feat: add remixGenerate const

feat: gemini 3 thinking level gemini-3-pro-preview-high

.env.example

@@ -63,10 +63,13 @@
 # 是否统计图片token
 # GET_MEDIA_TOKEN=true
 # 是否在非流（stream=false）情况下统计图片token
-# GET_MEDIA_TOKEN_NOT_STREAM=true
+# GET_MEDIA_TOKEN_NOT_STREAM=false
 # 设置 Dify 渠道是否输出工作流和节点信息到客户端
 # DIFY_DEBUG=true
 
+# LinuxDo相关配置
+LINUX_DO_TOKEN_ENDPOINT=https://connect.linux.do/oauth2/token
+LINUX_DO_USER_ENDPOINT=https://connect.linux.do/api/user
 
 # 节点类型
 # 如果是主节点则为master

.github/workflows/release.yml

@@ -22,6 +22,10 @@ jobs:
         uses: actions/checkout@v3
         with:
           fetch-depth: 0
+      - name: Determine Version
+        run: |
+          VERSION=$(git describe --tags)
+          echo "VERSION=$VERSION" >> $GITHUB_ENV
       - uses: oven-sh/setup-bun@v2
         with:
           bun-version: latest
@@ -31,7 +35,7 @@ jobs:
         run: |
           cd web
           bun install
-          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(git describe --tags) bun run build
+          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
         uses: actions/setup-go@v3
@@ -40,13 +44,11 @@ jobs:
       - name: Build Backend (amd64)
         run: |
           go mod download
-          VERSION=$(git describe --tags)
           go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION' -extldflags '-static'" -o new-api-$VERSION
       - name: Build Backend (arm64)
         run: |
           sudo apt-get update
           DEBIAN_FRONTEND=noninteractive sudo apt-get install -y gcc-aarch64-linux-gnu
-          VERSION=$(git describe --tags)
           CC=aarch64-linux-gnu-gcc CGO_ENABLED=1 GOOS=linux GOARCH=arm64 go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION' -extldflags '-static'" -o new-api-arm64-$VERSION
       - name: Release
         uses: softprops/action-gh-release@v2
@@ -65,6 +67,10 @@ jobs:
         uses: actions/checkout@v3
         with:
           fetch-depth: 0
+      - name: Determine Version
+        run: |
+          VERSION=$(git describe --tags)
+          echo "VERSION=$VERSION" >> $GITHUB_ENV
       - uses: oven-sh/setup-bun@v2
         with:
           bun-version: latest
@@ -75,7 +81,7 @@ jobs:
         run: |
           cd web
           bun install
-          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(git describe --tags) bun run build
+          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
         uses: actions/setup-go@v3
@@ -84,7 +90,6 @@ jobs:
       - name: Build Backend
         run: |
           go mod download
-          VERSION=$(git describe --tags)
           go build -ldflags "-X 'new-api/common.Version=$VERSION'" -o new-api-macos-$VERSION
       - name: Release
         uses: softprops/action-gh-release@v2
@@ -105,6 +110,10 @@ jobs:
         uses: actions/checkout@v3
         with:
           fetch-depth: 0
+      - name: Determine Version
+        run: |
+          VERSION=$(git describe --tags)
+          echo "VERSION=$VERSION" >> $GITHUB_ENV
       - uses: oven-sh/setup-bun@v2
         with:
           bun-version: latest
@@ -114,7 +123,7 @@ jobs:
         run: |
           cd web
           bun install
-          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(git describe --tags) bun run build
+          DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
         uses: actions/setup-go@v3
@@ -123,7 +132,6 @@ jobs:
       - name: Build Backend
         run: |
           go mod download
-          VERSION=$(git describe --tags)
           go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION'" -o new-api-$VERSION.exe
       - name: Release
         uses: softprops/action-gh-release@v2
@@ -132,5 +140,3 @@ jobs:
           files: new-api-*.exe
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-
-

.gitignore

@@ -16,6 +16,9 @@ new-api
 tiktoken_cache
 .eslintcache
 .gocache
+.cache
+web/bun.lock
 
 electron/node_modules
 electron/dist
+data/

README.en.md

@@ -193,6 +193,7 @@ docker run --name new-api -d --restart always \
 
 ### 🔐 Authorization and Security
 
+- 😈 Discord authorization login
 - 🤖 LinuxDO authorization login
 - 📱 Telegram authorization login
 - 🔑 OIDC unified authentication
@@ -237,6 +238,7 @@ docker run --name new-api -d --restart always \
 - `gemini-2.5-flash-nothinking` - Disable thinking mode
 - `gemini-2.5-pro-thinking` - Enable thinking mode
 - `gemini-2.5-pro-thinking-128` - Enable thinking mode with thinking budget of 128 tokens
+- You can also append `-low`, `-medium`, or `-high` to any Gemini model name to request the corresponding reasoning effort (no extra thinking-budget suffix needed).
 
 </details>
 
@@ -302,6 +304,7 @@ docker run --name new-api -d --restart always \
 | `SQL_DSN` | Database connection string | - |
 | `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` |
 | `AZURE_DEFAULT_API_VERSION` | Azure API version | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Error log switch | `false` |
 

README.fr.md

@@ -234,6 +234,7 @@ docker run --name new-api -d --restart always \
 - `gemini-2.5-flash-nothinking` - Désactiver le mode de pensée
 - `gemini-2.5-pro-thinking` - Activer le mode de pensée
 - `gemini-2.5-pro-thinking-128` - Activer le mode de pensée avec budget de pensée de 128 tokens
+- Vous pouvez également ajouter les suffixes `-low`, `-medium` ou `-high` aux modèles Gemini pour fixer le niveau d’effort de raisonnement (sans suffixe de budget supplémentaire).
 
 </details>
 
@@ -299,6 +300,7 @@ docker run --name new-api -d --restart always \
 | `SQL_DSN` | Chaine de connexion à la base de données | - |
 | `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` |
 | `AZURE_DEFAULT_API_VERSION` | Version de l'API Azure | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Interrupteur du journal d'erreurs | `false` |
 
@@ -438,4 +440,4 @@ Si ce projet vous est utile, bienvenue à nous donner une ⭐️ Étoile！
 
 <sub>Construit avec ❤️ par QuantumNous</sub>
 
-</div>
+</div>

README.ja.md

@@ -243,6 +243,7 @@ docker run --name new-api -d --restart always \
 - `gemini-2.5-flash-nothinking` - 思考モードを無効にする
 - `gemini-2.5-pro-thinking` - 思考モードを有効にする
 - `gemini-2.5-pro-thinking-128` - 思考モードを有効にし、思考予算を128トークンに設定する
+- Gemini モデル名の末尾に `-low` / `-medium` / `-high` を付けることで推論強度を直接指定できます（追加の思考予算サフィックスは不要です）。
 
 </details>
 
@@ -308,6 +309,7 @@ docker run --name new-api -d --restart always \
 | `SQL_DSN** | データベース接続文字列 | - |
 | `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` |
 | `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` |
 

README.md

@@ -193,6 +193,7 @@ docker run --name new-api -d --restart always \
 
 ### 🔐 授权与安全
 
+- 😈 Discord 授权登录
 - 🤖 LinuxDO 授权登录
 - 📱 Telegram 授权登录
 - 🔑 OIDC 统一认证
@@ -238,6 +239,7 @@ docker run --name new-api -d --restart always \
 - `gemini-2.5-flash-nothinking` - 禁用思考模式
 - `gemini-2.5-pro-thinking` - 启用思考模式
 - `gemini-2.5-pro-thinking-128` - 启用思考模式，并设置思考预算为128tokens
+- 也可以直接在 Gemini 模型名称后追加 `-low` / `-medium` / `-high` 来控制思考力度（无需再设置思考预算后缀）
 
 </details>
 
@@ -296,15 +298,16 @@ docker run --name new-api -d --restart always \
 <details>
 <summary>常用环境变量配置</summary>
 
-| 变量名 | 说明 | 默认值 |
-|--------|------|--------|
-| `SESSION_SECRET` | 会话密钥（多机部署必须） | - |
-| `CRYPTO_SECRET` | 加密密钥（Redis 必须） | - |
-| `SQL_DSN` | 数据库连接字符串 | - |
-| `REDIS_CONN_STRING` | Redis 连接字符串 | - |
-| `STREAMING_TIMEOUT` | 流式超时时间（秒） | `300` |
-| `AZURE_DEFAULT_API_VERSION` | Azure API 版本 | `2025-04-01-preview` |
-| `ERROR_LOG_ENABLED` | 错误日志开关 | `false` |
+| 变量名 | 说明                                                           | 默认值 |
+|--------|--------------------------------------------------------------|--------|
+| `SESSION_SECRET` | 会话密钥（多机部署必须）                                                 | - |
+| `CRYPTO_SECRET` | 加密密钥（Redis 必须）                                               | - |
+| `SQL_DSN` | 数据库连接字符串                                                     | - |
+| `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` |
+| `AZURE_DEFAULT_API_VERSION` | Azure API 版本                                                 | `2025-04-01-preview` |
+| `ERROR_LOG_ENABLED` | 错误日志开关                                                       | `false` |
 
 📖 **完整配置：** [环境变量文档](https://docs.newapi.pro/installation/environment-variables)
 

common/constants.go

@@ -121,6 +121,9 @@ var BatchUpdateInterval int
 
 var RelayTimeout int // unit is second
 
+var RelayMaxIdleConns int
+var RelayMaxIdleConnsPerHost int
+
 var GeminiSafetySetting string
 
 // https://docs.cohere.com/docs/safety-modes Type; NONE/CONTEXTUAL/STRICT

common/embed-file-system.go

@@ -4,6 +4,7 @@ import (
 	"embed"
 	"io/fs"
 	"net/http"
+	"os"
 
 	"github.com/gin-contrib/static"
 )
@@ -14,7 +15,7 @@ type embedFileSystem struct {
 	http.FileSystem
 }
 
-func (e embedFileSystem) Exists(prefix string, path string) bool {
+func (e *embedFileSystem) Exists(prefix string, path string) bool {
 	_, err := e.Open(path)
 	if err != nil {
 		return false
@@ -22,12 +23,21 @@ func (e embedFileSystem) Exists(prefix string, path string) bool {
 	return true
 }
 
+func (e *embedFileSystem) Open(name string) (http.File, error) {
+	if name == "/" {
+		// This will make sure the index page goes to NoRouter handler,
+		// which will use the replaced index bytes with analytic codes.
+		return nil, os.ErrNotExist
+	}
+	return e.FileSystem.Open(name)
+}
+
 func EmbedFolder(fsEmbed embed.FS, targetPath string) static.ServeFileSystem {
 	efs, err := fs.Sub(fsEmbed, targetPath)
 	if err != nil {
 		panic(err)
 	}
-	return embedFileSystem{
+	return &embedFileSystem{
 		FileSystem: http.FS(efs),
 	}
 }

common/gin.go

@@ -2,7 +2,9 @@ package common
 
 import (
 	"bytes"
+	"errors"
 	"io"
+	"mime"
 	"mime/multipart"
 	"net/http"
 	"net/url"
@@ -128,13 +130,13 @@ func ParseMultipartFormReusable(c *gin.Context) (*multipart.Form, error) {
 	}
 
 	contentType := c.Request.Header.Get("Content-Type")
-	boundary := ""
-	if idx := strings.Index(contentType, "boundary="); idx != -1 {
-		boundary = contentType[idx+9:]
+	boundary, err := parseBoundary(contentType)
+	if err != nil {
+		return nil, err
 	}
 
 	reader := multipart.NewReader(bytes.NewReader(requestBody), boundary)
-	form, err := reader.ReadForm(32 << 20) // 32 MB max memory
+	form, err := reader.ReadForm(multipartMemoryLimit())
 	if err != nil {
 		return nil, err
 	}
@@ -177,17 +179,16 @@ func parseFormData(data []byte, v any) error {
 
 func parseMultipartFormData(c *gin.Context, data []byte, v any) error {
 	contentType := c.Request.Header.Get("Content-Type")
-	boundary := ""
-	if idx := strings.Index(contentType, "boundary="); idx != -1 {
-		boundary = contentType[idx+9:]
-	}
-
-	if boundary == "" {
-		return Unmarshal(data, v) // Fallback to JSON
+	boundary, err := parseBoundary(contentType)
+	if err != nil {
+		if errors.Is(err, errBoundaryNotFound) {
+			return Unmarshal(data, v) // Fallback to JSON
+		}
+		return err
 	}
 
 	reader := multipart.NewReader(bytes.NewReader(data), boundary)
-	form, err := reader.ReadForm(32 << 20) // 32 MB max memory
+	form, err := reader.ReadForm(multipartMemoryLimit())
 	if err != nil {
 		return err
 	}
@@ -203,3 +204,31 @@ func parseMultipartFormData(c *gin.Context, data []byte, v any) error {
 
 	return processFormMap(formMap, v)
 }
+
+var errBoundaryNotFound = errors.New("multipart boundary not found")
+
+// parseBoundary extracts the multipart boundary from the Content-Type header using mime.ParseMediaType
+func parseBoundary(contentType string) (string, error) {
+	if contentType == "" {
+		return "", errBoundaryNotFound
+	}
+	// Boundary-UUID / boundary-------xxxxxx
+	_, params, err := mime.ParseMediaType(contentType)
+	if err != nil {
+		return "", err
+	}
+	boundary, ok := params["boundary"]
+	if !ok || boundary == "" {
+		return "", errBoundaryNotFound
+	}
+	return boundary, nil
+}
+
+// multipartMemoryLimit returns the configured multipart memory limit in bytes
+func multipartMemoryLimit() int64 {
+	limitMB := constant.MaxFileDownloadMB
+	if limitMB <= 0 {
+		limitMB = 32
+	}
+	return int64(limitMB) << 20
+}

common/init.go

@@ -30,6 +30,11 @@ func printHelp() {
 func InitEnv() {
 	flag.Parse()
 
+	envVersion := os.Getenv("VERSION")
+	if envVersion != "" {
+		Version = envVersion
+	}
+
 	if *PrintVersion {
 		fmt.Println(Version)
 		os.Exit(0)
@@ -85,6 +90,8 @@ func InitEnv() {
 	SyncFrequency = GetEnvOrDefault("SYNC_FREQUENCY", 60)
 	BatchUpdateInterval = GetEnvOrDefault("BATCH_UPDATE_INTERVAL", 5)
 	RelayTimeout = GetEnvOrDefault("RELAY_TIMEOUT", 0)
+	RelayMaxIdleConns = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS", 500)
+	RelayMaxIdleConnsPerHost = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS_PER_HOST", 100)
 
 	// Initialize string variables with GetEnvOrDefaultString
 	GeminiSafetySetting = GetEnvOrDefaultString("GEMINI_SAFETY_SETTING", "BLOCK_NONE")
@@ -109,10 +116,12 @@ func initConstantEnv() {
 	constant.StreamingTimeout = GetEnvOrDefault("STREAMING_TIMEOUT", 300)
 	constant.DifyDebug = GetEnvOrDefaultBool("DIFY_DEBUG", true)
 	constant.MaxFileDownloadMB = GetEnvOrDefault("MAX_FILE_DOWNLOAD_MB", 20)
+	constant.StreamScannerMaxBufferMB = GetEnvOrDefault("STREAM_SCANNER_MAX_BUFFER_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)
 	constant.GetMediaToken = GetEnvOrDefaultBool("GET_MEDIA_TOKEN", true)
-	constant.GetMediaTokenNotStream = GetEnvOrDefaultBool("GET_MEDIA_TOKEN_NOT_STREAM", true)
+	constant.GetMediaTokenNotStream = GetEnvOrDefaultBool("GET_MEDIA_TOKEN_NOT_STREAM", false)
 	constant.UpdateTask = GetEnvOrDefaultBool("UPDATE_TASK", true)
 	constant.AzureDefaultAPIVersion = GetEnvOrDefaultString("AZURE_DEFAULT_API_VERSION", "2025-04-01-preview")
 	constant.GeminiVisionMaxImageNum = GetEnvOrDefault("GEMINI_VISION_MAX_IMAGE_NUM", 16)
@@ -122,6 +131,8 @@ func initConstantEnv() {
 	constant.GenerateDefaultToken = GetEnvOrDefaultBool("GENERATE_DEFAULT_TOKEN", false)
 	// 是否启用错误日志
 	constant.ErrorLogEnabled = GetEnvOrDefaultBool("ERROR_LOG_ENABLED", false)
+	// 任务轮询时查询的最大数量
+	constant.TaskQueryLimit = GetEnvOrDefault("TASK_QUERY_LIMIT", 1000)
 
 	soraPatchStr := GetEnvOrDefaultString("TASK_PRICE_PATCH", "")
 	if soraPatchStr != "" {

common/model.go

@@ -17,6 +17,13 @@ var (
 		"flux-",
 		"flux.1-",
 	}
+	OpenAITextModels = []string{
+		"gpt-",
+		"o1",
+		"o3",
+		"o4",
+		"chatgpt",
+	}
 )
 
 func IsOpenAIResponseOnlyModel(modelName string) bool {
@@ -40,3 +47,13 @@ func IsImageGenerationModel(modelName string) bool {
 	}
 	return false
 }
+
+func IsOpenAITextModel(modelName string) bool {
+	modelName = strings.ToLower(modelName)
+	for _, m := range OpenAITextModels {
+		if strings.Contains(modelName, m) {
+			return true
+		}
+	}
+	return false
+}

constant/channel.go

@@ -180,3 +180,27 @@ func GetChannelTypeName(channelType int) string {
 	}
 	return "Unknown"
 }
+
+type ChannelSpecialBase struct {
+	ClaudeBaseURL string
+	OpenAIBaseURL string
+}
+
+var ChannelSpecialBases = map[string]ChannelSpecialBase{
+	"glm-coding-plan": {
+		ClaudeBaseURL: "https://open.bigmodel.cn/api/anthropic",
+		OpenAIBaseURL: "https://open.bigmodel.cn/api/coding/paas/v4",
+	},
+	"glm-coding-plan-international": {
+		ClaudeBaseURL: "https://api.z.ai/api/anthropic",
+		OpenAIBaseURL: "https://api.z.ai/api/coding/paas/v4",
+	},
+	"kimi-coding-plan": {
+		ClaudeBaseURL: "https://api.kimi.com/coding",
+		OpenAIBaseURL: "https://api.kimi.com/coding/v1",
+	},
+	"doubao-coding-plan": {
+		ClaudeBaseURL: "https://ark.cn-beijing.volces.com/api/coding",
+		OpenAIBaseURL: "https://ark.cn-beijing.volces.com/api/coding/v3",
+	},
+}

constant/context_key.go

@@ -3,8 +3,9 @@ package constant
 type ContextKey string
 
 const (
-	ContextKeyTokenCountMeta ContextKey = "token_count_meta"
-	ContextKeyPromptTokens   ContextKey = "prompt_tokens"
+	ContextKeyTokenCountMeta  ContextKey = "token_count_meta"
+	ContextKeyPromptTokens    ContextKey = "prompt_tokens"
+	ContextKeyEstimatedTokens ContextKey = "estimated_tokens"
 
