio de modèle doit être défini",
     "开启后未登录用户无法访问模型广场": "Lorsqu'il est activé, les utilisateurs non authentifiés ne peuvent pas accéder à la place du marché des modèles",
     "开启批量操作": "Activer la sélection par lots",
@@ -997,7 +1001,7 @@
     "支付地址": "Adresse de paiement",
     "支付宝": "Alipay",
     "支付方式": "Mode de paiement",
-    "支付设置": "Paramètres de paiement",
+    "支付设置": "Paiement",
     "支付请求失败": "Échec de la demande de paiement",
     "支付金额": "Montant payé",
     "支持6位TOTP验证码或8位备用码，可到`个人设置-安全设置-两步验证设置`配置或查看。": "Prend en charge le code de vérification TOTP à 6 chiffres ou le code de sauvegarde à 8 chiffres, peut être configuré ou consulté dans `Paramètres personnels - Paramètres de sécurité - Paramètres d'authentification à deux facteurs`.",
@@ -1027,9 +1031,9 @@
     "数据格式错误": "Erreur de format de données",
     "数据看板": "Tableau de bord",
     "数据看板更新间隔": "Intervalle de mise à jour du tableau de bord des données",
-    "数据看板设置": "Paramètres du tableau de bord des données",
+    "数据看板设置": "Tableau de bord",
     "数据看板默认时间粒度": "Granularité temporelle par défaut du tableau de bord des données",
-    "数据管理和日志查看": "Gestion des données et affichage des journaux",
+    "数据管理和日志查看": "Données et journaux",
     "文件上传": "Téléchargement de fichier",
     "文件搜索价格：{{symbol}}{{price}} / 1K 次": "Prix de recherche de fichier : {{symbol}}{{price}} / 1K fois",
     "文字提示 {{input}} tokens / 1M tokens * {{symbol}}{{price}} + 文字补全 {{completion}} tokens / 1M tokens * {{symbol}}{{compPrice}} = {{symbol}}{{total}}": "Invite texte {{input}} tokens / 1M tokens * {{symbol}}{{price}} + Complétion texte {{completion}} tokens / 1M tokens * {{symbol}}{{compPrice}} = {{symbol}}{{total}}",
@@ -1065,7 +1069,7 @@
     "无限额度": "Quota illimité",
     "日志清理失败：": "Échec du nettoyage des journaux :",
     "日志类型": "Type de journal",
-    "日志设置": "Paramètres du journal",
+    "日志设置": "Config. journaux",
     "日志详情": "Détails du journal",
     "旧格式（直接覆盖）：": "Ancien format (remplacement direct) :",
     "旧格式模板": "Modèle d'ancien format",
@@ -1219,7 +1223,7 @@
     "模型倍率值": "Valeur du ratio de modèle",
     "模型倍率和补全倍率": "Ratio de modèle et ratio de complétion",
     "模型倍率和补全倍率同时设置": "Le ratio de modèle et le ratio de complétion sont définis simultanément",
-    "模型倍率设置": "Paramètres de ratio de modèle",
+    "模型倍率设置": "Ratio modèle",
     "模型关键字": "mot-clé du modèle",
     "模型列表已复制到剪贴板": "Liste des modèles copiée dans le presse-papiers",
     "模型列表已更新": "La liste des modèles a été mise à jour",
@@ -1229,7 +1233,7 @@
     "模型固定价格": "Prix du modèle par appel",
     "模型图标": "Icône du modèle",
     "模型定价，需要登录访问": "Tarification du modèle, nécessite une connexion pour y accéder",
-    "模型广场": "Place du marché des modèles",
+    "模型广场": "Marché des modèles",
     "模型支持的接口端点信息": "Informations sur les points de terminaison de l'API pris en charge par le modèle",
     "模型数据分析": "Analyse des données du modèle",
     "模型映射必须是合法的 JSON 格式！": "Le mappage de modèles doit être au format JSON valide !",
@@ -1241,7 +1245,7 @@
     "模型的详细描述和基本特性": "Description détaillée et caractéristiques de base du modèle",
     "模型相关设置": "Paramètres liés au modèle",
     "模型社区需要大家的共同维护，如发现数据有误或想贡献新的模型数据，请访问：": "La communauté des modèles a besoin de la contribution de tous. Si vous trouvez des données incorrectes ou si vous souhaitez contribuer à de nouvelles données de modèle, veuillez visiter :",
-    "模型管理": "Gestion des modèles",
+    "模型管理": "Modèles",
     "模型组": "Groupe de modèles",
     "模型补全倍率（仅对自定义模型有效）": "Ratio d'achèvement de modèle (uniquement efficace pour les modèles personnalisés)",
     "模型请求速率限制": "Limite de débit de requête de modèle",
@@ -1367,7 +1371,7 @@
     "渠道的基本配置信息": "Informations de configuration de base du canal",
     "渠道的模型测试": "Test de modèle de canal",
     "渠道的高级配置选项": "Options de configuration avancées du canal",
-    "渠道管理": "Gestion des canaux",
+    "渠道管理": "Canaux",
     "渠道额外设置": "Paramètres supplémentaires du canal",