 	ContextKeyOriginalModel    ContextKey = "original_model"
 	ContextKeyRequestStartTime ContextKey = "request_start_time"
@@ -46,5 +47,7 @@ const (
 	ContextKeyUsingGroup  ContextKey = "group"
 	ContextKeyUserName    ContextKey = "username"
 
+	ContextKeyLocalCountTokens ContextKey = "local_count_tokens"
+
 	ContextKeySystemPromptOverride ContextKey = "system_prompt_override"
 )

constant/env.go

@@ -3,7 +3,9 @@ package constant
 var StreamingTimeout int
 var DifyDebug bool
 var MaxFileDownloadMB int
+var StreamScannerMaxBufferMB int
 var ForceStreamOption bool
+var CountToken bool
 var GetMediaToken bool
 var GetMediaTokenNotStream bool
 var UpdateTask bool
@@ -13,6 +15,7 @@ var NotifyLimitCount int
 var NotificationLimitDurationMinute int
 var GenerateDefaultToken bool
 var ErrorLogEnabled bool
+var TaskQueryLimit int
 
 // temporary variable for sora patch, will be removed in future
 var TaskPricePatches []string

constant/task.go

@@ -15,6 +15,7 @@ const (
 	TaskActionTextGenerate      = "textGenerate"
 	TaskActionFirstTailGenerate = "firstTailGenerate"
 	TaskActionReferenceGenerate = "referenceGenerate"
+	TaskActionRemix             = "remixGenerate"
 )
 
 var SunoModel2Action = map[string]string{

controller/channel-test.go

@@ -351,7 +351,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string)
 			newAPIError: types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError),
 		}
 	}
-	info.PromptTokens = usage.PromptTokens
+	info.SetEstimatePromptTokens(usage.PromptTokens)
 
 	quota := 0
 	if !priceData.UsePrice {

controller/channel.go

@@ -91,7 +91,7 @@ func GetAllChannels(c *gin.Context) {
 			if tag == nil || *tag == "" {
 				continue
 			}
-			tagChannels, err := model.GetChannelsByTag(*tag, idSort)
+			tagChannels, err := model.GetChannelsByTag(*tag, idSort, false)
 			if err != nil {
 				continue
 			}
@@ -191,14 +191,40 @@ func FetchUpstreamModels(c *gin.Context) {
 	case constant.ChannelTypeAli:
 		url = fmt.Sprintf("%s/compatible-mode/v1/models", baseURL)
 	case constant.ChannelTypeZhipu_v4:
-		url = fmt.Sprintf("%s/api/paas/v4/models", baseURL)
+		if plan, ok := constant.ChannelSpecialBases[baseURL]; ok && plan.OpenAIBaseURL != "" {
+			url = fmt.Sprintf("%s/models", plan.OpenAIBaseURL)
+		} else {
+			url = fmt.Sprintf("%s/api/paas/v4/models", baseURL)
+		}
+	case constant.ChannelTypeVolcEngine:
+		if plan, ok := constant.ChannelSpecialBases[baseURL]; ok && plan.OpenAIBaseURL != "" {
+			url = fmt.Sprintf("%s/v1/models", plan.OpenAIBaseURL)
+		} else {
+			url = fmt.Sprintf("%s/v1/models", baseURL)
+		}
+	case constant.ChannelTypeMoonshot:
+		if plan, ok := constant.ChannelSpecialBases[baseURL]; ok && plan.OpenAIBaseURL != "" {
+			url = fmt.Sprintf("%s/models", plan.OpenAIBaseURL)
+		} else {
+			url = fmt.Sprintf("%s/v1/models", baseURL)
+		}
 	default:
 		url = fmt.Sprintf("%s/v1/models", baseURL)
 	}
 
+	// 获取用于请求的可用密钥（多密钥渠道优先使用启用状态的密钥）
+	key, _, apiErr := channel.GetNextEnabledKey()
+	if apiErr != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": fmt.Sprintf("获取渠道密钥失败: %s", apiErr.Error()),
+		})
+		return
+	}
+	key = strings.TrimSpace(key)
+
 	// 获取响应体 - 根据渠道类型决定是否添加 AuthHeader
 	var body []byte
-	key := strings.Split(channel.Key, "\n")[0]
 	switch channel.Type {
 	case constant.ChannelTypeAnthropic:
 		body, err = GetResponseBody("GET", url, channel, GetClaudeAuthHeader(key))
@@ -271,7 +297,7 @@ func SearchChannels(c *gin.Context) {
 		}
 		for _, tag := range tags {
 			if tag != nil && *tag != "" {
-				tagChannel, err := model.GetChannelsByTag(*tag, idSort)
+				tagChannel, err := model.GetChannelsByTag(*tag, idSort, false)
 				if err == nil {
 					channelData = append(channelData, tagChannel...)
 				}
@@ -1021,7 +1047,7 @@ func GetTagModels(c *gin.Context) {
 		return
 	}
 
-	channels, err := model.GetChannelsByTag(tag, false) // Assuming false for idSort is fine here
+	channels, err := model.GetChannelsByTag(tag, false, false) // idSort=false, selectAll=false
 	if err != nil {
 		c.JSON(http.StatusInternalServerError, gin.H{
 			"success": false,

controller/discord.go

@@ -0,0 +1,223 @@
+package controller
+
+import (
+	"encoding/json"
+	"errors"
+	"fmt"
+	"net/http"
+	"net/url"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/model"
+	"github.com/QuantumNous/new-api/setting/system_setting"
+
+	"github.com/gin-contrib/sessions"
+	"github.com/gin-gonic/gin"
+)
+
+type DiscordResponse struct {
+	AccessToken  string `json:"access_token"`
+	IDToken      string `json:"id_token"`
+	RefreshToken string `json:"refresh_token"`
+	TokenType    string `json:"token_type"`
+	ExpiresIn    int    `json:"expires_in"`
+	Scope        string `json:"scope"`
+}
+
+type DiscordUser struct {
+	UID  string `json:"id"`
+	ID   string `json:"username"`
+	Name string `json:"global_name"`
+}
+
+func getDiscordUserInfoByCode(code string) (*DiscordUser, error) {
+	if code == "" {
+		return nil, errors.New("无效的参数")
+	}
+
+	values := url.Values{}
+	values.Set("client_id", system_setting.GetDiscordSettings().ClientId)
+	values.Set("client_secret", system_setting.GetDiscordSettings().ClientSecret)
+	values.Set("code", code)
+	values.Set("grant_type", "authorization_code")
+	values.Set("redirect_uri", fmt.Sprintf("%s/oauth/discord", system_setting.ServerAddress))
+	formData := values.Encode()
+	req, err := http.NewRequest("POST", "https://discord.com/api/v10/oauth2/token", strings.NewReader(formData))
+	if err != nil {
+		return nil, err
+	}
+	req.Header.Set("Content-Type", "application/x-www-form-urlencoded")
+	req.Header.Set("Accept", "application/json")
+	client := http.Client{
+		Timeout: 5 * time.Second,
+	}
+	res, err := client.Do(req)
+	if err != nil {
+		common.SysLog(err.Error())
+		return nil, errors.New("无法连接至 Discord 服务器，请稍后重试！")
+	}
+	defer res.Body.Close()
+	var discordResponse DiscordResponse
+	err = json.NewDecoder(res.Body).Decode(&discordResponse)
+	if err != nil {
+		return nil, err
+	}
+
+	if discordResponse.AccessToken == "" {
+		common.SysError("Discord 获取 Token 失败，请检查设置！")
+		return nil, errors.New("Discord 获取 Token 失败，请检查设置！")
+	}
+
+	req, err = http.NewRequest("GET", "https://discord.com/api/v10/users/@me", nil)
+	if err != nil {
+		return nil, err
+	}
+	req.Header.Set("Authorization", "Bearer "+discordResponse.AccessToken)
+	res2, err := client.Do(req)
+	if err != nil {
+		common.SysLog(err.Error())
+		return nil, errors.New("无法连接至 Discord 服务器，请稍后重试！")
+	}
+	defer res2.Body.Close()
+	if res2.StatusCode != http.StatusOK {
+		common.SysError("Discord 获取用户信息失败！请检查设置！")
+		return nil, errors.New("Discord 获取用户信息失败！请检查设置！")
+	}
+
+	var discordUser DiscordUser
+	err = json.NewDecoder(res2.Body).Decode(&discordUser)
+	if err != nil {
+		return nil, err
+	}
+	if discordUser.UID == "" || discordUser.ID == "" {
+		common.SysError("Discord 获取用户信息为空！请检查设置！")
+		return nil, errors.New("Discord 获取用户信息为空！请检查设置！")
+	}
+	return &discordUser, nil
+}
+
+func DiscordOAuth(c *gin.Context) {
+	session := sessions.Default(c)
+	state := c.Query("state")
+	if state == "" || session.Get("oauth_state") == nil || state != session.Get("oauth_state").(string) {
+		c.JSON(http.StatusForbidden, gin.H{
+			"success": false,
+			"message": "state is empty or not same",
+		})
+		return
+	}
+	username := session.Get("username")
+	if username != nil {
+		DiscordBind(c)
+		return
+	}
+		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",
+		})
+		return
+	}
+	code := c.Query("code")
+	discordUser, err := getDiscordUserInfoByCode(code)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	user := model.User{
+		DiscordId: discordUser.UID,
+	}
+	if model.IsDiscordIdAlreadyTaken(user.DiscordId) {
+		err := user.FillUserByDiscordId()
+		if err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": err.Error(),
+			})
+			return
+		}
+	} else {
+		if common.RegisterEnabled {
+			if discordUser.ID != "" {
+				user.Username = discordUser.ID
+			} else {
+				user.Username = "discord_" + strconv.Itoa(model.GetMaxUserId()+1)
+			}
+			if discordUser.Name != "" {
+				user.DisplayName = discordUser.Name
+			} else {
+				user.DisplayName = "Discord User"
+			}
+			err := user.Insert(0)
+			if err != nil {
+				c.JSON(http.StatusOK, gin.H{
+					"success": false,
+					"message": err.Error(),
+				})
+				return
+			}
+		} else {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "管理员关闭了新用户注册",
+			})
+			return
+		}
+	}
+
+	if user.Status != common.UserStatusEnabled {
+		c.JSON(http.StatusOK, gin.H{
+			"message": "用户已被封禁",
+			"success": false,
+		})
+		return
+	}
+	setupLogin(&user, c)
+}
+
+func DiscordBind(c *gin.Context) {
+	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",
+		})
+		return
+	}
+	code := c.Query("code")
+	discordUser, err := getDiscordUserInfoByCode(code)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	user := model.User{
+		DiscordId: discordUser.UID,
+	}
+	if model.IsDiscordIdAlreadyTaken(user.DiscordId) {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "该 Discord 账户已被绑定",
+		})
+		return
+	}
+	session := sessions.Default(c)
+	id := session.Get("id")
+	user.Id = id.(int)
+	err = user.FillUserById()
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	user.DiscordId = discordUser.UID
+	err = user.Update(false)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "bind",
+	})
+}

controller/github.go

@@ -44,7 +44,7 @@ func getGitHubUserInfoByCode(code string) (*GitHubUser, error) {
 	req.Header.Set("Content-Type", "application/json")
 	req.Header.Set("Accept", "application/json")
 	client := http.Client{
-		Timeout: 5 * time.Second,
+		Timeout: 20 * time.Second,
 	}
 	res, err := client.Do(req)
 	if err != nil {

controller/linuxdo.go

@@ -84,7 +84,7 @@ func getLinuxdoUserInfoByCode(code string, c *gin.Context) (*LinuxdoUser, error)
 	}
 
 	// Get access token using Basic auth
-	tokenEndpoint := "https://connect.linux.do/oauth2/token"
+	tokenEndpoint := common.GetEnvOrDefaultString("LINUX_DO_TOKEN_ENDPOINT", "https://connect.linux.do/oauth2/token")
 	credentials := common.LinuxDOClientId + ":" + common.LinuxDOClientSecret
 	basicAuth := "Basic " + base64.StdEncoding.EncodeToString([]byte(credentials))
 
@@ -129,7 +129,7 @@ func getLinuxdoUserInfoByCode(code string, c *gin.Context) (*LinuxdoUser, error)
 	}
 
 	// Get user info
-	userEndpoint := "https://connect.linux.do/api/user"
+	userEndpoint := common.GetEnvOrDefaultString("LINUX_DO_USER_ENDPOINT", "https://connect.linux.do/api/user")
 	req, err = http.NewRequest("GET", userEndpoint, nil)
 	if err != nil {
 		return nil, err

controller/misc.go

@@ -52,6 +52,8 @@ func GetStatus(c *gin.Context) {
 		"email_verification":          common.EmailVerificationEnabled,
 		"github_oauth":                common.GitHubOAuthEnabled,
 		"github_client_id":            common.GitHubClientId,
+		"discord_oauth":               system_setting.GetDiscordSettings().Enabled,
+		"discord_client_id":           system_setting.GetDiscordSettings().ClientId,
 		"linuxdo_oauth":               common.LinuxDOOAuthEnabled,
 		"linuxdo_client_id":           common.LinuxDOClientId,
 		"linuxdo_minimum_trust_level": common.LinuxDOMinimumTrustLevel,

controller/model.go

@@ -16,6 +16,8 @@ import (
 	"github.com/QuantumNous/new-api/relay/channel/moonshot"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/setting/operation_setting"
+	"github.com/QuantumNous/new-api/setting/ratio_setting"
 	"github.com/gin-gonic/gin"
 	"github.com/samber/lo"
 )
@@ -109,6 +111,17 @@ func init() {
 func ListModels(c *gin.Context, modelType int) {
 	userOpenAiModels := make([]dto.OpenAIModels, 0)
 
+	acceptUnsetRatioModel := operation_setting.SelfUseModeEnabled
+	if !acceptUnsetRatioModel {
+		userId := c.GetInt("id")
+		if userId > 0 {
+			userSettings, _ := model.GetUserSetting(userId, false)
+			if userSettings.AcceptUnsetRatioModel {
+				acceptUnsetRatioModel = true
+			}
+		}
+	}
+
 	modelLimitEnable := common.GetContextKeyBool(c, constant.ContextKeyTokenModelLimitEnabled)
 	if modelLimitEnable {
 		s, ok := common.GetContextKey(c, constant.ContextKeyTokenModelLimit)
@@ -119,6 +132,12 @@ func ListModels(c *gin.Context, modelType int) {
 			tokenModelLimit = map[string]bool{}
 		}
 		for allowModel, _ := range tokenModelLimit {
+			if !acceptUnsetRatioModel {
+				_, _, exist := ratio_setting.GetModelRatioOrPrice(allowModel)
+				if !exist {
+					continue
+				}
+			}
 			if oaiModel, ok := openAIModelsMap[allowModel]; ok {
 				oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(allowModel)
 				userOpenAiModels = append(userOpenAiModels, oaiModel)
@@ -161,6 +180,12 @@ func ListModels(c *gin.Context, modelType int) {
 			models = model.GetGroupEnabledModels(group)
 		}
 		for _, modelName := range models {
+			if !acceptUnsetRatioModel {
+				_, _, exist := ratio_setting.GetModelRatioOrPrice(modelName)
+				if !exist {
+					continue
+				}
+			}
 			if oaiModel, ok := openAIModelsMap[modelName]; ok {
 				oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(modelName)
 				userOpenAiModels = append(userOpenAiModels, oaiModel)
@@ -175,6 +200,7 @@ func ListModels(c *gin.Context, modelType int) {
 			}
 		}
 	}
+
 	switch modelType {
 	case constant.ChannelTypeAnthropic:
 		useranthropicModels := make([]dto.AnthropicModel, len(userOpenAiModels))

controller/option.go

@@ -71,6 +71,14 @@ func UpdateOption(c *gin.Context) {
 			})
 			return
 		}
+	case "discord.enabled":
+		if option.Value == "true" && system_setting.GetDiscordSettings().ClientId == "" {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "无法启用 Discord OAuth，请先填入 Discord Client Id 以及 Discord Client Secret！",
+			})
+			return
+		}
 	case "oidc.enabled":
 		if option.Value == "true" && system_setting.GetOIDCSettings().ClientId == "" {
 			c.JSON(http.StatusOK, gin.H{

controller/relay.go

@@ -125,13 +125,13 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 		}
 	}
 
-	tokens, err := service.CountRequestToken(c, meta, relayInfo)
+	tokens, err := service.EstimateRequestToken(c, meta, relayInfo)
 	if err != nil {
 		newAPIError = types.NewError(err, types.ErrorCodeCountTokenFailed)
 		return
 	}
 