     "源地址": "Adresse source",
     "演示站点": "Site de démonstration",
@@ -1410,7 +1414,7 @@
     "用户信息": "Informations utilisateur",
     "用户信息更新成功！": "Informations utilisateur mises à jour avec succès !",
     "用户分组": "Votre groupe par défaut",
-    "用户分组和额度管理": "Gestion des groupes d'utilisateurs et des quotas",
+    "用户分组和额度管理": "Groupes et quotas",
     "用户分组配置": "Configuration du groupe d'utilisateurs",
     "用户协议": "Accord utilisateur",
     "用户协议已更新": "L'accord utilisateur a été mis à jour",
@@ -1425,10 +1429,10 @@
     "用户每周期最多请求次数": "Nombre maximal de requêtes utilisateur par période",
     "用户注册时看到的网站名称，比如'我的网站'": "Nom du site Web que les utilisateurs voient lors de l'inscription, par exemple 'Mon site Web'",
     "用户的基本账户信息": "Informations de base du compte utilisateur",
-    "用户管理": "Gestion des utilisateurs",
+    "用户管理": "Utilisateurs",
     "用户组": "Groupe d'utilisateurs",
     "用户账户创建成功！": "Compte utilisateur créé avec succès !",
-    "用户账户管理": "Gestion des comptes utilisateurs",
+    "用户账户管理": "Comptes utilisateurs",
     "用时/首字": "Temps/premier mot",
     "留空则使用账号绑定的邮箱": "Si ce champ est laissé vide, l'adresse e-mail liée au compte sera utilisée",
     "留空则使用默认端点；支持 {path, method}": "Laissez vide pour utiliser le point de terminaison par défaut ; prend en charge {path, method}",
@@ -1439,7 +1443,7 @@
     "登录过期，请重新登录！": "Session expirée, veuillez vous reconnecter !",
     "白名单": "Liste blanche",
     "的前提下使用。": "doit être utilisé conformément aux conditions.",
-    "监控设置": "Paramètres de surveillance",
+    "监控设置": "Surveillance",
     "目标用户：{{username}}": "Utilisateur cible : {{username}}",
     "直接提交": "Soumettre directement",
     "相关项目": "Projets connexes",
@@ -1520,6 +1524,7 @@
     "私有IP访问详细说明": "⚠️ Avertissement de sécurité : l'activation de cette option autorise l'accès aux ressources du réseau interne (localhost, réseaux privés). N'activez cette option que si vous devez accéder à des services internes et que vous comprenez les implications en matière de sécurité.",
     "私有部署地址": "Adresse de déploiement privée",
     "秒": "Seconde",
+    "移除 functionResponse.id 字段": "Supprimer le champ functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "La suppression de la marque de copyright de One API doit d'abord être autorisée. La maintenance du projet demande beaucoup d'efforts. Si ce projet a du sens pour vous, veuillez le soutenir activement.",
     "窗口处理": "gestion des fenêtres",
     "窗口等待": "attente de la fenêtre",
@@ -1552,14 +1557,14 @@
     "精确": "Exact",
     "系统": "Système",
     "系统令牌已复制到剪切板": "Le jeton système a été copié dans le presse-papiers",
-    "系统任务记录": "Enregistrements de tâches système",
+    "系统任务记录": "Tâches système",
     "系统信息": "Informations système",
     "系统公告": "Avis système",
-    "系统公告管理，可以发布系统通知和重要消息（最多100个，前端显示最新20条）": "Gestion des avis système, vous pouvez publier des avis système et des messages importants (maximum 100, afficher les 20 derniers sur le front-end)",
+    "系统公告管理，可以发布系统通知和重要消息（最多100个，前端显示最新20条）": "Avis système, vous pouvez publier des avis système et des messages importants (maximum 100, afficher les 20 derniers sur le front-end)",
     "系统初始化": "Initialisation du système",
     "系统初始化失败，请重试": "L'initialisation du système a échoué, veuillez réessayer",
     "系统初始化成功，正在跳转...": "Initialisation du système réussie, redirection en cours...",
-    "系统参数配置": "Configuration des paramètres système",
+    "系统参数配置": "Paramètres système",
     "系统名称": "Nom du système",
     "系统名称已更新": "Nom du système mis à jour",
     "系统名称更新失败": "Échec de la mise à jour du nom du système",
@@ -1570,7 +1575,7 @@
     "系统文档和帮助信息": "Documentation système et informations d'aide",
     "系统消息": "Messages système",
     "系统管理功能": "Fonctions de gestion du système",
-    "系统设置": "Paramètres système",
+    "系统设置": "Système",
     "系统访问令牌": "Jeton d'accès au système",
     "约": "Environ",
     "索引": "Index",
@@ -1589,9 +1594,9 @@
     "结束时间": "Heure de fin",
     "结果图片": "Résultat",