-	relayInfo.SetPromptTokens(tokens)
+	relayInfo.SetEstimatePromptTokens(tokens)
 
 	priceData, err := helper.ModelPriceHelper(c, relayInfo, tokens, meta)
 	if err != nil {

controller/task.go

@@ -29,7 +29,7 @@ func UpdateTaskBulk() {
 		time.Sleep(time.Duration(15) * time.Second)
 		common.SysLog("任务进度轮询开始")
 		ctx := context.TODO()
-		allTasks := model.GetAllUnFinishSyncTasks(500)
+		allTasks := model.GetAllUnFinishSyncTasks(constant.TaskQueryLimit)
 		platformTask := make(map[constant.TaskPlatform][]*model.Task)
 		for _, t := range allTasks {
 			platformTask[t.Platform] = append(platformTask[t.Platform], t)

controller/task_video.go

@@ -52,6 +52,7 @@ func updateVideoTaskAll(ctx context.Context, platform constant.TaskPlatform, cha
 	info.ChannelMeta = &relaycommon.ChannelMeta{
 		ChannelBaseUrl: cacheGetChannel.GetBaseURL(),
 	}
+	info.ApiKey = cacheGetChannel.Key
 	adaptor.Init(info)
 	for _, taskId := range taskIds {
 		if err := updateVideoSingleTask(ctx, adaptor, cacheGetChannel, taskId, taskM); err != nil {

controller/user.go

@@ -453,6 +453,7 @@ func GetSelf(c *gin.Context) {
 		"status":            user.Status,
 		"email":             user.Email,
 		"github_id":         user.GitHubId,
+		"discord_id":        user.DiscordId,
 		"oidc_id":           user.OidcId,
 		"wechat_id":         user.WeChatId,
 		"telegram_id":       user.TelegramId,

controller/video_proxy.go

@@ -117,13 +117,12 @@ func VideoProxy(c *gin.Context) {
 			return
 		}
 		req.Header.Set("x-goog-api-key", apiKey)
-	case constant.ChannelTypeAli:
-		// Video URL is directly in task.FailReason
-		videoURL = task.FailReason
-	default:
-		// Default (Sora, etc.): Use original logic
+	case constant.ChannelTypeOpenAI, constant.ChannelTypeSora:
 		videoURL = fmt.Sprintf("%s/v1/videos/%s/content", baseURL, task.TaskID)
 		req.Header.Set("Authorization", "Bearer "+channel.Key)
+	default:
+		// Video URL is directly in task.FailReason
+		videoURL = task.FailReason
 	}
 
 	req.URL, err = url.Parse(videoURL)

docs/api/api_auth.md

@@ -1,53 +0,0 @@
-# API 鉴权文档
-
-## 认证方式
-
-### Access Token
-
-对于需要鉴权的 API 接口，必须同时提供以下两个请求头来进行 Access Token 认证：
-
-1. **请求头中的 `Authorization` 字段**
-
-    将 Access Token 放置于 HTTP 请求头部的 `Authorization` 字段中，格式如下：
-
-    ```
-    Authorization: <your_access_token>
-    ```
-
-    其中 `<your_access_token>` 需要替换为实际的 Access Token 值。
-
-2. **请求头中的 `New-Api-User` 字段**
-
-    将用户 ID 放置于 HTTP 请求头部的 `New-Api-User` 字段中，格式如下：
-
-    ```
-    New-Api-User: <your_user_id>
-    ```
-
-    其中 `<your_user_id>` 需要替换为实际的用户 ID。
-
-**注意：**
-
-*   **必须同时提供 `Authorization` 和 `New-Api-User` 两个请求头才能通过鉴权。**
-*   如果只提供其中一个请求头，或者两个请求头都未提供，则会返回 `401 Unauthorized` 错误。
-*   如果 `Authorization` 中的 Access Token 无效，则会返回 `401 Unauthorized` 错误，并提示“无权进行此操作，access token 无效”。
-*   如果 `New-Api-User` 中的用户 ID 与 Access Token 不匹配，则会返回 `401 Unauthorized` 错误，并提示“无权进行此操作，与登录用户不匹配，请重新登录”。
-*   如果没有提供 `New-Api-User` 请求头，则会返回 `401 Unauthorized` 错误，并提示“无权进行此操作，未提供 New-Api-User”。
-*   如果 `New-Api-User` 请求头格式错误，则会返回 `401 Unauthorized` 错误，并提示“无权进行此操作，New-Api-User 格式错误”。
-*   如果用户已被禁用，则会返回 `403 Forbidden` 错误，并提示“用户已被封禁”。
-*   如果用户权限不足，则会返回 `403 Forbidden` 错误，并提示“无权进行此操作，权限不足”。
-*   如果用户信息无效，则会返回 `403 Forbidden` 错误，并提示“无权进行此操作，用户信息无效”。
-
-## Curl 示例
-
-假设您的 Access Token 为 `access_token`，用户 ID 为 `123`，要访问的 API 接口为 `/api/user/self`，则可以使用以下 curl 命令：
-
-```bash
-curl -X GET \
-  -H "Authorization: access_token" \
-  -H "New-Api-User: 123" \
-  https://your-domain.com/api/user/self
-```
-
-请将 `access_token`、`123` 和 `https://your-domain.com` 替换为实际的值。
-

docs/api/web_api.md

@@ -1,197 +0,0 @@
-# New API – Web 界面后端接口文档
-
-> 本文档汇总了 **New API** 后端提供给前端 Web 界面的全部 REST 接口（不含 *Relay* 相关接口）。
->
-> 接口前缀统一为 `https://<your-domain>`，以下仅列出 **路径**、**HTTP 方法**、**鉴权要求** 与 **功能简介**。
->
-> 鉴权级别说明：
-> * **公开** – 不需要登录即可调用
-> * **用户** – 需携带用户 Token（`middleware.UserAuth`）
-> * **管理员** – 需管理员 Token（`middleware.AdminAuth`）
-> * **Root** – 仅限最高权限 Root 用户（`middleware.RootAuth`）
-
----
-
-## 1. 初始化 / 系统状态
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET  | /api/setup | 公开 | 获取系统初始化状态 |
-| POST | /api/setup | 公开 | 完成首次安装向导 |
-| GET  | /api/status | 公开 | 获取运行状态摘要 |
-| GET  | /api/uptime/status | 公开 | Uptime-Kuma 兼容状态探针 |
-| GET  | /api/status/test | 管理员 | 测试后端与依赖组件是否正常 |
-
-## 2. 公共信息
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/models | 用户 | 获取前端可用模型列表 |
-| GET | /api/notice | 公开 | 获取公告栏内容 |
-| GET | /api/about | 公开 | 关于页面信息 |
-| GET | /api/home_page_content | 公开 | 首页自定义内容 |
-| GET | /api/pricing | 可匿名/用户 | 价格与套餐信息 |
-| GET | /api/ratio_config | 公开 | 模型倍率配置（仅公开字段） |
-
-## 3. 邮件 / 身份验证
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/verification | 公开 (限流) | 发送邮箱验证邮件 |
-| GET | /api/reset_password | 公开 (限流) | 发送重置密码邮件 |
-| POST | /api/user/reset | 公开 | 提交重置密码请求 |
-
-## 4. OAuth / 第三方登录
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/oauth/github | 公开 | GitHub OAuth 跳转 |
-| GET | /api/oauth/oidc | 公开 | OIDC 通用 OAuth 跳转 |
-| GET | /api/oauth/linuxdo | 公开 | LinuxDo OAuth 跳转 |
-| GET | /api/oauth/wechat | 公开 | 微信扫码登录跳转 |
-| GET | /api/oauth/wechat/bind | 公开 | 微信账户绑定 |
-| GET | /api/oauth/email/bind | 公开 | 邮箱绑定 |
-| GET | /api/oauth/telegram/login | 公开 | Telegram 登录 |
-| GET | /api/oauth/telegram/bind | 公开 | Telegram 账户绑定 |
-| GET | /api/oauth/state | 公开 | 获取随机 state（防 CSRF） |
-
-## 5. 用户模块
-### 5.1 账号注册/登录
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| POST | /api/user/register | 公开 | 注册新账号 |
-| POST | /api/user/login | 公开 | 用户登录 |
-| GET  | /api/user/logout | 用户 | 退出登录 |
-| GET  | /api/user/epay/notify | 公开 | Epay 支付回调 |
-| GET  | /api/user/groups | 公开 | 列出所有分组（无鉴权版） |
-
-### 5.2 用户自身操作 (需登录)
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/user/self/groups | 用户 | 获取自己所在分组 |
-| GET | /api/user/self | 用户 | 获取个人资料 |
-| GET | /api/user/models | 用户 | 获取模型可见性 |
-| PUT | /api/user/self | 用户 | 修改个人资料 |
-| DELETE | /api/user/self | 用户 | 注销账号 |
-| GET | /api/user/token | 用户 | 生成用户级别 Access Token |
-| GET | /api/user/aff | 用户 | 获取推广码信息 |
-| POST | /api/user/topup | 用户 | 余额直充 |
-| POST | /api/user/pay | 用户 | 提交支付订单 |
-| POST | /api/user/amount | 用户 | 余额支付 |
-| POST | /api/user/aff_transfer | 用户 | 推广额度转账 |
-| PUT | /api/user/setting | 用户 | 更新用户设置 |
-
-### 5.3 管理员用户管理
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/user/ | 管理员 | 获取全部用户列表 |
-| GET | /api/user/search | 管理员 | 搜索用户 |
-| GET | /api/user/:id | 管理员 | 获取单个用户信息 |
-| POST | /api/user/ | 管理员 | 创建用户 |
-| POST | /api/user/manage | 管理员 | 冻结/重置等管理操作 |
-| PUT | /api/user/ | 管理员 | 更新用户 |
-| DELETE | /api/user/:id | 管理员 | 删除用户 |
-
-## 6. 站点选项 (Root)
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/option/ | Root | 获取全局配置 |
-| PUT | /api/option/ | Root | 更新全局配置 |
-| POST | /api/option/rest_model_ratio | Root | 重置模型倍率 |
-| POST | /api/option/migrate_console_setting | Root | 迁移旧版控制台配置 |
-
-## 7. 模型倍率同步 (Root)
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/ratio_sync/channels | Root | 获取可同步渠道列表 |
-| POST | /api/ratio_sync/fetch | Root | 从上游拉取倍率 |
-
-## 8. 渠道管理 (管理员)
-| 方法 | 路径 | 说明 |
-|------|------|------|
-| GET | /api/channel/ | 获取渠道列表 |
-| GET | /api/channel/search | 搜索渠道 |
-| GET | /api/channel/models | 查询渠道模型能力 |
-| GET | /api/channel/models_enabled | 查询启用模型能力 |
-| GET | /api/channel/:id | 获取单个渠道 |
-| GET | /api/channel/test | 批量测试渠道连通性 |
-| GET | /api/channel/test/:id | 单个渠道测试 |
-| GET | /api/channel/update_balance | 批量刷新余额 |
-| GET | /api/channel/update_balance/:id | 单个刷新余额 |
-| POST | /api/channel/ | 新增渠道 |
-| PUT | /api/channel/ | 更新渠道 |
-| DELETE | /api/channel/disabled | 删除已禁用渠道 |
-| POST | /api/channel/tag/disabled | 批量禁用标签渠道 |
-| POST | /api/channel/tag/enabled | 批量启用标签渠道 |
-| PUT | /api/channel/tag | 编辑渠道标签 |
-| DELETE | /api/channel/:id | 删除渠道 |
-| POST | /api/channel/batch | 批量删除渠道 |
-| POST | /api/channel/fix | 修复渠道能力表 |
-| GET | /api/channel/fetch_models/:id | 拉取单渠道模型 |
-| POST | /api/channel/fetch_models | 拉取全部渠道模型 |
-| POST | /api/channel/batch/tag | 批量设置渠道标签 |
-| GET | /api/channel/tag/models | 根据标签获取模型 |
-| POST | /api/channel/copy/:id | 复制渠道 |
-
-## 9. Token 管理
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/token/ | 用户 | 获取全部 Token |
-| GET | /api/token/search | 用户 | 搜索 Token |
-| GET | /api/token/:id | 用户 | 获取单个 Token |
-| POST | /api/token/ | 用户 | 创建 Token |
-| PUT | /api/token/ | 用户 | 更新 Token |
-| DELETE | /api/token/:id | 用户 | 删除 Token |
-| POST | /api/token/batch | 用户 | 批量删除 Token |
-
-## 10. 兑换码管理 (管理员)
-| 方法 | 路径 | 说明 |
-|------|------|------|
-| GET | /api/redemption/ | 获取兑换码列表 |
-| GET | /api/redemption/search | 搜索兑换码 |
-| GET | /api/redemption/:id | 获取单个兑换码 |
-| POST | /api/redemption/ | 创建兑换码 |
-| PUT | /api/redemption/ | 更新兑换码 |
-| DELETE | /api/redemption/invalid | 删除无效兑换码 |
-| DELETE | /api/redemption/:id | 删除兑换码 |
-
-## 11. 日志
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/log/ | 管理员 | 获取全部日志 |
-| DELETE | /api/log/ | 管理员 | 删除历史日志 |
-| GET | /api/log/stat | 管理员 | 日志统计 |
-| GET | /api/log/self/stat | 用户 | 我的日志统计 |
-| GET | /api/log/search | 管理员 | 搜索全部日志 |
-| GET | /api/log/self | 用户 | 获取我的日志 |
-| GET | /api/log/self/search | 用户 | 搜索我的日志 |
-| GET | /api/log/token | 公开 | 根据 Token 查询日志（支持 CORS） |
-
-## 12. 数据统计
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/data/ | 管理员 | 全站用量按日期统计 |
-| GET | /api/data/self | 用户 | 我的用量按日期统计 |
-
-## 13. 分组
-| GET | /api/group/ | 管理员 | 获取全部分组列表 |
-
-## 14. Midjourney 任务
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/mj/self | 用户 | 获取自己的 MJ 任务 |
-| GET | /api/mj/ | 管理员 | 获取全部 MJ 任务 |
-
-## 15. 任务中心
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /api/task/self | 用户 | 获取我的任务 |
-| GET | /api/task/ | 管理员 | 获取全部任务 |
-
-## 16. 账户计费面板 (Dashboard)
-| 方法 | 路径 | 鉴权 | 说明 |
-|------|------|------|------|
-| GET | /dashboard/billing/subscription | 用户 Token | 获取订阅额度信息 |
-| GET | /v1/dashboard/billing/subscription | 同上 | 兼容 OpenAI SDK 路径 |
-| GET | /dashboard/billing/usage | 用户 Token | 获取使用量信息 |
-| GET | /v1/dashboard/billing/usage | 同上 | 兼容 OpenAI SDK 路径 |
-
----
-
-> **更新日期**：2025.07.17

docs/models/Midjourney.md

@@ -1,82 +0,0 @@
-# Midjourney Proxy API文档
-
-**简介**:Midjourney Proxy API文档
-
-## 接口列表
-支持的接口如下：
-+ [x] /mj/submit/imagine
-+ [x] /mj/submit/change
-+ [x] /mj/submit/blend
-+ [x] /mj/submit/describe
-+ [x] /mj/image/{id} （通过此接口获取图片，**请必须在系统设置中填写服务器地址！！**）
-+ [x] /mj/task/{id}/fetch （此接口返回的图片地址为经过One API转发的地址）
-+ [x] /task/list-by-condition
-+ [x] /mj/submit/action （仅midjourney-proxy-plus支持，下同）
-+ [x] /mj/submit/modal
-+ [x] /mj/submit/shorten
-+ [x] /mj/task/{id}/image-seed
-+ [x] /mj/insight-face/swap （InsightFace）
-
-## 模型列表
-
-### midjourney-proxy支持
-
-- mj_imagine (绘图)
-- mj_variation (变换)
-- mj_reroll (重绘)
-- mj_blend (混合)
-- mj_upscale (放大)
-- mj_describe (图生文)
-
-### 仅midjourney-proxy-plus支持
-
-- mj_zoom (比例变焦)
-- mj_shorten (提示词缩短)
-- mj_modal (窗口提交，局部重绘和自定义比例变焦必须和mj_modal一同添加)
-- mj_inpaint (局部重绘提交，必须和mj_modal一同添加)
-- mj_custom_zoom (自定义比例变焦，必须和mj_modal一同添加)
-- mj_high_variation (强变换)
-- mj_low_variation (弱变换)
-- mj_pan (平移)
-- swap_face (换脸)
-
-## 模型价格设置（在设置-运营设置-模型固定价格设置中设置）
-```json
-{
-  "mj_imagine": 0.1,
-  "mj_variation": 0.1,
-  "mj_reroll": 0.1,
-  "mj_blend": 0.1,
-  "mj_modal": 0.1,
-  "mj_zoom": 0.1,
-  "mj_shorten": 0.1,
-  "mj_high_variation": 0.1,
-  "mj_low_variation": 0.1,
-  "mj_pan": 0.1,
-  "mj_inpaint": 0,
-  "mj_custom_zoom": 0,
-  "mj_describe": 0.05,
-  "mj_upscale": 0.05,
-  "swap_face": 0.05
-}
-```
-其中mj_inpaint和mj_custom_zoom的价格设置为0，是因为这两个模型需要搭配mj_modal使用，所以价格由mj_modal决定。
-
-## 渠道设置
-
-### 对接 midjourney-proxy(plus)
-
-1.
-
-部署Midjourney-Proxy，并配置好midjourney账号等（强烈建议设置密钥），[项目地址](https://github.com/novicezk/midjourney-proxy)
-
-2. 在渠道管理中添加渠道，渠道类型选择**Midjourney Proxy**，如果是plus版本选择**Midjourney Proxy Plus**
-   ，模型请参考上方模型列表
-3. **代理**填写midjourney-proxy部署的地址，例如：http://localhost:8080
-4. 密钥填写midjourney-proxy的密钥，如果没有设置密钥，可以随便填
-
-### 对接上游new api
-
-1. 在渠道管理中添加渠道，渠道类型选择**Midjourney Proxy Plus**，模型请参考上方模型列表
-2. **代理**填写上游new api的地址，例如：http://localhost:3000
-3. 密钥填写上游new api的密钥

docs/models/Rerank.md

@@ -1,62 +0,0 @@
-# Rerank API文档
-
-**简介**:Rerank API文档
-
-## 接入Dify
-模型供应商选择Jina，按要求填写模型信息即可接入Dify。
-
-## 请求方式
-
-Post: /v1/rerank
-
-Request:
-
-```json
-{
-  "model": "jina-reranker-v2-base-multilingual",
-  "query": "What is the capital of the United States?",
-  "top_n": 3,
-  "documents": [
-    "Carson City is the capital city of the American state of Nevada.",
-    "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean. Its capital is Saipan.",
-    "Washington, D.C. (also known as simply Washington or D.C., and officially as the District of Columbia) is the capital of the United States. It is a federal district.",
-    "Capitalization or capitalisation in English grammar is the use of a capital letter at the start of a word. English usage varies from capitalization in other languages.",
-    "Capital punishment (the death penalty) has existed in the United States since beforethe United States was a country. As of 2017, capital punishment is legal in 30 of the 50 states."
-  ]
-}
-```
-
-Response:
-
-```json
-{
-  "results": [
-    {
-      "document": {
-        "text": "Washington, D.C. (also known as simply Washington or D.C., and officially as the District of Columbia) is the capital of the United States. It is a federal district."
-      },
-      "index": 2,
-      "relevance_score": 0.9999702
-    },
-    {
-      "document": {
-        "text": "Carson City is the capital city of the American state of Nevada."
-      },
-      "index": 0,
-      "relevance_score": 0.67800725
-    },
-    {
-      "document": {
-        "text": "Capitalization or capitalisation in English grammar is the use of a capital letter at the start of a word. English usage varies from capitalization in other languages."
-      },
-      "index": 3,
-      "relevance_score": 0.02800752
-    }
-  ],
-  "usage": {
-    "prompt_tokens": 158,
-    "completion_tokens": 0,
-    "total_tokens": 158
-  }
-}
-```