     "绘图": "Dessin",
-    "绘图任务记录": "Enregistrements de tâches de dessin",
-    "绘图日志": "Journaux de dessin",
-    "绘图设置": "Paramètres de dessin",
+    "绘图任务记录": "Tâches dessin",
+    "绘图日志": "Dessins",
+    "绘图设置": "Dessin",
     "统一的": "La Passerelle",
     "统计Tokens": "Jetons statistiques",
     "统计次数": "Nombre de statistiques",
@@ -1638,11 +1643,11 @@
     "置信度": "Confiance",
     "美元": "Dollar américain",
     "聊天": "Discuter",
-    "聊天会话管理": "Gestion des sessions de discussion",
+    "聊天会话管理": "Sessions de discussion",
     "聊天区域": "Zone de discussion",
     "聊天应用名称": "Nom de l'application de discussion",
     "聊天应用名称已存在，请使用其他名称": "Le nom de l'application de discussion existe déjà, veuillez utiliser un autre nom",
-    "聊天设置": "Paramètres de discussion",
+    "聊天设置": "Discussion",
     "聊天配置": "Configuration de la discussion",
     "聊天链接配置错误，请联系管理员": "Erreur de configuration du lien de discussion, veuillez contacter l'administrateur",
     "联系我们": "Contactez-nous",
@@ -1762,7 +1767,7 @@
     "请先阅读并同意用户协议和隐私政策": "Veuillez d'abord lire et accepter l'accord utilisateur et la politique de confidentialité",
     "请再次输入新密码": "Veuillez saisir à nouveau le nouveau mot de passe",
     "请前往个人设置 → 安全设置进行配置。": "Veuillez aller dans Paramètres personnels → Paramètres de sécurité pour configurer.",
-    "请勿过度信任此功能，IP可能被伪造": "Ne faites pas trop confiance à cette fonctionnalité, l'IP peut être usurpée",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Ne faites pas trop confiance à cette fonctionnalité, l'IP peut être usurpée, veuillez l'utiliser en conjonction avec des passerelles telles que nginx et cdn",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Veuillez modifier les ratios de groupe dans les paramètres système pour ajouter de nouveaux groupes :",
     "请填写完整的产品信息": "Veuillez renseigner l'ensemble des informations produit",
     "请填写完整的管理员账号信息": "Veuillez remplir les informations complètes du compte administrateur",
@@ -1986,19 +1991,19 @@
     "输出价格": "Prix de sortie",
     "输出价格：{{symbol}}{{price}} * {{completionRatio}} = {{symbol}}{{total}} / 1M tokens (补全倍率: {{completionRatio}})": "Prix de sortie : {{symbol}}{{price}} * {{completionRatio}} = {{symbol}}{{total}} / 1M tokens (ratio d'achèvement : {{completionRatio}})",
     "输出倍率 {{completionRatio}}": "Ratio de sortie {{completionRatio}}",
-    "边栏设置": "Paramètres de la barre latérale",
+    "边栏设置": "Barre latérale",
     "过期时间": "Date d'expiration",
     "过期时间不能早于当前时间！": "La date d'expiration ne peut pas être antérieure à l'heure actuelle !",
     "过期时间快捷设置": "Paramètres rapides de la date d'expiration",
     "过期时间格式错误！": "Erreur de format de la date d'expiration !",
-    "运营设置": "Paramètres de fonctionnement",
+    "运营设置": "Opérations",
     "返回修改": "Revenir pour modifier",
     "返回登录": "Retour à la connexion",
     "这是重复键中的最后一个，其值将被使用": "Ceci est la dernière clé dupliquée, sa valeur sera utilisée",
     "进度": "calendrier",
     "进行中": "En cours",
     "进行该操作时，可能导致渠道访问错误，请仅在数据库出现问题时使用": "Lors de cette opération, cela peut entraîner des erreurs d'accès au canal. Veuillez ne l'utiliser que lorsqu'il y a un problème avec la base de données.",
-    "连接保活设置": "Paramètres de maintien de connexion",
+    "连接保活设置": "Maintien connexion",
     "连接已断开": "Connexion interrompue",
     "追加到现有密钥": "Ajouter aux clés existantes",
     "追加模式：将新密钥添加到现有密钥列表末尾": "Mode d'ajout : ajouter les nouvelles clés à la fin de la liste de clés existantes",
@@ -2030,7 +2035,7 @@
     "选择过期时间（可选，留空为永久）": "Sélectionnez la date d'expiration (facultatif, laissez vide pour permanent)",
     "透传请求体": "Corps de transmission",
     "通义千问": "Qwen",
-    "通用设置": "Paramètres généraux",
+    "通用设置": "Général",
     "通知": "Avis",
     "通知、价格和隐私相关设置": "Paramètres de notification, de prix et de confidentialité",
     "通知内容": "Contenu de la notification",
@@ -2039,13 +2044,13 @@
     "通知标题": "Titre de la notification",
     "通知类型 (quota_exceed: 额度预警)": "Type de notification (quota_exceed : avertissement de quota)",
     "通知邮箱": "E-mail de notification",
-    "通知配置": "Configuration des notifications",
+    "通知配置": "Notifications",
     "通过划转功能将奖励额度转入到您的账户余额中": "Transférez le montant de la récompense sur le solde de votre compte via la fonction de virement",
     "通过密码注册时需要进行邮箱验证": "La vérification par e-mail est requise lors de l'inscription via mot de passe",
     "通道 ${name} 余额更新成功！": "Le quota du canal ${name} a été mis à jour avec succès !",
     "通道 ${name} 测试成功，模型 ${model} 耗时 ${time.toFixed(2)} 秒。": "Test du canal ${name} réussi, modèle ${model} a pris ${time.toFixed(2)} secondes.",
     "通道 ${name} 测试成功，耗时 ${time.toFixed(2)} 秒。": "Test du canal ${name} réussi, a pris ${time.toFixed(2)} secondes.",
-    "速率限制设置": "Paramètres de limitation de débit",
+    "速率限制设置": "Limitation débit",
     "邀请": "Invitations",
     "邀请人": "Inviteur",
     "邀请人数": "Nombre de personnes invitées",
@@ -2107,7 +2112,7 @@
     "重置邮件发送成功，请检查邮箱！": "L'e-mail de réinitialisation a été envoyé avec succès, veuillez vérifier votre e-mail !",
     "重置配置": "Réinitialiser la configuration",
     "重试": "Réessayer",
-    "钱包管理": "Gestion du portefeuille",
+    "钱包管理": "Portefeuille",
     "链接中的{key}将自动替换为sk-xxxx，{address}将自动替换为系统设置的服务器地址，末尾不带/和/v1": "Le {key} dans le lien sera automatiquement remplacé par sk-xxxx, le {address} sera automatiquement remplacé par l'adresse du serveur dans les paramètres système, et la fin n'aura pas / et /v1",
     "错误": "Erreur",
     "键为分组名称，值为另一个 JSON 对象，键为分组名称，值为该分组的用户的特殊分组倍率，例如：{\"vip\": {\"default\": 0.5, \"test\": 1}}，表示 vip 分组的用户在使用default分组的令牌时倍率为0.5，使用test分组时倍率为1": "La clé est le nom du groupe, la valeur est un autre objet JSON, la clé est le nom du groupe, la valeur est le ratio de groupe spécial des utilisateurs de ce groupe, par exemple : {\"vip\": {\"default\": 0.5, \"test\": 1}}, ce qui signifie que les utilisateurs du groupe vip ont un ratio de 0.5 lors de l'utilisation de jetons du groupe default et un ratio de 1 lors de l'utilisation du groupe test",
@@ -2125,7 +2130,7 @@
     "隐私政策": "Politique de confidentialité",
     "隐私政策已更新": "La politique de confidentialité a été mise à jour",
     "隐私政策更新失败": "Échec de la mise à jour de la politique de confidentialité",
-    "隐私设置": "Paramètres de confidentialité",
+    "隐私设置": "Confidentialité",
     "隐藏操作项": "Masquer les actions",
     "隐藏调试": "Masquer le débogage",
     "随机": "Aléatoire",
@@ -2146,7 +2151,7 @@
     "音频输出补全相关的倍率设置，键为模型名称，值为倍率": "Paramètres de ratio liés à l'achèvement de la sortie audio, la clé est le nom du modèle, la valeur est le ratio",
     "页脚": "Pied de page",
     "页面未找到，请检查您的浏览器地址是否正确": "Page non trouvée, veuillez vérifier si l'adresse de votre navigateur est correcte",
-    "顶栏管理": "Gestion de l'en-tête",
+    "顶栏管理": "En-tête",
     "项目": "Élément",
     "项目内容": "Contenu de l'élément",
     "项目操作按钮组": "Groupe de boutons d'action du projet",
@@ -2161,7 +2166,7 @@
     "额度必须大于0": "Le quota doit être supérieur à 0",
     "额度提醒阈值": "Seuil de rappel de quota",
     "额度查询接口返回令牌额度而非用户额度": "Affiche le quota de jetons au lieu du quota utilisateur",
-    "额度设置": "Paramètres de quota",
+    "额度设置": "Quota",
     "额度预警阈值": "Seuil d'avertissement de quota",
     "首尾生视频": "Vidéo de début et de fin",
     "首页": "Accueil",
@@ -2226,6 +2231,9 @@
     "默认助手消息": "Bonjour ! Comment puis-je vous aider aujourd'hui ?",
     "可选，用于复现结果": "Optionnel, pour des résultats reproductibles",
     "随机种子 (留空为随机)": "Graine aléatoire (laisser vide pour aléatoire)",
-    "默认补全倍率": "Taux de complétion par défaut"
+    "默认补全倍率": "Taux de complétion par défaut",
+    "跨分组重试": "Nouvelle tentative inter-groupes",
+    "跨分组": "Inter-groupes",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "Après activation, lorsque le canal du groupe actuel échoue, il essaiera le canal du groupe suivant dans l'ordre"
   }
 }

web/src/i18n/locales/ja.json

@@ -82,7 +82,7 @@
     "Homepage URL 填": "ホームページURLを入力してください",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "IPホワイトリスト",