docs/models/Suno.md

@@ -1,44 +0,0 @@
-# Suno API文档
-
-**简介**:Suno API文档
-
-## 接口列表
-支持的接口如下：
-+ [x] /suno/submit/music
-+ [x] /suno/submit/lyrics
-+ [x] /suno/fetch
-+ [x] /suno/fetch/:id
-
-## 模型列表
-
-### Suno API支持
-
-- suno_music (自定义模式、灵感模式、续写)
-- suno_lyrics (生成歌词)
-
-
-## 模型价格设置（在设置-运营设置-模型固定价格设置中设置）
-```json
-{
-  "suno_music": 0.3,
-  "suno_lyrics": 0.01
-}
-```
-
-## 渠道设置
-
-### 对接 Suno API
-
-1.
-部署 Suno API，并配置好suno账号等（强烈建议设置密钥），[项目地址](https://github.com/Suno-API/Suno-API)
-
-2. 在渠道管理中添加渠道，渠道类型选择**Suno API**
-   ，模型请参考上方模型列表
-3. **代理**填写 Suno API 部署的地址，例如：http://localhost:8080
-4. 密钥填写 Suno API 的密钥，如果没有设置密钥，可以随便填
-
-### 对接上游new api
-
-1. 在渠道管理中添加渠道，渠道类型选择**Suno API**，或任意类型，只需模型包含上方模型列表的模型
-2. **代理**填写上游new api的地址，例如：http://localhost:3000
-3. 密钥填写上游new api的密钥

dto/claude.go

@@ -203,6 +203,9 @@ type ClaudeRequest struct {
 	Stream            bool            `json:"stream,omitempty"`
 	Tools             any             `json:"tools,omitempty"`
 	ContextManagement json.RawMessage `json:"context_management,omitempty"`
+	OutputConfig      json.RawMessage `json:"output_config,omitempty"`
+	OutputFormat      json.RawMessage `json:"output_format,omitempty"`
+	Container         json.RawMessage `json:"container,omitempty"`
 	ToolChoice        any             `json:"tool_choice,omitempty"`
 	Thinking          *Thinking       `json:"thinking,omitempty"`
 	McpServers        json.RawMessage `json:"mcp_servers,omitempty"`

dto/gemini.go

@@ -141,6 +141,39 @@ func (r *GeminiChatRequest) SetTools(tools []GeminiChatTool) {
 type GeminiThinkingConfig struct {
 	IncludeThoughts bool `json:"includeThoughts,omitempty"`
 	ThinkingBudget  *int `json:"thinkingBudget,omitempty"`
+	// TODO Conflict with thinkingbudget.
+	ThinkingLevel string `json:"thinkingLevel,omitempty"`
+}
+
+// UnmarshalJSON allows GeminiThinkingConfig to accept both snake_case and camelCase fields.
+func (c *GeminiThinkingConfig) UnmarshalJSON(data []byte) error {
+	type Alias GeminiThinkingConfig
+	var aux struct {
+		Alias
+		IncludeThoughtsSnake *bool  `json:"include_thoughts,omitempty"`
+		ThinkingBudgetSnake  *int   `json:"thinking_budget,omitempty"`
+		ThinkingLevelSnake   string `json:"thinking_level,omitempty"`
+	}
+
+	if err := common.Unmarshal(data, &aux); err != nil {
+		return err
+	}
+
+	*c = GeminiThinkingConfig(aux.Alias)
+
+	if aux.IncludeThoughtsSnake != nil {
+		c.IncludeThoughts = *aux.IncludeThoughtsSnake
+	}
+
+	if aux.ThinkingBudgetSnake != nil {
+		c.ThinkingBudget = aux.ThinkingBudgetSnake
+	}
+
+	if aux.ThinkingLevelSnake != "" {
+		c.ThinkingLevel = aux.ThinkingLevelSnake
+	}
+
+	return nil
 }
 
 func (c *GeminiThinkingConfig) SetThinkingBudget(budget int) {
@@ -182,8 +215,12 @@ type FunctionCall struct {
 }
 
 type GeminiFunctionResponse struct {
-	Name     string                 `json:"name"`
-	Response map[string]interface{} `json:"response"`
+	Name         string                 `json:"name"`
+	Response     map[string]interface{} `json:"response"`
+	WillContinue json.RawMessage        `json:"willContinue,omitempty"`
+	Scheduling   json.RawMessage        `json:"scheduling,omitempty"`
+	Parts        json.RawMessage        `json:"parts,omitempty"`
+	ID           json.RawMessage        `json:"id,omitempty"`
 }
 
 type GeminiPartExecutableCode struct {
@@ -202,11 +239,15 @@ type GeminiFileData struct {
 }
 
 type GeminiPart struct {
-	Text                string                         `json:"text,omitempty"`
-	Thought             bool                           `json:"thought,omitempty"`
-	InlineData          *GeminiInlineData              `json:"inlineData,omitempty"`
-	FunctionCall        *FunctionCall                  `json:"functionCall,omitempty"`
-	FunctionResponse    *GeminiFunctionResponse        `json:"functionResponse,omitempty"`
+	Text             string                  `json:"text,omitempty"`
+	Thought          bool                    `json:"thought,omitempty"`
+	InlineData       *GeminiInlineData       `json:"inlineData,omitempty"`
+	FunctionCall     *FunctionCall           `json:"functionCall,omitempty"`
+	ThoughtSignature json.RawMessage         `json:"thoughtSignature,omitempty"`
+	FunctionResponse *GeminiFunctionResponse `json:"functionResponse,omitempty"`
+	// Optional. Media resolution for the input media.
+	MediaResolution     json.RawMessage                `json:"mediaResolution,omitempty"`
+	VideoMetadata       json.RawMessage                `json:"videoMetadata,omitempty"`
 	FileData            *GeminiFileData                `json:"fileData,omitempty"`
 	ExecutableCode      *GeminiPartExecutableCode      `json:"executableCode,omitempty"`
 	CodeExecutionResult *GeminiPartCodeExecutionResult `json:"codeExecutionResult,omitempty"`

dto/openai_request.go

@@ -66,10 +66,11 @@ type GeneralOpenAIRequest struct {
 	// 注意：默认过滤此字段以保护用户隐私，但过滤后可能导致 Codex 无法正常使用
 	Store json.RawMessage `json:"store,omitempty"`
 	// Used by OpenAI to cache responses for similar requests to optimize your cache hit rates. Replaces the user field
-	PromptCacheKey string          `json:"prompt_cache_key,omitempty"`
-	LogitBias      json.RawMessage `json:"logit_bias,omitempty"`
-	Metadata       json.RawMessage `json:"metadata,omitempty"`
-	Prediction     json.RawMessage `json:"prediction,omitempty"`
+	PromptCacheKey       string          `json:"prompt_cache_key,omitempty"`
+	PromptCacheRetention json.RawMessage `json:"prompt_cache_retention,omitempty"`
+	LogitBias            json.RawMessage `json:"logit_bias,omitempty"`
+	Metadata             json.RawMessage `json:"metadata,omitempty"`
+	Prediction           json.RawMessage `json:"prediction,omitempty"`
 	// gemini
 	ExtraBody json.RawMessage `json:"extra_body,omitempty"`
 	//xai
@@ -798,19 +799,20 @@ type OpenAIResponsesRequest struct {
 	PreviousResponseID string          `json:"previous_response_id,omitempty"`
 	Reasoning          *Reasoning      `json:"reasoning,omitempty"`
 	// 服务层级字段，用于指定 API 服务等级。允许透传可能导致实际计费高于预期，默认应过滤
-	ServiceTier    string          `json:"service_tier,omitempty"`
-	Store          json.RawMessage `json:"store,omitempty"`
-	PromptCacheKey json.RawMessage `json:"prompt_cache_key,omitempty"`
-	Stream         bool            `json:"stream,omitempty"`
-	Temperature    float64         `json:"temperature,omitempty"`
-	Text           json.RawMessage `json:"text,omitempty"`
-	ToolChoice     json.RawMessage `json:"tool_choice,omitempty"`
-	Tools          json.RawMessage `json:"tools,omitempty"` // 需要处理的参数很少，MCP 参数太多不确定，所以用 map
-	TopP           float64         `json:"top_p,omitempty"`
-	Truncation     string          `json:"truncation,omitempty"`
-	User           string          `json:"user,omitempty"`
-	MaxToolCalls   uint            `json:"max_tool_calls,omitempty"`
-	Prompt         json.RawMessage `json:"prompt,omitempty"`
+	ServiceTier          string          `json:"service_tier,omitempty"`
+	Store                json.RawMessage `json:"store,omitempty"`
+	PromptCacheKey       json.RawMessage `json:"prompt_cache_key,omitempty"`
+	PromptCacheRetention json.RawMessage `json:"prompt_cache_retention,omitempty"`
+	Stream               bool            `json:"stream,omitempty"`
+	Temperature          float64         `json:"temperature,omitempty"`
+	Text                 json.RawMessage `json:"text,omitempty"`
+	ToolChoice           json.RawMessage `json:"tool_choice,omitempty"`
+	Tools                json.RawMessage `json:"tools,omitempty"` // 需要处理的参数很少，MCP 参数太多不确定，所以用 map
+	TopP                 float64         `json:"top_p,omitempty"`
+	Truncation           string          `json:"truncation,omitempty"`
+	User                 string          `json:"user,omitempty"`
+	MaxToolCalls         uint            `json:"max_tool_calls,omitempty"`
+	Prompt               json.RawMessage `json:"prompt,omitempty"`
 }
 
 func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta {
@@ -895,6 +897,12 @@ type Reasoning struct {
 	Summary string `json:"summary,omitempty"`
 }
 
+type Input struct {
+	Type    string          `json:"type,omitempty"`
+	Role    string          `json:"role,omitempty"`
+	Content json.RawMessage `json:"content,omitempty"`
+}
+
 type MediaInput struct {
 	Type     string `json:"type"`
 	Text     string `json:"text,omitempty"`
@@ -913,7 +921,7 @@ func (r *OpenAIResponsesRequest) ParseInput() []MediaInput {
 		return nil
 	}
 
-	var inputs []MediaInput
+	var mediaInputs []MediaInput
 
 	// Try string first
 	// if str, ok := common.GetJsonType(r.Input); ok {
@@ -923,60 +931,74 @@ func (r *OpenAIResponsesRequest) ParseInput() []MediaInput {
 	if common.GetJsonType(r.Input) == "string" {
 		var str string
 		_ = common.Unmarshal(r.Input, &str)
-		inputs = append(inputs, MediaInput{Type: "input_text", Text: str})
-		return inputs
+		mediaInputs = append(mediaInputs, MediaInput{Type: "input_text", Text: str})
+		return mediaInputs
 	}
 
 	// Try array of parts
 	if common.GetJsonType(r.Input) == "array" {
-		var array []any
-		_ = common.Unmarshal(r.Input, &array)
-		for _, itemAny := range array {
-			// Already parsed MediaInput
-			if media, ok := itemAny.(MediaInput); ok {
-				inputs = append(inputs, media)
-				continue
+		var inputs []Input
+		_ = common.Unmarshal(r.Input, &inputs)
+		for _, input := range inputs {
+			if common.GetJsonType(input.Content) == "string" {
+				var str string
+				_ = common.Unmarshal(input.Content, &str)
+				mediaInputs = append(mediaInputs, MediaInput{Type: "input_text", Text: str})
 			}
-			// Generic map
-			item, ok := itemAny.(map[string]any)
-			if !ok {
-				continue
-			}
-			typeVal, ok := item["type"].(string)
-			if !ok {
-				continue
-			}
-			switch typeVal {
-			case "input_text":
-				text, _ := item["text"].(string)
-				inputs = append(inputs, MediaInput{Type: "input_text", Text: text})
-			case "input_image":
-				// image_url may be string or object with url field
-				var imageUrl string
-				switch v := item["image_url"].(type) {
-				case string:
-					imageUrl = v
-				case map[string]any:
-					if url, ok := v["url"].(string); ok {
-						imageUrl = url
+
+			if common.GetJsonType(input.Content) == "array" {
+				var array []any
+				_ = common.Unmarshal(input.Content, &array)
+				for _, itemAny := range array {
+					// Already parsed MediaContent
+					if media, ok := itemAny.(MediaInput); ok {
+						mediaInputs = append(mediaInputs, media)
+						continue
+					}
+
+					// Generic map
+					item, ok := itemAny.(map[string]any)
+					if !ok {
+						continue
+					}
+
+					typeVal, ok := item["type"].(string)
+					if !ok {
+						continue
+					}
+					switch typeVal {
+					case "input_text":
+						text, _ := item["text"].(string)
+						mediaInputs = append(mediaInputs, MediaInput{Type: "input_text", Text: text})
+					case "input_image":
+						// image_url may be string or object with url field
+						var imageUrl string
+						switch v := item["image_url"].(type) {
+						case string:
+							imageUrl = v
+						case map[string]any:
+							if url, ok := v["url"].(string); ok {
+								imageUrl = url
+							}
+						}
+						mediaInputs = append(mediaInputs, MediaInput{Type: "input_image", ImageUrl: imageUrl})
+					case "input_file":
+						// file_url may be string or object with url field
+						var fileUrl string
+						switch v := item["file_url"].(type) {
+						case string:
+							fileUrl = v
+						case map[string]any:
+							if url, ok := v["url"].(string); ok {
+								fileUrl = url
+							}
+						}
+						mediaInputs = append(mediaInputs, MediaInput{Type: "input_file", FileUrl: fileUrl})
 					}
 				}
-				inputs = append(inputs, MediaInput{Type: "input_image", ImageUrl: imageUrl})
-			case "input_file":
-				// file_url may be string or object with url field
-				var fileUrl string
-				switch v := item["file_url"].(type) {
-				case string:
-					fileUrl = v
-				case map[string]any:
-					if url, ok := v["url"].(string); ok {
-						fileUrl = url
-					}
-				}
-				inputs = append(inputs, MediaInput{Type: "input_file", FileUrl: fileUrl})
 			}
 		}
 	}
 
-	return inputs
+	return mediaInputs
 }

electron/package-lock.json

@@ -2784,9 +2784,9 @@
       }
     },
     "node_modules/js-yaml": {
-      "version": "4.1.0",
-      "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.1.0.tgz",
-      "integrity": "sha512-wpxZs9NoxZaJESJGIZTyDEaYpl0FKSA+FB9aJiyemKhMwkxQg63h4T1KJgUGHpTqPDNRcmmYLugrRjJlBtWvRA==",
+      "version": "4.1.1",
+      "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.1.1.tgz",
+      "integrity": "sha512-qQKT4zQxXl8lLwBtHMWwaTcGfFOZviOJet3Oy/xmGk2gZH677CJM9EvtfdSkgWcATZhj/55JZ0rmy3myCT5lsA==",
       "dev": true,
       "license": "MIT",
       "dependencies": {

go.mod

@@ -43,10 +43,10 @@ require (
 	github.com/tidwall/sjson v1.2.5
 	github.com/tiktoken-go/tokenizer v0.6.2
 	github.com/yapingcat/gomedia v0.0.0-20240906162731-17feea57090c
-	golang.org/x/crypto v0.42.0
+	golang.org/x/crypto v0.45.0
 	golang.org/x/image v0.23.0
-	golang.org/x/net v0.43.0
-	golang.org/x/sync v0.17.0
+	golang.org/x/net v0.47.0
+	golang.org/x/sync v0.18.0
 	gorm.io/driver/mysql v1.4.3
 	gorm.io/driver/postgres v1.5.2
 	gorm.io/gorm v1.25.2
@@ -111,8 +111,8 @@ require (
 	github.com/yusufpapurcu/wmi v1.2.3 // indirect
 	golang.org/x/arch v0.21.0 // indirect
 	golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 // indirect
-	golang.org/x/sys v0.36.0 // indirect
-	golang.org/x/text v0.29.0 // indirect
+	golang.org/x/sys v0.38.0 // indirect
+	golang.org/x/text v0.31.0 // indirect
 	google.golang.org/protobuf v1.34.2 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 	modernc.org/libc v1.22.5 // indirect

go.sum

@@ -281,18 +281,18 @@ go.uber.org/mock v0.6.0/go.mod h1:KiVJ4BqZJaMj4svdfmHM0AUx4NJYO8ZNpPnZn1Z+BBU=
 golang.org/x/arch v0.21.0 h1:iTC9o7+wP6cPWpDWkivCvQFGAHDQ59SrSxsLPcnkArw=
 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 v0.42.0 h1:chiH31gIWm57EkTXpwnqf8qeuMUi0yekh6mT2AvFlqI=
-golang.org/x/crypto v0.42.0/go.mod h1:4+rDnOTJhQCx2q7/j6rAN5XDw8kPjeaXEUR2eL94ix8=
+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/image v0.23.0 h1:HseQ7c2OpPKTPVzNjG5fwJsOTCiiwS4QdsYi5XU6H68=
 golang.org/x/image v0.23.0/go.mod h1:wJJBTdLfCCf3tiHa1fNxpZmUI4mmoZvwMCPP0ddoNKY=
 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 v0.43.0 h1:lat02VYK2j4aLzMzecihNvTlJNQUq316m2Mr9rnM6YE=
-golang.org/x/net v0.43.0/go.mod h1:vhO1fvI4dGsIjh73sWfUVjj3N7CA9WkKJNQm2svM6Jg=
-golang.org/x/sync v0.17.0 h1:l60nONMj9l5drqw6jlhIELNv9I0A4OFgRsG9k2oT9Ug=
-golang.org/x/sync v0.17.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
+golang.org/x/net v0.47.0 h1:Mx+4dIFzqraBXUugkia1OOvlD6LemFo1ALMHjrXDOhY=
+golang.org/x/net v0.47.0/go.mod h1:/jNxtkgq5yWUGYkaZGqo27cfGZ1c5Nen03aYrrKpVRU=
+golang.org/x/sync v0.18.0 h1:kr88TuHDroi+UVf+0hZnirlk8o8T+4MrK6mr60WkH/I=
+golang.org/x/sync v0.18.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
 golang.org/x/sys v0.0.0-20190726091711-fc99dfbffb4e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
@@ -304,15 +304,15 @@ golang.org/x/sys v0.0.0-20210806184541-e5e7981a1069/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.8.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.11.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.36.0 h1:KVRy2GtZBrk1cBYA7MKu5bEZFxQk4NIDV6RLVcC8o0k=
-golang.org/x/sys v0.36.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/sys v0.38.0 h1:3yZWxaJjBmCWXqhN1qh02AkOnCQ1poK6oF+a7xWL6Gc=
+golang.org/x/sys v0.38.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
-golang.org/x/text v0.29.0 h1:1neNs90w9YzJ9BocxfsQNHKuAT4pkghyXc4nhZ6sJvk=
-golang.org/x/text v0.29.0/go.mod h1:7MhJOA9CD2qZyOKYazxdYMF85OwPdEr9jTtBpO7ydH4=
+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/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=

middleware/distributor.go

@@ -181,6 +181,10 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 		}
 		c.Set("platform", string(constant.TaskPlatformSuno))
 		c.Set("relay_mode", relayMode)
+	} else if strings.Contains(c.Request.URL.Path, "/v1/videos/") && strings.HasSuffix(c.Request.URL.Path, "/remix") {
+		relayMode := relayconstant.RelayModeVideoSubmit
+		c.Set("relay_mode", relayMode)
+		shouldSelectChannel = false
 	} else if strings.Contains(c.Request.URL.Path, "/v1/videos") {
 		//curl https://api.openai.com/v1/videos \
 		//  -H "Authorization: Bearer $OPENAI_API_KEY" \

model/channel.go

@@ -272,13 +272,17 @@ func GetAllChannels(startIdx int, num int, selectAll bool, idSort bool) ([]*Chan
 	return channels, err
 }
 
-func GetChannelsByTag(tag string, idSort bool) ([]*Channel, error) {
+func GetChannelsByTag(tag string, idSort bool, selectAll bool) ([]*Channel, error) {
 	var channels []*Channel
 	order := "priority desc"
 	if idSort {
 		order = "id desc"
 	}
-	err := DB.Where("tag = ?", tag).Order(order).Find(&channels).Error
+	query := DB.Where("tag = ?", tag).Order(order)
+	if !selectAll {
+		query = query.Omit("key")
+	}
+	err := query.Find(&channels).Error
 	return channels, err
 }
 
@@ -728,7 +732,7 @@ func EditChannelByTag(tag string, newTag *string, modelMapping *string, models *
 		return err
 	}
 	if shouldReCreateAbilities {
-		channels, err := GetChannelsByTag(updatedTag, false)
+		channels, err := GetChannelsByTag(updatedTag, false, false)
 		if err == nil {
 			for _, channel := range channels {
 				err = channel.UpdateAbilities(nil)

model/task.go

@@ -429,3 +429,14 @@ func TaskCountAllUserTask(userId int, queryParams SyncTaskQueryParams) int64 {
 	_ = query.Count(&total).Error
 	return total
 }
+func (t *Task) ToOpenAIVideo() *dto.OpenAIVideo {
+	openAIVideo := dto.NewOpenAIVideo()
+	openAIVideo.ID = t.TaskID
+	openAIVideo.Status = t.Status.ToVideoStatus()
+	openAIVideo.Model = t.Properties.OriginModelName
+	openAIVideo.SetProgressStr(t.Progress)
+	openAIVideo.CreatedAt = t.CreatedAt
+	openAIVideo.CompletedAt = t.UpdatedAt
+	openAIVideo.SetMetadata("url", t.FailReason)
+	return openAIVideo
+}

model/user.go

@@ -27,6 +27,7 @@ type User struct {
 	Status           int            `json:"status" gorm:"type:int;default:1"` // enabled, disabled
 	Email            string         `json:"email" gorm:"index" validate:"max=50"`
 	GitHubId         string         `json:"github_id" gorm:"column:github_id;index"`
+	DiscordId        string         `json:"discord_id" gorm:"column:discord_id;index"`
 	OidcId           string         `json:"oidc_id" gorm:"column:oidc_id;index"`
 	WeChatId         string         `json:"wechat_id" gorm:"column:wechat_id;index"`
 	TelegramId       string         `json:"telegram_id" gorm:"column:telegram_id;index"`
@@ -539,6 +540,14 @@ func (user *User) FillUserByGitHubId() error {
 	return nil
 }
 
+func (user *User) FillUserByDiscordId() error {
+	if user.DiscordId == "" {
+		return errors.New("discord id 为空！")
+	}
+	DB.Where(User{DiscordId: user.DiscordId}).First(user)
+	return nil
+}
+
 func (user *User) FillUserByOidcId() error {
 	if user.OidcId == "" {
 		return errors.New("oidc id 为空！")
@@ -578,6 +587,10 @@ func IsGitHubIdAlreadyTaken(githubId string) bool {
 	return DB.Unscoped().Where("github_id = ?", githubId).Find(&User{}).RowsAffected == 1
 }
 
+func IsDiscordIdAlreadyTaken(discordId string) bool {
+	return DB.Unscoped().Where("discord_id = ?", discordId).Find(&User{}).RowsAffected == 1
+}
+
 func IsOidcIdAlreadyTaken(oidcId string) bool {
 	return DB.Where("oidc_id = ?", oidcId).Find(&User{}).RowsAffected == 1
 }

relay/channel/ali/adaptor.go

@@ -47,7 +47,11 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		case constant.RelayModeImagesGenerations:
 			fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/text2image/image-synthesis", info.ChannelBaseUrl)
 		case constant.RelayModeImagesEdits:
-			fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
+			if isWanModel(info.OriginModelName) {
+				fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/image2image/image-synthesis", info.ChannelBaseUrl)
+			} else {
+				fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
+			}
 		case constant.RelayModeCompletions:
 			fullRequestURL = fmt.Sprintf("%s/compatible-mode/v1/completions", info.ChannelBaseUrl)
 		default:
@@ -71,6 +75,9 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
 		req.Set("X-DashScope-Async", "enable")
 	}
 	if info.RelayMode == constant.RelayModeImagesEdits {
+		if isWanModel(info.OriginModelName) {
+			req.Set("X-DashScope-Async", "enable")
+		}
 		req.Set("Content-Type", "application/json")
 	}
 	return nil
@@ -82,15 +89,15 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 	}
 	// docs: https://bailian.console.aliyun.com/?tab=api#/api/?type=model&url=2712216
 	// fix: InternalError.Algo.InvalidParameter: The value of the enable_thinking parameter is restricted to True.
-	if strings.Contains(request.Model, "thinking") {
-		request.EnableThinking = true
-		request.Stream = true
-		info.IsStream = true
-	}
-	// fix: ali parameter.enable_thinking must be set to false for non-streaming calls
-	if !info.IsStream {
-		request.EnableThinking = false
-	}
+	//if strings.Contains(request.Model, "thinking") {
+	//	request.EnableThinking = true
+	//	request.Stream = true
+	//	info.IsStream = true
+	//}
+	//// fix: ali parameter.enable_thinking must be set to false for non-streaming calls
+	//if !info.IsStream {
+	//	request.EnableThinking = false
+	//}
 
 	switch info.RelayMode {
 	default:
@@ -107,6 +114,9 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 		}
 		return aliRequest, nil
 	} else if info.RelayMode == constant.RelayModeImagesEdits {
+		if isWanModel(info.OriginModelName) {
+			return oaiFormEdit2WanxImageEdit(c, info, request)
+		}
 		// ali image edit https://bailian.console.aliyun.com/?tab=api#/api/?type=model&url=2976416
 		// 如果用户使用表单，则需要解析表单数据
 		if strings.Contains(c.Request.Header.Get("Content-Type"), "multipart/form-data") {
@@ -161,7 +171,11 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 		case constant.RelayModeImagesGenerations:
 			err, usage = aliImageHandler(c, resp, info)
 		case constant.RelayModeImagesEdits:
-			err, usage = aliImageEditHandler(c, resp, info)
+			if isWanModel(info.OriginModelName) {
+				err, usage = aliImageHandler(c, resp, info)
+			} else {
+				err, usage = aliImageEditHandler(c, resp, info)
+			}
 		case constant.RelayModeRerank:
 			err, usage = RerankHandler(c, resp, info)
 		default:

relay/channel/ali/dto.go

@@ -112,6 +112,19 @@ type AliImageInput struct {
 	Messages       []AliMessage `json:"messages,omitempty"`
 }
 
+type WanImageInput struct {
+	Prompt         string   `json:"prompt"`                    // 必需：文本提示词，描述生成图像中期望包含的元素和视觉特点
+	Images         []string `json:"images"`                    // 必需：图像URL数组，长度不超过2，支持HTTP/HTTPS URL或Base64编码
+	NegativePrompt string   `json:"negative_prompt,omitempty"` // 可选：反向提示词，描述不希望在画面中看到的内容
+}
+
+type WanImageParameters struct {
+	N         int     `json:"n,omitempty"`         // 生成图片数量，取值范围1-4，默认4
+	Watermark *bool   `json:"watermark,omitempty"` // 是否添加水印标识，默认false
+	Seed      int     `json:"seed,omitempty"`      // 随机数种子，取值范围[0, 2147483647]
+	Strength  float64 `json:"strength,omitempty"`  // 修改幅度 0.0-1.0，默认0.5（部分模型支持）
+}
+
 type AliRerankParameters struct {
 	TopN            *int  `json:"top_n,omitempty"`
 	ReturnDocuments *bool `json:"return_documents,omitempty"`

relay/channel/ali/image.go

@@ -58,11 +58,7 @@ func oaiImage2Ali(request dto.ImageRequest) (*AliImageRequest, error) {
 	return &imageRequest, nil
 }
 
-func oaiFormEdit2AliImageEdit(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (*AliImageRequest, error) {
-	var imageRequest AliImageRequest
-	imageRequest.Model = request.Model
-	imageRequest.ResponseFormat = request.ResponseFormat
-
+func getImageBase64sFromForm(c *gin.Context, fieldName string) ([]string, error) {
 	mf := c.Request.MultipartForm
 	if mf == nil {
 		if _, err := c.MultipartForm(); err != nil {
@@ -127,7 +123,18 @@ func oaiFormEdit2AliImageEdit(c *gin.Context, info *relaycommon.RelayInfo, reque
 		imageBase64s = append(imageBase64s, dataURL)
 		image.Close()
 	}
+	return imageBase64s, nil
+}
 
+func oaiFormEdit2AliImageEdit(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (*AliImageRequest, error) {
+	var imageRequest AliImageRequest
+	imageRequest.Model = request.Model
+	imageRequest.ResponseFormat = request.ResponseFormat
+
+	imageBase64s, err := getImageBase64sFromForm(c, "image")
+	if err != nil {
+		return nil, fmt.Errorf("get image base64s from form failed: %w", err)
+	}
 	//dto.MediaContent{}
 	mediaContents := make([]AliMediaContent, len(imageBase64s))
 	for i, b64 := range imageBase64s {

relay/channel/ali/image_wan.go

@@ -0,0 +1,39 @@
+package ali
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/dto"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+
+	"github.com/gin-gonic/gin"
+)
+
+func oaiFormEdit2WanxImageEdit(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (*AliImageRequest, error) {
+	var err error
+	var imageRequest AliImageRequest
+	imageRequest.Model = request.Model
+	imageRequest.ResponseFormat = request.ResponseFormat
+	wanInput := WanImageInput{
+		Prompt: request.Prompt,
+	}
+
+	if err := common.UnmarshalBodyReusable(c, &wanInput); err != nil {
+		return nil, err
+	}
+	if wanInput.Images, err = getImageBase64sFromForm(c, "image"); err != nil {
+		return nil, fmt.Errorf("get image base64s from form failed: %w", err)
+	}
+	wanParams := WanImageParameters{
+		N: int(request.N),
+	}
+	imageRequest.Input = wanInput
+	imageRequest.Parameters = wanParams
+	return &imageRequest, nil
+}
+
+func isWanModel(modelName string) bool {
+	return strings.Contains(modelName, "wan")
+}

relay/channel/aws/constants.go

@@ -18,6 +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",
 	// Nova models
 	"nova-micro-v1:0":   "amazon.nova-micro-v1:0",
 	"nova-lite-v1:0":    "amazon.nova-lite-v1:0",
@@ -76,6 +77,11 @@ var awsModelCanCrossRegionMap = map[string]map[string]bool{
 		"ap": true,
 		"eu": true,
 	},
+	"anthropic.claude-opus-4-5-20251101-v1:0": {
+		"us": true,
+		"ap": true,
+		"eu": true,
+	},
 	"anthropic.claude-haiku-4-5-20251001-v1:0": {
 		"us": true,
 		"ap": true,

relay/channel/aws/dto.go

@@ -1,15 +1,21 @@
 package aws
 
 import (
+	"context"
+	"encoding/json"
 	"io"
+	"net/http"
+	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
 )
 
 type AwsClaudeRequest struct {
 	// AnthropicVersion should be "bedrock-2023-05-31"
 	AnthropicVersion string              `json:"anthropic_version"`
+	AnthropicBeta    json.RawMessage     `json:"anthropic_beta,omitempty"`
 	System           any                 `json:"system,omitempty"`
 	Messages         []dto.ClaudeMessage `json:"messages"`
 	MaxTokens        uint                `json:"max_tokens,omitempty"`
@@ -22,29 +28,28 @@ type AwsClaudeRequest struct {
 	Thinking         *dto.Thinking       `json:"thinking,omitempty"`
 }
 
-func copyRequest(req *dto.ClaudeRequest) *AwsClaudeRequest {
-	return &AwsClaudeRequest{
-		AnthropicVersion: "bedrock-2023-05-31",
-		System:           req.System,
-		Messages:         req.Messages,
-		MaxTokens:        req.MaxTokens,
-		Temperature:      req.Temperature,
-		TopP:             req.TopP,
-		TopK:             req.TopK,
-		StopSequences:    req.StopSequences,
-		Tools:            req.Tools,
-		ToolChoice:       req.ToolChoice,
-		Thinking:         req.Thinking,
-	}
-}
-
-func formatRequest(requestBody io.Reader) (*AwsClaudeRequest, error) {
+func formatRequest(requestBody io.Reader, requestHeader http.Header) (*AwsClaudeRequest, error) {
 	var awsClaudeRequest AwsClaudeRequest
 	err := common.DecodeJson(requestBody, &awsClaudeRequest)
 	if err != nil {
 		return nil, err
 	}
 	awsClaudeRequest.AnthropicVersion = "bedrock-2023-05-31"
+
+	// check header anthropic-beta
+	anthropicBetaValues := requestHeader.Get("anthropic-beta")
+	if len(anthropicBetaValues) > 0 {
+		var tempArray []string
+		tempArray = strings.Split(anthropicBetaValues, ",")
+		if len(tempArray) > 0 {
+			betaJson, err := json.Marshal(tempArray)
+			if err != nil {
+				return nil, err
+			}
+			awsClaudeRequest.AnthropicBeta = betaJson
+		}
+	}
+	logger.LogJson(context.Background(), "json", awsClaudeRequest)
 	return &awsClaudeRequest, nil
 }
 

relay/channel/aws/relay-aws.go

@@ -25,6 +25,17 @@ import (
 	"github.com/aws/smithy-go/auth/bearer"
 )
 
+// getAwsErrorStatusCode extracts HTTP status code from AWS SDK error
+func getAwsErrorStatusCode(err error) int {
+	// Check for HTTP response error which contains status code
+	var httpErr interface{ HTTPStatusCode() int }
+	if errors.As(err, &httpErr) {
+		return httpErr.HTTPStatusCode()
+	}
+	// Default to 500 if we can't determine the status code
+	return http.StatusInternalServerError
+}
+
 func newAwsClient(c *gin.Context, info *relaycommon.RelayInfo) (*bedrockruntime.Client, error) {
 	var (
 		httpClient *http.Client
@@ -73,7 +84,6 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 	}
 	a.AwsClient = awsCli
 
-	println(info.UpstreamModelName)
 	// 获取对应的AWS模型ID
 	awsModelId := getAwsModelID(info.UpstreamModelName)
 
@@ -83,6 +93,10 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 		awsModelId = awsModelCrossRegion(awsModelId, awsRegionPrefix)
 	}
 
+	// init empty request.header
+	requestHeader := http.Header{}
+	a.SetupRequestHeader(c, &requestHeader, info)
+
 	if isNovaModel(awsModelId) {
 		var novaReq *NovaRequest
 		err = common.DecodeJson(requestBody, &novaReq)
@@ -104,7 +118,7 @@ func doAwsClientRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor,
 		awsReq.Body = reqBody
 		return nil, nil
 	} else {
-		awsClaudeReq, err := formatRequest(requestBody)
+		awsClaudeReq, err := formatRequest(requestBody, requestHeader)
 		if err != nil {
 			return nil, types.NewError(errors.Wrap(err, "format aws request fail"), types.ErrorCodeBadRequestBody)
 		}
@@ -170,7 +184,8 @@ func awsHandler(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor) (*types
 
 	awsResp, err := a.AwsClient.InvokeModel(c.Request.Context(), a.AwsReq.(*bedrockruntime.InvokeModelInput))
 	if err != nil {
-		return types.NewOpenAIError(errors.Wrap(err, "InvokeModel"), types.ErrorCodeAwsInvokeError, http.StatusInternalServerError), nil
+		statusCode := getAwsErrorStatusCode(err)
+		return types.NewOpenAIError(errors.Wrap(err, "InvokeModel"), types.ErrorCodeAwsInvokeError, statusCode), nil
 	}
 
 	claudeInfo := &claude.ClaudeResponseInfo{
@@ -196,7 +211,8 @@ func awsHandler(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor) (*types
 func awsStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor) (*types.NewAPIError, *dto.Usage) {
 	awsResp, err := a.AwsClient.InvokeModelWithResponseStream(c.Request.Context(), a.AwsReq.(*bedrockruntime.InvokeModelWithResponseStreamInput))
 	if err != nil {
-		return types.NewOpenAIError(errors.Wrap(err, "InvokeModelWithResponseStream"), types.ErrorCodeAwsInvokeError, http.StatusInternalServerError), nil
+		statusCode := getAwsErrorStatusCode(err)
+		return types.NewOpenAIError(errors.Wrap(err, "InvokeModelWithResponseStream"), types.ErrorCodeAwsInvokeError, statusCode), nil
 	}
 	stream := awsResp.GetStream()
 	defer stream.Close()
@@ -235,7 +251,8 @@ func handleNovaRequest(c *gin.Context, info *relaycommon.RelayInfo, a *Adaptor)
 
 	awsResp, err := a.AwsClient.InvokeModel(c.Request.Context(), a.AwsReq.(*bedrockruntime.InvokeModelInput))
 	if err != nil {
-		return types.NewError(errors.Wrap(err, "InvokeModel"), types.ErrorCodeChannelAwsClientError), nil
+		statusCode := getAwsErrorStatusCode(err)
+		return types.NewOpenAIError(errors.Wrap(err, "InvokeModel"), types.ErrorCodeAwsInvokeError, statusCode), nil
 	}
 