+    "IP白名单（支持CIDR表达式）": "IPホワイトリスト（CIDR表記に対応）",
     "IP限制": "IP制限",
     "IP黑名单": "IPブラックリスト",
     "JSON": "JSON",
@@ -136,6 +136,7 @@
     "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Uptime Kumaの監視分類管理：サービスステータス表示用に、複数の監視分類を設定できます（最大20個）",
     "URL链接": "URL",
     "User Info Endpoint": "User Info Endpoint",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AIはfunctionResponse.idフィールドをサポートしていません。有効にすると、このフィールドは自動的に削除されます",
     "Webhook 签名密钥": "Webhook署名シークレット",
     "Webhook地址": "Webhook URL",
     "Webhook地址必须以https://开头": "Webhook URLは、https://で始まることが必須です",
@@ -795,6 +796,9 @@
     "开启后，仅\"消费\"和\"错误\"日志将记录您的客户端IP地址": "有効にすると、「消費」と「エラー」のログにのみ、クライアントIPアドレスが記録されます",
     "开启后，将定期发送ping数据保持连接活跃": "有効にすると、接続をアクティブに保つためにpingデータが定期的に送信されます",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "有効にすると、すべてのリクエストは直接アップストリームにパススルーされ、いかなる処理も行われません（リダイレクトとチャネルの自動調整も無効になります）。有効にする際はご注意ください",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "このチャネルではリクエストのパススルーが有効です。パラメータ上書き、モデルリダイレクト、チャネル適応などの NewAPI 内蔵機能は無効になります。ベストプラクティスではありません。これにより問題が発生しても issue を投稿しないでください。",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "全体のリクエストパススルーが有効です。パラメータ上書き、モデルリダイレクト、チャネル適応などの NewAPI 内蔵機能は無効になります。ベストプラクティスではありません。これにより問題が発生しても issue を投稿しないでください。",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "このチャネルではリクエストのパススルーが有効です。パラメータ上書きやモデルリダイレクトなどの NewAPI 内蔵機能は無効になります。ベストプラクティスではありません。",
     "开启后不限制：必须设置模型倍率": "有効化後は制限なし：モデル倍率の設定が必須",
     "开启后未登录用户无法访问模型广场": "有効にすると、ログインしていないユーザーはモデルマーケットプレイスにアクセスできなくなります",
     "开启批量操作": "一括操作を有効にする",
@@ -1440,6 +1444,7 @@
     "私有IP访问详细说明": "プライベートIPアクセスの詳細説明",
     "私有部署地址": "プライベートデプロイ先URL",
     "秒": "秒",
+    "移除 functionResponse.id 字段": "functionResponse.idフィールドを削除",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "One APIの著作権表示を削除するには、事前の許可が必要です。プロジェクトの維持には多大な労力がかかります。もしこのプロジェクトがあなたにとって有意義でしたら、積極的なご支援をお願いいたします",
     "窗口处理": "ウィンドウ処理",
     "窗口等待": "ウィンドウ待機中",
@@ -1669,7 +1674,7 @@
     "请先阅读并同意用户协议和隐私政策": "まずユーザー利用規約とプライバシーポリシーをご確認の上、同意してください",
     "请再次输入新密码": "新しいパスワードを再入力してください",
     "请前往个人设置 → 安全设置进行配置。": "アカウント設定 → セキュリティ設定 にて設定してください。",
-    "请勿过度信任此功能，IP可能被伪造": "IPは偽装される可能性があるため、この機能を過信しないでください",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "IPは偽装される可能性があるため、この機能を過信しないでください。nginxやCDNなどのゲートウェイと組み合わせて使用してください。",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "新規グループを追加するには、システム設定ページでグループ倍率を編集してください：",
     "请填写完整的管理员账号信息": "管理者アカウント情報をすべて入力してください",
     "请填写密钥": "APIキーを入力してください",
@@ -2125,6 +2130,9 @@
     "默认用户消息": "こんにちは",
     "默认助手消息": "こんにちは！何かお手伝いできることはありますか？",
     "可选，用于复现结果": "オプション、結果の再現用",
-    "随机种子 (留空为随机)": "ランダムシード（空欄でランダム）"
+    "随机种子 (留空为随机)": "ランダムシード（空欄でランダム）",
+    "跨分组重试": "グループ間リトライ",
+    "跨分组": "グループ間",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "有効にすると、現在のグループチャネルが失敗した場合、次のグループのチャネルを順番に試行します"
   }
 }

web/src/i18n/locales/ru.json

@@ -101,7 +101,7 @@
     "Homepage URL 填": "URL домашней страницы:",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "Белый список IP",
+    "IP白名单（支持CIDR表达式）": "Белый список IP (поддерживает выражения CIDR)",
     "IP限制": "Ограничения IP",
     "IP黑名单": "Черный список IP",
     "JSON": "JSON",
@@ -156,6 +156,7 @@
     "URL链接": "URL ссылка",
     "USD (美元)": "USD (доллар США)",
     "User Info Endpoint": "Конечная точка информации о пользователе",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI не поддерживает поле functionResponse.id. При включении это поле будет автоматически удалено",
     "Webhook 密钥": "Секрет вебхука",
     "Webhook 签名密钥": "Ключ подписи Webhook",
     "Webhook地址": "Адрес Webhook",
@@ -856,6 +857,9 @@
     "开启后，对免费模型（倍率为0，或者价格为0）的模型也会预消耗额度": "После включения бесплатные модели (коэффициент 0 или цена 0) тоже будут предварительно расходовать квоту",