 	// 解析Nova响应

relay/channel/claude/constants.go

@@ -21,6 +21,8 @@ var ModelList = []string{
 	"claude-opus-4-1-20250805-thinking",
 	"claude-sonnet-4-5-20250929",
 	"claude-sonnet-4-5-20250929-thinking",
+	"claude-opus-4-5-20251101",
+	"claude-opus-4-5-20251101-thinking",
 }
 
 var ChannelName = "claude"

relay/channel/claude/relay-claude.go

@@ -673,7 +673,7 @@ func HandleStreamResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 func HandleStreamFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, claudeInfo *ClaudeResponseInfo, requestMode int) {
 
 	if requestMode == RequestModeCompletion {
-		claudeInfo.Usage = service.ResponseText2Usage(claudeInfo.ResponseText.String(), info.UpstreamModelName, info.PromptTokens)
+		claudeInfo.Usage = service.ResponseText2Usage(c, claudeInfo.ResponseText.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
 	} else {
 		if claudeInfo.Usage.PromptTokens == 0 {
 			//上游出错
@@ -682,7 +682,7 @@ func HandleStreamFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, clau
 			if common.DebugEnabled {
 				common.SysLog("claude response usage is not complete, maybe upstream error")
 			}
-			claudeInfo.Usage = service.ResponseText2Usage(claudeInfo.ResponseText.String(), info.UpstreamModelName, claudeInfo.Usage.PromptTokens)
+			claudeInfo.Usage = service.ResponseText2Usage(c, claudeInfo.ResponseText.String(), info.UpstreamModelName, claudeInfo.Usage.PromptTokens)
 		}
 	}
 
@@ -734,10 +734,7 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 		return types.WithClaudeError(*claudeError, http.StatusInternalServerError)
 	}
 	if requestMode == RequestModeCompletion {
-		completionTokens := service.CountTextToken(claudeResponse.Completion, info.OriginModelName)
-		claudeInfo.Usage.PromptTokens = info.PromptTokens
-		claudeInfo.Usage.CompletionTokens = completionTokens
-		claudeInfo.Usage.TotalTokens = info.PromptTokens + completionTokens
+		claudeInfo.Usage = service.ResponseText2Usage(c, claudeResponse.Completion, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	} else {
 		claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
 		claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens

relay/channel/cloudflare/relay_cloudflare.go

@@ -74,7 +74,7 @@ func cfStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Res
 	if err := scanner.Err(); err != nil {
 		logger.LogError(c, "error_scanning_stream_response: "+err.Error())
 	}
-	usage := service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+	usage := service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	if info.ShouldIncludeUsage {
 		response := helper.GenerateFinalUsageResponse(id, info.StartTime.Unix(), info.UpstreamModelName, *usage)
 		err := helper.ObjectData(c, response)
@@ -105,7 +105,7 @@ func cfHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response)
 	for _, choice := range response.Choices {
 		responseText += choice.Message.StringContent()
 	}
-	usage := service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+	usage := service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	response.Usage = *usage
 	response.Id = helper.GetResponseID(c)
 	jsonResponse, err := json.Marshal(response)
@@ -142,10 +142,6 @@ func cfSTTHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respon
 	c.Writer.WriteHeader(resp.StatusCode)
 	_, _ = c.Writer.Write(jsonResponse)
 
-	usage := &dto.Usage{}
-	usage.PromptTokens = info.PromptTokens
-	usage.CompletionTokens = service.CountTextToken(cfResp.Result.Text, info.UpstreamModelName)
-	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
-
+	usage := service.ResponseText2Usage(c, cfResp.Result.Text, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	return nil, usage
 }

relay/channel/cohere/relay-cohere.go

@@ -165,7 +165,7 @@ func cohereStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 		}
 	})
 	if usage.PromptTokens == 0 {
-		usage = service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	}
 	return usage, nil
 }
@@ -225,9 +225,9 @@ func cohereRerankHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 	}
 	usage := dto.Usage{}
 	if cohereResp.Meta.BilledUnits.InputTokens == 0 {
-		usage.PromptTokens = info.PromptTokens
+		usage.PromptTokens = info.GetEstimatePromptTokens()
 		usage.CompletionTokens = 0
-		usage.TotalTokens = info.PromptTokens
+		usage.TotalTokens = info.GetEstimatePromptTokens()
 	} else {
 		usage.PromptTokens = cohereResp.Meta.BilledUnits.InputTokens
 		usage.CompletionTokens = cohereResp.Meta.BilledUnits.OutputTokens

relay/channel/coze/relay-coze.go

@@ -142,7 +142,7 @@ func cozeChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *ht
 	helper.Done(c)
 
 	if usage.TotalTokens == 0 {
-		usage = service.ResponseText2Usage(responseText, info.UpstreamModelName, c.GetInt("coze_input_count"))
+		usage = service.ResponseText2Usage(c, responseText, info.UpstreamModelName, c.GetInt("coze_input_count"))
 	}
 
 	return usage, nil

relay/channel/dify/relay-dify.go

@@ -246,7 +246,7 @@ func difyStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.R
 	})
 	helper.Done(c)
 	if usage.TotalTokens == 0 {
-		usage = service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	}
 	usage.CompletionTokens += nodeToken
 	return usage, nil

relay/channel/gemini/adaptor.go

@@ -137,6 +137,8 @@ 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 != "" {
+			info.UpstreamModelName = baseModel
 		}
 	}
 
@@ -177,7 +179,7 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 		return nil, errors.New("request is nil")
 	}
 
-	geminiRequest, err := CovertGemini2OpenAI(c, *request, info)
+	geminiRequest, err := CovertOpenAI2Gemini(c, *request, info)
 	if err != nil {
 		return nil, err
 	}

relay/channel/gemini/constant.go

@@ -8,6 +8,7 @@ var ModelList = []string{
 	"gemini-1.5-pro-latest", "gemini-1.5-flash-latest",
 	// preview version
 	"gemini-2.0-flash-lite-preview",
+	"gemini-3-pro-preview",
 	// gemini exp
 	"gemini-exp-1206",
 	// flash exp
@@ -31,7 +32,7 @@ var SafetySettingList = []string{
 	"HARM_CATEGORY_HATE_SPEECH",
 	"HARM_CATEGORY_SEXUALLY_EXPLICIT",
 	"HARM_CATEGORY_DANGEROUS_CONTENT",
-	"HARM_CATEGORY_CIVIC_INTEGRITY",
+	//"HARM_CATEGORY_CIVIC_INTEGRITY", This item is deprecated!
 }
 
 var ChannelName = "google gemini"

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

@@ -3,7 +3,6 @@ package gemini
 import (
 	"io"
 	"net/http"
-	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/dto"
@@ -13,8 +12,6 @@ import (
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/types"
 
-	"github.com/pkg/errors"
-
 	"github.com/gin-gonic/gin"
 )
 
@@ -72,10 +69,7 @@ func NativeGeminiEmbeddingHandler(c *gin.Context, resp *http.Response, info *rel
 		println(string(responseBody))
 	}
 
-	usage := &dto.Usage{
-		PromptTokens: info.PromptTokens,
-		TotalTokens:  info.PromptTokens,
-	}
+	usage := service.ResponseText2Usage(c, "", info.UpstreamModelName, info.GetEstimatePromptTokens())
 
 	if info.IsGeminiBatchEmbedding {
 		var geminiResponse dto.GeminiBatchEmbeddingResponse
@@ -97,80 +91,15 @@ func NativeGeminiEmbeddingHandler(c *gin.Context, resp *http.Response, info *rel
 }
 
 func GeminiTextGenerationStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	var usage = &dto.Usage{}
-	var imageCount int
-
 	helper.SetEventStreamHeaders(c)
 
-	responseText := strings.Builder{}
-
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
-		var geminiResponse dto.GeminiChatResponse
-		err := common.UnmarshalJsonStr(data, &geminiResponse)
-		if err != nil {
-			logger.LogError(c, "error unmarshalling stream response: "+err.Error())
-			return false
-		}
-
-		// 统计图片数量
-		for _, candidate := range geminiResponse.Candidates {
-			for _, part := range candidate.Content.Parts {
-				if part.InlineData != nil && part.InlineData.MimeType != "" {
-					imageCount++
-				}
-				if part.Text != "" {
-					responseText.WriteString(part.Text)
-				}
-			}
-		}
-
-		// 更新使用量统计
-		if geminiResponse.UsageMetadata.TotalTokenCount != 0 {
-			usage.PromptTokens = geminiResponse.UsageMetadata.PromptTokenCount
-			usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount + geminiResponse.UsageMetadata.ThoughtsTokenCount
-			usage.TotalTokens = geminiResponse.UsageMetadata.TotalTokenCount
-			usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
-			for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
-				if detail.Modality == "AUDIO" {
-					usage.PromptTokensDetails.AudioTokens = detail.TokenCount
-				} else if detail.Modality == "TEXT" {
-					usage.PromptTokensDetails.TextTokens = detail.TokenCount
-				}
-			}
-		}
-
+	return geminiStreamHandler(c, info, resp, func(data string, geminiResponse *dto.GeminiChatResponse) bool {
 		// 直接发送 GeminiChatResponse 响应
-		err = helper.StringData(c, data)
+		err := helper.StringData(c, data)
 		if err != nil {
 			logger.LogError(c, err.Error())
 		}
 		info.SendResponseCount++
 		return true
 	})
-
-	if info.SendResponseCount == 0 {
-		return nil, types.NewOpenAIError(errors.New("no response received from Gemini API"), types.ErrorCodeEmptyResponse, http.StatusInternalServerError)
-	}
-
-	if imageCount != 0 {
-		if usage.CompletionTokens == 0 {
-			usage.CompletionTokens = imageCount * 258
-		}
-	}
-
-	// 如果usage.CompletionTokens为0，则使用本地统计的completion tokens
-	if usage.CompletionTokens == 0 {
-		str := responseText.String()
-		if len(str) > 0 {
-			usage = service.ResponseText2Usage(responseText.String(), info.UpstreamModelName, info.PromptTokens)
-		} else {
-			// 空补全，不需要使用量
-			usage = &dto.Usage{}
-		}
-	}
-
-	// 移除流式响应结尾的[Done]，因为Gemini API没有发送Done的行为
-	//helper.Done(c)
-
-	return usage, nil
 }

relay/channel/gemini/relay-gemini.go

@@ -19,8 +19,8 @@ import (
 	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
 	"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"
 )
 
@@ -44,6 +44,8 @@ var geminiSupportedMimeTypes = map[string]bool{
 	"video/flv":       true,
 }
 
+const thoughtSignatureBypassValue = "context_engineering_is_the_way_to_go"
+
 // Gemini 允许的思考预算范围
 const (
 	pro25MinBudget       = 128
@@ -120,6 +122,14 @@ func clampThinkingBudgetByEffort(modelName string, effort string) int {
 	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
@@ -176,12 +186,18 @@ func ThinkingAdaptor(geminiRequest *dto.GeminiChatRequest, info *relaycommon.Rel
 					ThinkingBudget: common.GetPointer(0),
 				}
 			}
+		} else if _, level := parseThinkingLevelSuffix(modelName); level != "" {
+			geminiRequest.GenerationConfig.ThinkingConfig = &dto.GeminiThinkingConfig{
+				IncludeThoughts: true,
+				ThinkingLevel:   level,
+			}
+			info.ReasoningEffort = level
 		}
 	}
 }
 
 // Setting safety to the lowest possible values since Gemini is already powerless enough
-func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, info *relaycommon.RelayInfo) (*dto.GeminiChatRequest, error) {
+func CovertOpenAI2Gemini(c *gin.Context, textRequest dto.GeneralOpenAIRequest, info *relaycommon.RelayInfo) (*dto.GeminiChatRequest, error) {
 
 	geminiRequest := dto.GeminiChatRequest{
 		Contents: make([]dto.GeminiChatContent, 0, len(textRequest.Messages)),
@@ -193,6 +209,10 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 		},
 	}
 
+	attachThoughtSignature := (info.ChannelType == constant.ChannelTypeGemini ||
+		info.ChannelType == constant.ChannelTypeVertexAi) &&
+		model_setting.GetGeminiSettings().FunctionCallThoughtSignatureEnabled
+
 	if model_setting.IsGeminiModelSupportImagine(info.UpstreamModelName) {
 		geminiRequest.GenerationConfig.ResponseModalities = []string{
 			"TEXT",
@@ -202,6 +222,7 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 
 	adaptorWithExtraBody := false
 
+	// patch extra_body
 	if len(textRequest.ExtraBody) > 0 {
 		if !strings.HasSuffix(info.UpstreamModelName, "-nothinking") {
 			var extraBody map[string]interface{}
@@ -233,6 +254,39 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 						}
 					}
 				}
+
+				// check error param name like imageConfig, should be image_config
+				if _, hasErrorParam := googleBody["imageConfig"]; hasErrorParam {
+					return nil, errors.New("extra_body.google.imageConfig is not supported, use extra_body.google.image_config instead")
+				}
+
+				if imageConfig, ok := googleBody["image_config"].(map[string]interface{}); ok {
+					// check error param name like aspectRatio, should be aspect_ratio
+					if _, hasErrorParam := imageConfig["aspectRatio"]; hasErrorParam {
+						return nil, errors.New("extra_body.google.image_config.aspectRatio is not supported, use extra_body.google.image_config.aspect_ratio instead")
+					}
+					// check error param name like imageSize, should be image_size
+					if _, hasErrorParam := imageConfig["imageSize"]; hasErrorParam {
+						return nil, errors.New("extra_body.google.image_config.imageSize is not supported, use extra_body.google.image_config.image_size instead")
+					}
+
+					// convert snake_case to camelCase for Gemini API
+					geminiImageConfig := make(map[string]interface{})
+					if aspectRatio, ok := imageConfig["aspect_ratio"]; ok {
+						geminiImageConfig["aspectRatio"] = aspectRatio
+					}
+					if imageSize, ok := imageConfig["image_size"]; ok {
+						geminiImageConfig["imageSize"] = imageSize
+					}
+
+					if len(geminiImageConfig) > 0 {
+						imageConfigBytes, err := common.Marshal(geminiImageConfig)
+						if err != nil {
+							return nil, fmt.Errorf("failed to marshal image_config: %w", err)
+						}
+						geminiRequest.GenerationConfig.ImageConfig = imageConfigBytes
+					}
+				}
 			}
 		}
 	}
@@ -371,6 +425,8 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 		content := dto.GeminiChatContent{
 			Role: message.Role,
 		}
+		shouldAttachThoughtSignature := attachThoughtSignature && (message.Role == "assistant" || message.Role == "model")
+		signatureAttached := false
 		// isToolCall := false
 		if message.ToolCalls != nil {
 			// message.Role = "model"
@@ -388,6 +444,10 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 						Arguments:    args,
 					},
 				}
+				if shouldAttachThoughtSignature && !signatureAttached && hasFunctionCallContent(toolCall.FunctionCall) && len(toolCall.ThoughtSignature) == 0 {
+					toolCall.ThoughtSignature = json.RawMessage(strconv.Quote(thoughtSignatureBypassValue))
+					signatureAttached = true
+				}
 				parts = append(parts, toolCall)
 				tool_call_ids[call.ID] = call.Function.Name
 			}
@@ -400,9 +460,68 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 				if part.Text == "" {
 					continue
 				}
-				parts = append(parts, dto.GeminiPart{
-					Text: part.Text,
-				})
+				// check markdown image ![image](data:image/jpeg;base64,xxxxxxxxxxxx)
+				// 使用字符串查找而非正则，避免大文本性能问题
+				text := part.Text
+				hasMarkdownImage := false
+				for {
+					// 快速检查是否包含 markdown 图片标记
+					startIdx := strings.Index(text, "![")
+					if startIdx == -1 {
+						break
+					}
+					// 找到 ](
+					bracketIdx := strings.Index(text[startIdx:], "](data:")
+					if bracketIdx == -1 {
+						break
+					}
+					bracketIdx += startIdx
+					// 找到闭合的 )
+					closeIdx := strings.Index(text[bracketIdx+2:], ")")
+					if closeIdx == -1 {
+						break
+					}
+					closeIdx += bracketIdx + 2
+
+					hasMarkdownImage = true
+					// 添加图片前的文本
+					if startIdx > 0 {
+						textBefore := text[:startIdx]
+						if textBefore != "" {
+							parts = append(parts, dto.GeminiPart{
+								Text: textBefore,
+							})
+						}
+					}
+					// 提取 data URL (从 "](" 后面开始，到 ")" 之前)
+					dataUrl := text[bracketIdx+2 : closeIdx]
+					imageNum += 1
+					if constant.GeminiVisionMaxImageNum != -1 && imageNum > constant.GeminiVisionMaxImageNum {
+						return nil, fmt.Errorf("too many images in the message, max allowed is %d", constant.GeminiVisionMaxImageNum)
+					}
+					format, base64String, err := service.DecodeBase64FileData(dataUrl)
+					if err != nil {
+						return nil, fmt.Errorf("decode markdown base64 image data failed: %s", err.Error())
+					}
+					imgPart := dto.GeminiPart{
+						InlineData: &dto.GeminiInlineData{
+							MimeType: format,
+							Data:     base64String,
+						},
+					}
+					if shouldAttachThoughtSignature {
+						imgPart.ThoughtSignature = json.RawMessage(strconv.Quote(thoughtSignatureBypassValue))
+					}
+					parts = append(parts, imgPart)
+					// 继续处理剩余文本
+					text = text[closeIdx+1:]
+				}
+				// 添加剩余文本或原始文本（如果没有找到 markdown 图片）
+				if !hasMarkdownImage {
+					parts = append(parts, dto.GeminiPart{
+						Text: part.Text,
+					})
+				}
 			} else if part.Type == dto.ContentTypeImageURL {
 				imageNum += 1
 
@@ -472,6 +591,17 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 			}
 		}
 
+		// 如果需要附加签名但还没有附加（没有 tool_calls 或 tool_calls 为空），
+		// 则在第一个文本 part 上附加 thoughtSignature
+		if shouldAttachThoughtSignature && !signatureAttached && len(parts) > 0 {
+			for i := range parts {
+				if parts[i].Text != "" {
+					parts[i].ThoughtSignature = json.RawMessage(strconv.Quote(thoughtSignatureBypassValue))
+					break
+				}
+			}
+		}
+
 		content.Parts = parts
 
 		// there's no assistant role in gemini and API shall vomit if Role is not user or model
@@ -496,6 +626,28 @@ func CovertGemini2OpenAI(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 	return &geminiRequest, nil
 }
 
+func hasFunctionCallContent(call *dto.FunctionCall) bool {
+	if call == nil {
+		return false
+	}
+	if strings.TrimSpace(call.FunctionName) != "" {
+		return true
+	}
+
+	switch v := call.Arguments.(type) {
+	case nil:
+		return false
+	case string:
+		return strings.TrimSpace(v) != ""
+	case map[string]interface{}:
+		return len(v) > 0
+	case []interface{}:
+		return len(v) > 0
+	default:
+		return true
+	}
+}
+
 // Helper function to get a list of supported MIME types for error messages
 func getSupportedMimeTypesList() []string {
 	keys := make([]string, 0, len(geminiSupportedMimeTypes))
@@ -920,14 +1072,10 @@ func handleFinalStream(c *gin.Context, info *relaycommon.RelayInfo, resp *dto.Ch
 	return nil
 }
 
-func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	// responseText := ""
-	id := helper.GetResponseID(c)
-	createAt := common.GetTimestamp()
-	responseText := strings.Builder{}
+func geminiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response, callback func(data string, geminiResponse *dto.GeminiChatResponse) bool) (*dto.Usage, *types.NewAPIError) {
 	var usage = &dto.Usage{}
 	var imageCount int
-	finishReason := constant.FinishReasonStop
+	responseText := strings.Builder{}
 
 	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
 		var geminiResponse dto.GeminiChatResponse
@@ -937,6 +1085,7 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 			return false
 		}
 
+		// 统计图片数量
 		for _, candidate := range geminiResponse.Candidates {
 			for _, part := range candidate.Content.Parts {
 				if part.InlineData != nil && part.InlineData.MimeType != "" {
@@ -948,14 +1097,10 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 			}
 		}
 
-		response, isStop := streamResponseGeminiChat2OpenAI(&geminiResponse)
-
-		response.Id = id
-		response.Created = createAt
-		response.Model = info.UpstreamModelName
+		// 更新使用量统计
 		if geminiResponse.UsageMetadata.TotalTokenCount != 0 {
 			usage.PromptTokens = geminiResponse.UsageMetadata.PromptTokenCount
-			usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount
+			usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount + geminiResponse.UsageMetadata.ThoughtsTokenCount
 			usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
 			usage.TotalTokens = geminiResponse.UsageMetadata.TotalTokenCount
 			for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
@@ -966,6 +1111,45 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 				}
 			}
 		}
+
+		return callback(data, &geminiResponse)
+	})
+
+	if imageCount != 0 {
+		if usage.CompletionTokens == 0 {
+			usage.CompletionTokens = imageCount * 1400
+		}
+	}
+
+	usage.PromptTokensDetails.TextTokens = usage.PromptTokens
+	if usage.TotalTokens > 0 {
+		usage.CompletionTokens = usage.TotalTokens - usage.PromptTokens
+	}
+
+	if usage.CompletionTokens <= 0 {
+		str := responseText.String()
+		if len(str) > 0 {
+			usage = service.ResponseText2Usage(c, responseText.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
+		} else {
+			usage = &dto.Usage{}
+		}
+	}
+
+	return usage, nil
+}
+
+func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
+	id := helper.GetResponseID(c)
+	createAt := common.GetTimestamp()
+	finishReason := constant.FinishReasonStop
+
+	usage, err := geminiStreamHandler(c, info, resp, func(data string, geminiResponse *dto.GeminiChatResponse) bool {
+		response, isStop := streamResponseGeminiChat2OpenAI(geminiResponse)
+
+		response.Id = id
+		response.Created = createAt
+		response.Model = info.UpstreamModelName
+
 		logger.LogDebug(c, fmt.Sprintf("info.SendResponseCount = %d", info.SendResponseCount))
 		if info.SendResponseCount == 0 {
 			// send first response
@@ -981,7 +1165,7 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 					emptyResponse.Choices[0].Delta.ToolCalls = copiedToolCalls
 				}
 				finishReason = constant.FinishReasonToolCalls
-				err = handleStream(c, info, emptyResponse)
+				err := handleStream(c, info, emptyResponse)
 				if err != nil {
 					logger.LogError(c, err.Error())
 				}
@@ -991,14 +1175,14 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 					response.Choices[0].FinishReason = nil
 				}
 			} else {
-				err = handleStream(c, info, emptyResponse)
+				err := handleStream(c, info, emptyResponse)
 				if err != nil {
 					logger.LogError(c, err.Error())
 				}
 			}
 		}
 
-		err = handleStream(c, info, response)
+		err := handleStream(c, info, response)
 		if err != nil {
 			logger.LogError(c, err.Error())
 		}
@@ -1008,40 +1192,15 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 		return true
 	})
 
-	if info.SendResponseCount == 0 {
-		// 空补全，报错不计费
-		// empty response, throw an error
-		return nil, types.NewOpenAIError(errors.New("no response received from Gemini API"), types.ErrorCodeEmptyResponse, http.StatusInternalServerError)
-	}
-
-	if imageCount != 0 {
-		if usage.CompletionTokens == 0 {
-			usage.CompletionTokens = imageCount * 258
-		}
-	}
-
-	usage.PromptTokensDetails.TextTokens = usage.PromptTokens
-	usage.CompletionTokens = usage.TotalTokens - usage.PromptTokens
-
-	if usage.CompletionTokens == 0 {
-		str := responseText.String()
-		if len(str) > 0 {
-			usage = service.ResponseText2Usage(responseText.String(), info.UpstreamModelName, info.PromptTokens)
-		} else {
-			// 空补全，不需要使用量
-			usage = &dto.Usage{}
-		}
+	if err != nil {
+		return usage, err
 	}
 
 	response := helper.GenerateFinalUsageResponse(id, createAt, info.UpstreamModelName, *usage)
-	err := handleFinalStream(c, info, response)
-	if err != nil {
-		common.SysLog("send final response failed: " + err.Error())
+	handleErr := handleFinalStream(c, info, response)
+	if handleErr != nil {
+		common.SysLog("send final response failed: " + handleErr.Error())
 	}
-	//if info.RelayFormat == relaycommon.RelayFormatOpenAI {
-	//	helper.Done(c)
-	//}
-	//resp.Body.Close()
 	return usage, nil
 }
 
@@ -1143,11 +1302,7 @@ func GeminiEmbeddingHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *h
 	// Google has not yet clarified how embedding models will be billed
 	// refer to openai billing method to use input tokens billing
 	// https://platform.openai.com/docs/guides/embeddings#what-are-embeddings
-	usage := &dto.Usage{
-		PromptTokens:     info.PromptTokens,
-		CompletionTokens: 0,
-		TotalTokens:      info.PromptTokens,
-	}
+	usage := service.ResponseText2Usage(c, "", info.UpstreamModelName, info.GetEstimatePromptTokens())
 	openAIResponse.Usage = *usage
 
 	jsonResponse, jsonErr := common.Marshal(openAIResponse)

relay/channel/minimax/tts.go

@@ -163,7 +163,7 @@ func handleTTSResponse(c *gin.Context, resp *http.Response, info *relaycommon.Re
 	}
 
 	usage = &dto.Usage{
-		PromptTokens:     info.PromptTokens,
+		PromptTokens:     info.GetEstimatePromptTokens(),
 		CompletionTokens: 0,
 		TotalTokens:      int(minimaxResp.ExtraInfo.UsageCharacters),
 	}

relay/channel/moonshot/adaptor.go

@@ -6,6 +6,7 @@ import (
 	"io"
 	"net/http"
 
+	channelconstant "github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/relay/channel"
 	"github.com/QuantumNous/new-api/relay/channel/claude"
@@ -44,6 +45,16 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	baseURL := info.ChannelBaseUrl
+	if specialPlan, ok := channelconstant.ChannelSpecialBases[baseURL]; ok {
+		if info.RelayFormat == types.RelayFormatClaude {
+			return fmt.Sprintf("%s/v1/messages", specialPlan.ClaudeBaseURL), nil
+		}
+		if info.RelayFormat == types.RelayFormatOpenAI {
+			return fmt.Sprintf("%s/chat/completions", specialPlan.OpenAIBaseURL), nil
+		}
+	}
+
 	switch info.RelayFormat {
 	case types.RelayFormatClaude:
 		return fmt.Sprintf("%s/anthropic/v1/messages", info.ChannelBaseUrl), nil

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"}
+	effortSuffixes := []string{"-high", "-minimal", "-low", "-medium", "-none"}
 	for _, suffix := range effortSuffixes {
 		if strings.HasSuffix(model, suffix) {
 			effort := strings.TrimPrefix(suffix, "-")

relay/channel/openai/relay-openai.go

@@ -183,7 +183,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 	}
 
 	if !containStreamUsage {
-		usage = service.ResponseText2Usage(responseTextBuilder.String(), info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(c, responseTextBuilder.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
 		usage.CompletionTokens += toolCount * 7
 	}
 
@@ -245,9 +245,9 @@ func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 			}
 		}
 		simpleResponse.Usage = dto.Usage{
-			PromptTokens:     info.PromptTokens,
+			PromptTokens:     info.GetEstimatePromptTokens(),
 			CompletionTokens: completionTokens,
-			TotalTokens:      info.PromptTokens + completionTokens,
+			TotalTokens:      info.GetEstimatePromptTokens() + completionTokens,
 		}
 		usageModified = true
 	}
@@ -336,8 +336,8 @@ func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	// and can be terminated directly.
 	defer service.CloseResponseBodyGracefully(resp)
 	usage := &dto.Usage{}
-	usage.PromptTokens = info.PromptTokens
-	usage.TotalTokens = info.PromptTokens
+	usage.PromptTokens = info.GetEstimatePromptTokens()
+	usage.TotalTokens = info.GetEstimatePromptTokens()
 	for k, v := range resp.Header {
 		c.Writer.Header().Set(k, v[0])
 	}
@@ -383,7 +383,7 @@ func OpenaiSTTHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	}
 
 	usage := &dto.Usage{}
-	usage.PromptTokens = info.PromptTokens
+	usage.PromptTokens = info.GetEstimatePromptTokens()
 	usage.CompletionTokens = 0
 	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
 	return nil, usage

relay/channel/openai/relay_responses.go

@@ -141,7 +141,7 @@ func OaiResponsesStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp
 	}
 
 	if usage.PromptTokens == 0 && usage.CompletionTokens != 0 {
-		usage.PromptTokens = info.PromptTokens
+		usage.PromptTokens = info.GetEstimatePromptTokens()
 	}
 
 	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens

relay/channel/palm/adaptor.go

@@ -81,7 +81,7 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	if info.IsStream {
 		var responseText string
 		err, responseText = palmStreamHandler(c, resp)
-		usage = service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	} else {
 		usage, err = palmHandler(c, info, resp)
 	}

relay/channel/palm/relay-palm.go

@@ -121,13 +121,8 @@ func palmHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respons
 		}, resp.StatusCode)
 	}
 	fullTextResponse := responsePaLM2OpenAI(&palmResponse)
-	completionTokens := service.CountTextToken(palmResponse.Candidates[0].Content, info.UpstreamModelName)
-	usage := dto.Usage{
-		PromptTokens:     info.PromptTokens,
-		CompletionTokens: completionTokens,
-		TotalTokens:      info.PromptTokens + completionTokens,
-	}
-	fullTextResponse.Usage = usage
+	usage := service.ResponseText2Usage(c, palmResponse.Candidates[0].Content, info.UpstreamModelName, info.GetEstimatePromptTokens())
+	fullTextResponse.Usage = *usage
 	jsonResponse, err := common.Marshal(fullTextResponse)
 	if err != nil {
 		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
@@ -135,5 +130,5 @@ func palmHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respons
 	c.Writer.Header().Set("Content-Type", "application/json")
 	c.Writer.WriteHeader(resp.StatusCode)
 	service.IOCopyBytesGracefully(c, resp, jsonResponse)
-	return &usage, nil
+	return usage, nil
 }

relay/channel/task/hailuo/adaptor.go

@@ -0,0 +1,297 @@
+package hailuo
+
+import (
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/model"
+	"github.com/gin-gonic/gin"
+	"github.com/pkg/errors"
+
+	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/relay/channel"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/QuantumNous/new-api/service"
+)
+
+// https://platform.minimaxi.com/docs/api-reference/video-generation-intro
+type TaskAdaptor struct {
+	ChannelType int
+	apiKey      string
+	baseURL     string
+}
+
+func (a *TaskAdaptor) Init(info *relaycommon.RelayInfo) {
+	a.ChannelType = info.ChannelType
+	a.baseURL = info.ChannelBaseUrl
+	a.apiKey = info.ApiKey
+}
+
+func (a *TaskAdaptor) ValidateRequestAndSetAction(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.TaskError) {
+	return relaycommon.ValidateBasicTaskRequest(c, info, constant.TaskActionGenerate)
+}
+
+func (a *TaskAdaptor) BuildRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	return fmt.Sprintf("%s%s", a.baseURL, TextToVideoEndpoint), nil
+}
+
+func (a *TaskAdaptor) BuildRequestHeader(c *gin.Context, req *http.Request, info *relaycommon.RelayInfo) error {
+	req.Header.Set("Content-Type", "application/json")
+	req.Header.Set("Accept", "application/json")
+	req.Header.Set("Authorization", "Bearer "+a.apiKey)
+	return nil
+}
+
+func (a *TaskAdaptor) BuildRequestBody(c *gin.Context, info *relaycommon.RelayInfo) (io.Reader, error) {
+	v, exists := c.Get("task_request")
+	if !exists {
+		return nil, fmt.Errorf("request not found in context")
+	}
+	req, ok := v.(relaycommon.TaskSubmitReq)
+	if !ok {
+		return nil, fmt.Errorf("invalid request type in context")
+	}
+
+	body, err := a.convertToRequestPayload(&req)
+	if err != nil {
+		return nil, errors.Wrap(err, "convert request payload failed")
+	}
+
+	data, err := json.Marshal(body)
+	if err != nil {
+		return nil, err
+	}
+
+	return bytes.NewReader(data), nil
+}
+
+func (a *TaskAdaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (*http.Response, error) {
+	return channel.DoTaskApiRequest(a, c, info, requestBody)
+}
+
+func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (taskID string, taskData []byte, taskErr *dto.TaskError) {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		taskErr = service.TaskErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError)
+		return
+	}
+	_ = resp.Body.Close()
+
+	var hResp VideoResponse
+	if err := json.Unmarshal(responseBody, &hResp); err != nil {
+		taskErr = service.TaskErrorWrapper(errors.Wrapf(err, "body: %s", responseBody), "unmarshal_response_body_failed", http.StatusInternalServerError)
+		return
+	}
+
+	if hResp.BaseResp.StatusCode != StatusSuccess {
+		taskErr = service.TaskErrorWrapper(
+			fmt.Errorf("hailuo api error: %s", hResp.BaseResp.StatusMsg),
+			strconv.Itoa(hResp.BaseResp.StatusCode),
+			http.StatusBadRequest,
+		)
+		return
+	}
+
+	ov := dto.NewOpenAIVideo()
+	ov.ID = hResp.TaskID
+	ov.TaskID = hResp.TaskID
+	ov.CreatedAt = time.Now().Unix()
+	ov.Model = info.OriginModelName
+
+	c.JSON(http.StatusOK, ov)
+	return hResp.TaskID, responseBody, nil
+}
+
+func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any) (*http.Response, error) {
+	taskID, ok := body["task_id"].(string)
+	if !ok {
+		return nil, fmt.Errorf("invalid task_id")
+	}
+
+	uri := fmt.Sprintf("%s%s?task_id=%s", baseUrl, QueryTaskEndpoint, taskID)
+
+	req, err := http.NewRequest(http.MethodGet, uri, nil)
+	if err != nil {
+		return nil, err
+	}
+
+	req.Header.Set("Accept", "application/json")
+	req.Header.Set("Authorization", "Bearer "+key)
+
+	return service.GetHttpClient().Do(req)
+}
+
+func (a *TaskAdaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *TaskAdaptor) GetChannelName() string {
+	return ChannelName
+}
+
+func (a *TaskAdaptor) convertToRequestPayload(req *relaycommon.TaskSubmitReq) (*VideoRequest, error) {
+	modelConfig := GetModelConfig(req.Model)
+	duration := DefaultDuration
+	if req.Duration > 0 {
+		duration = req.Duration
+	}
+	resolution := modelConfig.DefaultResolution
+	if req.Size != "" {
+		resolution = a.parseResolutionFromSize(req.Size, modelConfig)
+	}
+
+	videoRequest := &VideoRequest{
+		Model:      req.Model,
+		Prompt:     req.Prompt,
+		Duration:   &duration,
+		Resolution: resolution,
+	}
+	if err := req.UnmarshalMetadata(&videoRequest); err != nil {
+		return nil, errors.Wrap(err, "unmarshal metadata to video request failed")
+	}
+
+	return videoRequest, nil
+}
+
+func (a *TaskAdaptor) parseResolutionFromSize(size string, modelConfig ModelConfig) string {
+	switch {
+	case strings.Contains(size, "1080"):
+		return Resolution1080P
+	case strings.Contains(size, "768"):
+		return Resolution768P
+	case strings.Contains(size, "720"):
+		return Resolution720P
+	case strings.Contains(size, "512"):
+		return Resolution512P
+	default:
+		return modelConfig.DefaultResolution
+	}
+}
+
+func (a *TaskAdaptor) ParseTaskResult(respBody []byte) (*relaycommon.TaskInfo, error) {
+	resTask := QueryTaskResponse{}
+	if err := json.Unmarshal(respBody, &resTask); err != nil {
+		return nil, errors.Wrap(err, "unmarshal task result failed")
+	}
+
+	taskResult := relaycommon.TaskInfo{}
+
+	if resTask.BaseResp.StatusCode == StatusSuccess {
+		taskResult.Code = 0
+	} else {
+		taskResult.Code = resTask.BaseResp.StatusCode
+		taskResult.Reason = resTask.BaseResp.StatusMsg
+		taskResult.Status = model.TaskStatusFailure
+		taskResult.Progress = "100%"
+	}
+
+	switch resTask.Status {
+	case TaskStatusPreparing, TaskStatusQueueing, TaskStatusProcessing:
+		taskResult.Status = model.TaskStatusInProgress
+		taskResult.Progress = "30%"
+		if resTask.Status == TaskStatusProcessing {
+			taskResult.Progress = "50%"
+		}
+	case TaskStatusSuccess:
+		taskResult.Status = model.TaskStatusSuccess
+		taskResult.Progress = "100%"
+		taskResult.Url = a.buildVideoURL(resTask.TaskID, resTask.FileID)
+	case TaskStatusFailed:
+		taskResult.Status = model.TaskStatusFailure
+		taskResult.Progress = "100%"
+		if taskResult.Reason == "" {
+			taskResult.Reason = "task failed"
+		}
+	default:
+		taskResult.Status = model.TaskStatusInProgress
+		taskResult.Progress = "30%"
+	}
+
+	return &taskResult, nil
+}
+
+func (a *TaskAdaptor) ConvertToOpenAIVideo(originTask *model.Task) ([]byte, error) {
+	var hailuoResp QueryTaskResponse
+	if err := json.Unmarshal(originTask.Data, &hailuoResp); err != nil {
+		return nil, errors.Wrap(err, "unmarshal hailuo task data failed")
+	}
+
+	openAIVideo := originTask.ToOpenAIVideo()
+	if hailuoResp.BaseResp.StatusCode != StatusSuccess {
+		openAIVideo.Error = &dto.OpenAIVideoError{
+			Message: hailuoResp.BaseResp.StatusMsg,
+			Code:    strconv.Itoa(hailuoResp.BaseResp.StatusCode),
+		}
+	}
+
+	jsonData, err := common.Marshal(openAIVideo)
+	if err != nil {
+		return nil, errors.Wrap(err, "marshal openai video failed")
+	}
+
+	return jsonData, nil
+}
+
+func (a *TaskAdaptor) buildVideoURL(_, fileID string) string {
+	if a.apiKey == "" || a.baseURL == "" {
+		return ""
+	}
+
+	url := fmt.Sprintf("%s/v1/files/retrieve?file_id=%s", a.baseURL, fileID)
+
+	req, err := http.NewRequest(http.MethodGet, url, nil)
+	if err != nil {
+		return ""
+	}
+
+	req.Header.Set("Accept", "application/json")
+	req.Header.Set("Authorization", "Bearer "+a.apiKey)
+
+	resp, err := service.GetHttpClient().Do(req)
+	if err != nil {
+		return ""
+	}
+	defer resp.Body.Close()
+
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return ""
+	}
+
+	var retrieveResp RetrieveFileResponse
+	if err := json.Unmarshal(responseBody, &retrieveResp); err != nil {
+		return ""
+	}
+
+	if retrieveResp.BaseResp.StatusCode != StatusSuccess {
+		return ""
+	}
+
+	return retrieveResp.File.DownloadURL
+}
+
+func contains(slice []string, item string) bool {
+	for _, s := range slice {
+		if s == item {
+			return true
+		}
+	}
+	return false
+}
+
+func containsInt(slice []int, item int) bool {
+	for _, s := range slice {
+		if s == item {
+			return true
+		}
+	}
+	return false
+}