     "开启后，将定期发送ping数据保持连接活跃": "После включения будет периодически отправляться ping-данные для поддержания активности соединения",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "После включения все запросы будут напрямую передаваться upstream без какой-либо обработки (перенаправление и адаптация каналов также будут отключены), включайте с осторожностью",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Для этого канала включена сквозная передача запросов. Встроенные возможности NewAPI, такие как переопределение параметров, перенаправление моделей и адаптация канала, будут отключены. Это не является лучшей практикой. Если из-за этого возникнут проблемы, пожалуйста, не создавайте issue.",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Глобальная сквозная передача запросов включена. Встроенные возможности NewAPI, такие как переопределение параметров, перенаправление моделей и адаптация канала, будут отключены. Это не является лучшей практикой. Если из-за этого возникнут проблемы, пожалуйста, не создавайте issue.",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "Для этого канала включена сквозная передача запросов; встроенные функции NewAPI, такие как переопределение параметров и перенаправление моделей, будут отключены. Это не является лучшей практикой.",
     "开启后不限制：必须设置模型倍率": "После включения без ограничений: необходимо установить множители моделей",
     "开启后未登录用户无法访问模型广场": "После включения незарегистрированные пользователи не смогут получить доступ к площади моделей",
     "开启批量操作": "Включить пакетные операции",
@@ -1531,6 +1535,7 @@
     "私有IP访问详细说明": "⚠️ Предупреждение безопасности: включение этой опции позволит доступ к ресурсам внутренней сети (localhost, частные сети). Включайте только при необходимости доступа к внутренним службам и понимании рисков безопасности.",
     "私有部署地址": "Адрес частного развёртывания",
     "秒": "секунда",
+    "移除 functionResponse.id 字段": "Удалить поле functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Удаление авторских знаков One API требует предварительного разрешения, поддержка проекта требует больших усилий, если этот проект важен для вас, пожалуйста, поддержите его",
     "窗口处理": "Обработка окна",
     "窗口等待": "Ожидание окна",
@@ -1773,7 +1778,7 @@
     "请先阅读并同意用户协议和隐私政策": "Пожалуйста, сначала прочтите и согласитесь с пользовательским соглашением и политикой конфиденциальности",
     "请再次输入新密码": "Пожалуйста, введите новый пароль ещё раз",
     "请前往个人设置 → 安全设置进行配置。": "Пожалуйста, перейдите в Личные настройки → Настройки безопасности для конфигурации.",
-    "请勿过度信任此功能，IP可能被伪造": "Не доверяйте этой функции чрезмерно, IP может быть подделан",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Не доверяйте этой функции чрезмерно, IP может быть подделан, используйте её вместе с nginx и CDN и другими шлюзами",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Пожалуйста, отредактируйте коэффициенты групп на странице системных настроек для добавления новой группы:",
     "请填写完整的产品信息": "Пожалуйста, заполните всю информацию о продукте",
     "请填写完整的管理员账号信息": "Пожалуйста, заполните полную информацию об учётной записи администратора",
@@ -2236,6 +2241,9 @@
     "默认用户消息": "Здравствуйте",
     "默认助手消息": "Здравствуйте! Чем я могу вам помочь?",
     "可选，用于复现结果": "Необязательно, для воспроизводимых результатов",
-    "随机种子 (留空为随机)": "Случайное зерно (оставьте пустым для случайного)"
+    "随机种子 (留空为随机)": "Случайное зерно (оставьте пустым для случайного)",
+    "跨分组重试": "Повторная попытка между группами",
+    "跨分组": "Межгрупповой",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "После включения, когда канал текущей группы не работает, он будет пытаться использовать канал следующей группы по порядку"
   }
 }

web/src/i18n/locales/vi.json

@@ -82,7 +82,7 @@
     "Homepage URL 填": "Điền URL trang chủ",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "Danh sách trắng IP",
+    "IP白名单（支持CIDR表达式）": "Danh sách trắng IP (hỗ trợ biểu thức CIDR)",
     "IP限制": "Hạn chế IP",
     "IP黑名单": "Danh sách đen IP",
     "JSON": "JSON",
@@ -136,6 +136,7 @@
     "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Quản lý danh mục giám sát Uptime Kuma, bạn có thể cấu hình nhiều danh mục giám sát để hiển thị trạng thái dịch vụ (tối đa 20)",
     "URL链接": "Liên kết URL",
     "User Info Endpoint": "User Info Endpoint",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI không hỗ trợ trường functionResponse.id. Khi bật, trường này sẽ tự động bị xóa",
     "Webhook 签名密钥": "Khóa chữ ký Webhook",
     "Webhook地址": "URL Webhook",
     "Webhook地址必须以https://开头": "URL Webhook phải bắt đầu bằng https://",
@@ -795,6 +796,9 @@
     "开启后，仅\"消费\"和\"错误\"日志将记录您的客户端IP地址": "Sau khi bật, chỉ nhật ký \"tiêu thụ\" và \"lỗi\" sẽ ghi lại địa chỉ IP máy khách của bạn",
     "开启后，将定期发送ping数据保持连接活跃": "Sau khi bật, dữ liệu ping sẽ được gửi định kỳ để giữ kết nối hoạt động",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "Khi bật, tất cả các yêu cầu sẽ được chuyển tiếp trực tiếp đến thượng nguồn mà không cần xử lý (chuyển hướng và thích ứng kênh cũng sẽ bị vô hiệu hóa). Vui lòng bật một cách thận trọng.",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Kênh này đã bật truyền qua yêu cầu. Các tính năng tích hợp của NewAPI như ghi đè tham số, chuyển hướng mô hình và thích ứng kênh sẽ bị vô hiệu hóa. Đây không phải là thực hành tốt nhất. Nếu phát sinh vấn đề, vui lòng không gửi issue.",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "Đã bật truyền qua yêu cầu toàn cục. Các tính năng tích hợp của NewAPI như ghi đè tham số, chuyển hướng mô hình và thích ứng kênh sẽ bị vô hiệu hóa. Đây không phải là thực hành tốt nhất. Nếu phát sinh vấn đề, vui lòng không gửi issue.",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "Kênh này đã bật truyền qua yêu cầu; các tính năng tích hợp của NewAPI như ghi đè tham số và chuyển hướng mô hình sẽ bị vô hiệu hóa. Đây không phải là thực hành tốt nhất.",
     "开启后不限制：必须设置模型倍率": "Sau khi bật, không giới hạn: phải đặt tỷ lệ mô hình",
     "开启后未登录用户无法访问模型广场": "Khi bật, người dùng chưa xác thực không thể truy cập thị trường mô hình",
     "开启批量操作": "Bật chọn hàng loạt",
@@ -1987,7 +1991,7 @@
     "请先阅读并同意用户协议和隐私政策": "Vui lòng đọc và đồng ý với thỏa thuận người dùng và chính sách bảo mật trước",
     "请再次输入新密码": "Vui lòng nhập lại mật khẩu mới",
     "请前往个人设置 → 安全设置进行配置。": "Vui lòng truy cập Cài đặt cá nhân → Cài đặt bảo mật để cấu hình.",
-    "请勿过度信任此功能，IP可能被伪造": "Đừng quá tin tưởng tính năng này, IP có thể bị giả mạo",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Đừng quá tin tưởng tính năng này, IP có thể bị giả mạo, vui lòng sử dụng cùng với nginx và các cổng khác như cdn",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Vui lòng chỉnh sửa tỷ lệ nhóm trên trang cài đặt hệ thống để thêm nhóm mới:",
     "请填写完整的管理员账号信息": "Vui lòng điền đầy đủ thông tin tài khoản quản trị viên",
     "请填写密钥": "Vui lòng điền khóa",
@@ -2648,6 +2652,7 @@
     "私有IP访问详细说明": "⚠️ Cảnh báo bảo mật: Bật tính năng này cho phép truy cập vào tài nguyên mạng nội bộ (localhost, mạng riêng). Chỉ bật nếu bạn cần truy cập các dịch vụ nội bộ và hiểu rõ các rủi ro bảo mật.",
     "私有部署地址": "Địa chỉ triển khai riêng",
     "秒": "Giây",
+    "移除 functionResponse.id 字段": "Xóa trường functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Việc xóa dấu bản quyền One API trước tiên phải được ủy quyền. Việc bảo trì dự án đòi hỏi rất nhiều nỗ lực. Nếu dự án này có ý nghĩa với bạn, vui lòng chủ động ủng hộ dự án này.",
     "窗口处理": "xử lý cửa sổ",
     "窗口等待": "chờ cửa sổ",
@@ -2736,6 +2741,9 @@
     "默认用户消息": "Xin chào",
     "默认助手消息": "Xin chào! Tôi có thể giúp gì cho bạn?",
     "可选，用于复现结果": "Tùy chọn, để tái tạo kết quả",
-    "随机种子 (留空为随机)": "Hạt giống ngẫu nhiên (để trống cho ngẫu nhiên)"
+    "随机种子 (留空为随机)": "Hạt giống ngẫu nhiên (để trống cho ngẫu nhiên)",
+    "跨分组重试": "Thử lại giữa các nhóm",
+    "跨分组": "Giữa các nhóm",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "Sau khi bật, khi kênh nhóm hiện tại thất bại, nó sẽ thử kênh của nhóm tiếp theo theo thứ tự"
   }
 }

web/src/i18n/locales/zh.json

@@ -95,7 +95,7 @@
     "Homepage URL 填": "Homepage URL 填",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单": "IP白名单",
+    "IP白名单（支持CIDR表达式）": "IP白名单（支持CIDR表达式）",
     "IP限制": "IP限制",
     "IP黑名单": "IP黑名单",
     "JSON": "JSON",
@@ -150,6 +150,7 @@
     "URL链接": "URL链接",
     "USD (美元)": "USD (美元)",
     "User Info Endpoint": "User Info Endpoint",
+    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段",
     "Webhook 密钥": "Webhook 密钥",
     "Webhook 签名密钥": "Webhook 签名密钥",
     "Webhook地址": "Webhook地址",
@@ -829,6 +830,9 @@
     "开启后，对免费模型（倍率为0，或者价格为0）的模型也会预消耗额度": "开启后，对免费模型（倍率为0，或者价格为0）的模型也会预消耗额度",
     "开启后，将定期发送ping数据保持连接活跃": "开启后，将定期发送ping数据保持连接活跃",
     "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启": "开启后，所有请求将直接透传给上游，不会进行任何处理（重定向和渠道适配也将失效）,请谨慎开启",
+    "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "该渠道已开启请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。",
+    "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。": "已开启全局请求透传：参数覆写、模型重定向、渠道适配等 NewAPI 内置功能将失效，非最佳实践；如因此产生问题，请勿提交 issue 反馈。",
+    "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。": "该渠道已开启请求透传，参数覆写、模型重定向等 NewAPI 内置功能将失效，非最佳实践。",
     "开启后不限制：必须设置模型倍率": "开启后不限制：必须设置模型倍率",
     "开启后未登录用户无法访问模型广场": "开启后未登录用户无法访问模型广场",
     "开启批量操作": "开启批量操作",
@@ -1498,6 +1502,7 @@
     "私有IP访问详细说明": "⚠️ 安全警告：启用此选项将允许访问内网资源（本地主机、私有网络）。仅在需要访问内部服务且了解安全风险的情况下启用。",
     "私有部署地址": "私有部署地址",
     "秒": "秒",
+    "移除 functionResponse.id 字段": "移除 functionResponse.id 字段",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目",
     "窗口处理": "窗口处理",
     "窗口等待": "窗口等待",
@@ -1740,7 +1745,7 @@
     "请先阅读并同意用户协议和隐私政策": "请先阅读并同意用户协议和隐私政策",
     "请再次输入新密码": "请再次输入新密码",
     "请前往个人设置 → 安全设置进行配置。": "请前往个人设置 → 安全设置进行配置。",
-    "请勿过度信任此功能，IP可能被伪造": "请勿过度信任此功能，IP可能被伪造",
+    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "请在系统设置页面编辑分组倍率以添加新的分组：",
     "请填写完整的产品信息": "请填写完整的产品信息",
     "请填写完整的管理员账号信息": "请填写完整的管理员账号信息",
@@ -2203,6 +2208,9 @@
     "默认用户消息": "你好",
     "默认助手消息": "你好！有什么我可以帮助你的吗？",
     "可选，用于复现结果": "可选，用于复现结果",
-    "随机种子 (留空为随机)": "随机种子 (留空为随机)"
+    "随机种子 (留空为随机)": "随机种子 (留空为随机)",
+    "跨分组重试": "跨分组重试",
+    "跨分组": "跨分组",
+    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道"
   }
 }

web/src/pages/Setting/Model/SettingGeminiModel.jsx

@@ -46,6 +46,7 @@ const DEFAULT_GEMINI_INPUTS = {
   'gemini.thinking_adapter_enabled': false,
   'gemini.thinking_adapter_budget_tokens_percentage': 0.6,
   'gemini.function_call_thought_signature_enabled': true,
+  'gemini.remove_function_response_id_enabled': true,
 };
 
 export default function SettingGeminiModel(props) {
@@ -186,6 +187,23 @@ export default function SettingGeminiModel(props) {
                 />
               </Col>
             </Row>
+            <Row>
+              <Col span={16}>
+                <Form.Switch
+                  label={t('移除 functionResponse.id 字段')}
+                  field={'gemini.remove_function_response_id_enabled'}
+                  extraText={t(
+                    'Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段',
+                  )}
+                  onChange={(value) =>
+                    setInputs({
+                      ...inputs,
+                      'gemini.remove_function_response_id_enabled': value,
+                    })
+                  }
+                />
+              </Col>
+            </Row>
             <Row>
               <Col xs={24} sm={12} md={8} lg={8} xl={8}>
                 <Form.TextArea

web/vite.config.js

@@ -21,6 +21,7 @@ import react from '@vitejs/plugin-react';
 import { defineConfig, transformWithEsbuild } from 'vite';
 import pkg from '@douyinfe/vite-plugin-semi';
 import path from 'path';
+import { codeInspectorPlugin } from 'code-inspector-plugin';
 const { vitePluginSemi } = pkg;
 
 // https://vitejs.dev/config/
@@ -31,6 +32,9 @@ export default defineConfig({
     },
   },
   plugins: [
+    codeInspectorPlugin({
+      bundler: 'vite',
+    }),
     {
       name: 'treat-js-files-as-jsx',
       async transform(code, id) {