relay/channel/task/hailuo/constants.go

@@ -0,0 +1,52 @@
+package hailuo
+
+const (
+	ChannelName = "hailuo-video"
+)
+
+var ModelList = []string{
+	"MiniMax-Hailuo-2.3",
+	"MiniMax-Hailuo-2.3-Fast",
+	"MiniMax-Hailuo-02",
+	"T2V-01-Director",
+	"T2V-01",
+	"I2V-01-Director",
+	"I2V-01-live",
+	"I2V-01",
+	"S2V-01",
+}
+
+const (
+	TextToVideoEndpoint = "/v1/video_generation"
+	QueryTaskEndpoint   = "/v1/query/video_generation"
+)
+
+const (
+	StatusSuccess    = 0
+	StatusRateLimit  = 1002
+	StatusAuthFailed = 1004
+	StatusNoBalance  = 1008
+	StatusSensitive  = 1026
+	StatusParamError = 2013
+	StatusInvalidKey = 2049
+)
+
+const (
+	TaskStatusPreparing  = "Preparing"
+	TaskStatusQueueing   = "Queueing"
+	TaskStatusProcessing = "Processing"
+	TaskStatusSuccess    = "Success"
+	TaskStatusFailed     = "Fail"
+)
+
+const (
+	Resolution512P  = "512P"
+	Resolution720P  = "720P"
+	Resolution768P  = "768P"
+	Resolution1080P = "1080P"
+)
+
+const (
+	DefaultDuration   = 6
+	DefaultResolution = Resolution720P
+)

relay/channel/task/hailuo/models.go

@@ -0,0 +1,170 @@
+package hailuo
+
+type SubjectReference struct {
+	Type  string   `json:"type"`  // Subject type, currently only supports "character"
+	Image []string `json:"image"` // Array of subject reference images (currently only supports single image)
+}
+
+type VideoRequest struct {
+	Model            string             `json:"model"`
+	Prompt           string             `json:"prompt,omitempty"`
+	PromptOptimizer  *bool              `json:"prompt_optimizer,omitempty"`
+	FastPretreatment *bool              `json:"fast_pretreatment,omitempty"`
+	Duration         *int               `json:"duration,omitempty"`
+	Resolution       string             `json:"resolution,omitempty"`
+	CallbackURL      string             `json:"callback_url,omitempty"`
+	AigcWatermark    *bool              `json:"aigc_watermark,omitempty"`
+	FirstFrameImage  string             `json:"first_frame_image,omitempty"` // For image-to-video and start-end-to-video
+	LastFrameImage   string             `json:"last_frame_image,omitempty"`  // For start-end-to-video
+	SubjectReference []SubjectReference `json:"subject_reference,omitempty"` // For subject-reference-to-video
+}
+
+type VideoResponse struct {
+	TaskID   string   `json:"task_id"`
+	BaseResp BaseResp `json:"base_resp"`
+}
+
+type BaseResp struct {
+	StatusCode int    `json:"status_code"`
+	StatusMsg  string `json:"status_msg"`
+}
+
+type QueryTaskRequest struct {
+	TaskID string `json:"task_id"`
+}
+
+type QueryTaskResponse struct {
+	TaskID      string   `json:"task_id"`
+	Status      string   `json:"status"`
+	FileID      string   `json:"file_id,omitempty"`
+	VideoWidth  int      `json:"video_width,omitempty"`
+	VideoHeight int      `json:"video_height,omitempty"`
+	BaseResp    BaseResp `json:"base_resp"`
+}
+
+type ErrorInfo struct {
+	StatusCode int    `json:"status_code"`
+	StatusMsg  string `json:"status_msg"`
+}
+
+type TaskStatusInfo struct {
+	TaskID    string `json:"task_id"`
+	Status    string `json:"status"`
+	FileID    string `json:"file_id,omitempty"`
+	VideoURL  string `json:"video_url,omitempty"`
+	ErrorCode int    `json:"error_code,omitempty"`
+	ErrorMsg  string `json:"error_msg,omitempty"`
+}
+
+type ModelConfig struct {
+	Name                 string
+	DefaultResolution    string
+	SupportedDurations   []int
+	SupportedResolutions []string
+	HasPromptOptimizer   bool
+	HasFastPretreatment  bool
+}
+
+type RetrieveFileResponse struct {
+	File     FileObject `json:"file"`
+	BaseResp BaseResp   `json:"base_resp"`
+}
+
+type FileObject struct {
+	FileID      int64  `json:"file_id"`
+	Bytes       int64  `json:"bytes"`
+	CreatedAt   int64  `json:"created_at"`
+	Filename    string `json:"filename"`
+	Purpose     string `json:"purpose"`
+	DownloadURL string `json:"download_url"`
+}
+
+func GetModelConfig(model string) ModelConfig {
+	configs := map[string]ModelConfig{
+		"MiniMax-Hailuo-2.3": {
+			Name:                 "MiniMax-Hailuo-2.3",
+			DefaultResolution:    Resolution768P,
+			SupportedDurations:   []int{6, 10},
+			SupportedResolutions: []string{Resolution768P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  true,
+		},
+		"MiniMax-Hailuo-2.3-Fast": {
+			Name:                 "MiniMax-Hailuo-2.3-Fast",
+			DefaultResolution:    Resolution768P,
+			SupportedDurations:   []int{6, 10},
+			SupportedResolutions: []string{Resolution768P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  true,
+		},
+		"MiniMax-Hailuo-02": {
+			Name:                 "MiniMax-Hailuo-02",
+			DefaultResolution:    Resolution768P,
+			SupportedDurations:   []int{6, 10},
+			SupportedResolutions: []string{Resolution512P, Resolution768P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  true,
+		},
+		"T2V-01-Director": {
+			Name:                 "T2V-01-Director",
+			DefaultResolution:    Resolution768P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution768P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+		"T2V-01": {
+			Name:                 "T2V-01",
+			DefaultResolution:    Resolution720P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution720P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+		"I2V-01-Director": {
+			Name:                 "I2V-01-Director",
+			DefaultResolution:    Resolution720P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution720P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+		"I2V-01-live": {
+			Name:                 "I2V-01-live",
+			DefaultResolution:    Resolution720P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution720P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+		"I2V-01": {
+			Name:                 "I2V-01",
+			DefaultResolution:    Resolution720P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution720P, Resolution1080P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+		"S2V-01": {
+			Name:                 "S2V-01",
+			DefaultResolution:    Resolution720P,
+			SupportedDurations:   []int{6},
+			SupportedResolutions: []string{Resolution720P},
+			HasPromptOptimizer:   true,
+			HasFastPretreatment:  false,
+		},
+	}
+
+	if config, exists := configs[model]; exists {
+		return config
+	}
+
+	return ModelConfig{
+		Name:                 model,
+		DefaultResolution:    DefaultResolution,
+		SupportedDurations:   []int{6},
+		SupportedResolutions: []string{DefaultResolution},
+		HasPromptOptimizer:   true,
+		HasFastPretreatment:  false,
+	}
+}

relay/channel/task/sora/adaptor.go

@@ -5,8 +5,10 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"strings"
 
 	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/relay/channel"
@@ -67,11 +69,30 @@ func (a *TaskAdaptor) Init(info *relaycommon.RelayInfo) {
 	a.apiKey = info.ApiKey
 }
 
+func validateRemixRequest(c *gin.Context) *dto.TaskError {
+	var req struct {
+		Prompt string `json:"prompt"`
+	}
+	if err := common.UnmarshalBodyReusable(c, &req); err != nil {
+		return service.TaskErrorWrapperLocal(err, "invalid_request", http.StatusBadRequest)
+	}
+	if strings.TrimSpace(req.Prompt) == "" {
+		return service.TaskErrorWrapperLocal(fmt.Errorf("field prompt is required"), "invalid_request", http.StatusBadRequest)
+	}
+	return nil
+}
+
 func (a *TaskAdaptor) ValidateRequestAndSetAction(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.TaskError) {
+	if info.Action == constant.TaskActionRemix {
+		return validateRemixRequest(c)
+	}
 	return relaycommon.ValidateMultipartDirect(c, info)
 }
 
 func (a *TaskAdaptor) BuildRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	if info.Action == constant.TaskActionRemix {
+		return fmt.Sprintf("%s/v1/videos/%s/remix", a.baseURL, info.OriginTaskID), nil
+	}
 	return fmt.Sprintf("%s/v1/videos", a.baseURL), nil
 }
 

relay/channel/task/vertex/adaptor.go

@@ -12,7 +12,6 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/model"
-
 	"github.com/gin-gonic/gin"
 
 	"github.com/QuantumNous/new-api/constant"
@@ -147,13 +146,40 @@ func (a *TaskAdaptor) BuildRequestBody(c *gin.Context, info *relaycommon.RelayIn
 			body.Parameters["storageUri"] = v
 		}
 		if v, ok := req.Metadata["sampleCount"]; ok {
-			body.Parameters["sampleCount"] = v
+			if i, ok := v.(int); ok {
+				body.Parameters["sampleCount"] = i
+			}
+			if f, ok := v.(float64); ok {
+				body.Parameters["sampleCount"] = int(f)
+			}
 		}
 	}
 	if _, ok := body.Parameters["sampleCount"]; !ok {
 		body.Parameters["sampleCount"] = 1
 	}
 
+	if body.Parameters["sampleCount"].(int) <= 0 {
+		return nil, fmt.Errorf("sampleCount must be greater than 0")
+	}
+
+	// if req.Duration > 0 {
+	// 	body.Parameters["durationSeconds"] = req.Duration
+	// } else if req.Seconds != "" {
+	// 	seconds, err := strconv.Atoi(req.Seconds)
+	// 	if err != nil {
+	// 		return nil, errors.Wrap(err, "convert seconds to int failed")
+	// 	}
+	// 	body.Parameters["durationSeconds"] = seconds
+	// }
+
+	info.PriceData.OtherRatios = map[string]float64{
+		"sampleCount": float64(body.Parameters["sampleCount"].(int)),
+	}
+
+	// if v, ok := body.Parameters["durationSeconds"]; ok {
+	// 	info.PriceData.OtherRatios["durationSeconds"] = float64(v.(int))
+	// }
+
 	data, err := json.Marshal(body)
 	if err != nil {
 		return nil, err

relay/channel/tencent/relay-tencent.go

@@ -105,7 +105,7 @@ func tencentStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *htt
 		data = strings.TrimPrefix(data, "data:")
 
 		var tencentResponse TencentChatResponse
-		err := json.Unmarshal([]byte(data), &tencentResponse)
+		err := common.Unmarshal([]byte(data), &tencentResponse)
 		if err != nil {
 			common.SysLog("error unmarshalling stream response: " + err.Error())
 			continue
@@ -130,7 +130,7 @@ func tencentStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *htt
 
 	service.CloseResponseBodyGracefully(resp)
 
-	return service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens), nil
+	return service.ResponseText2Usage(c, responseText, info.UpstreamModelName, info.GetEstimatePromptTokens()), nil
 }
 
 func tencentHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {

relay/channel/vertex/adaptor.go

@@ -39,6 +39,7 @@ var claudeModelMap = map[string]string{
 	"claude-opus-4-20250514":     "claude-opus-4@20250514",
 	"claude-opus-4-1-20250805":   "claude-opus-4-1@20250805",
 	"claude-sonnet-4-5-20250929": "claude-sonnet-4-5@20250929",
+	"claude-opus-4-5-20251101":   "claude-opus-4-5@20251101",
 }
 
 const anthropicVersion = "vertex-2023-10-16"
@@ -76,7 +77,9 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 	if strings.HasPrefix(info.UpstreamModelName, "claude") {
 		a.RequestMode = RequestModeClaude
-	} else if strings.Contains(info.UpstreamModelName, "llama") {
+	} else if strings.Contains(info.UpstreamModelName, "llama") ||
+		// open source models
+		strings.Contains(info.UpstreamModelName, "-maas") {
 		a.RequestMode = RequestModeLlama
 	} else {
 		a.RequestMode = RequestModeGemini
@@ -220,6 +223,9 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
 	if a.AccountCredentials.ProjectID != "" {
 		req.Set("x-goog-user-project", a.AccountCredentials.ProjectID)
 	}
+	if strings.Contains(info.UpstreamModelName, "claude") {
+		claude.CommonClaudeHeadersOperation(c, req, info)
+	}
 	return nil
 }
 
@@ -291,7 +297,7 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 		info.UpstreamModelName = claudeReq.Model
 		return vertexClaudeReq, nil
 	} else if a.RequestMode == RequestModeGemini {
-		geminiRequest, err := gemini.CovertGemini2OpenAI(c, *request, info)
+		geminiRequest, err := gemini.CovertOpenAI2Gemini(c, *request, info)
 		if err != nil {
 			return nil, err
 		}

relay/channel/volcengine/adaptor.go

@@ -13,6 +13,7 @@ import (
 	channelconstant "github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/relay/channel"
+	"github.com/QuantumNous/new-api/relay/channel/claude"
 	"github.com/QuantumNous/new-api/relay/channel/openai"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/relay/constant"
@@ -36,6 +37,10 @@ func (a *Adaptor) ConvertGeminiRequest(*gin.Context, *relaycommon.RelayInfo, *dt
 }
 
 func (a *Adaptor) ConvertClaudeRequest(c *gin.Context, info *relaycommon.RelayInfo, req *dto.ClaudeRequest) (any, error) {
+	if _, ok := channelconstant.ChannelSpecialBases[info.ChannelBaseUrl]; ok {
+		adaptor := claude.Adaptor{}
+		return adaptor.ConvertClaudeRequest(c, info, req)
+	}
 	adaptor := openai.Adaptor{}
 	return adaptor.ConvertClaudeRequest(c, info, req)
 }
@@ -235,9 +240,13 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 	if baseUrl == "" {
 		baseUrl = channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine]
 	}
+	specialPlan, hasSpecialPlan := channelconstant.ChannelSpecialBases[baseUrl]
 
 	switch info.RelayFormat {
 	case types.RelayFormatClaude:
+		if hasSpecialPlan && specialPlan.ClaudeBaseURL != "" {
+			return fmt.Sprintf("%s/v1/messages", specialPlan.ClaudeBaseURL), nil
+		}
 		if strings.HasPrefix(info.UpstreamModelName, "bot") {
 			return fmt.Sprintf("%s/api/v3/bots/chat/completions", baseUrl), nil
 		}
@@ -245,6 +254,9 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 	default:
 		switch info.RelayMode {
 		case constant.RelayModeChatCompletions:
+			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {
+				return fmt.Sprintf("%s/chat/completions", specialPlan.OpenAIBaseURL), nil
+			}
 			if strings.HasPrefix(info.UpstreamModelName, "bot") {
 				return fmt.Sprintf("%s/api/v3/bots/chat/completions", baseUrl), nil
 			}
@@ -331,6 +343,15 @@ func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, request
 }
 
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *types.NewAPIError) {
+	if info.RelayFormat == types.RelayFormatClaude {
+		if _, ok := channelconstant.ChannelSpecialBases[info.ChannelBaseUrl]; ok {
+			if info.IsStream {
+				return claude.ClaudeStreamHandler(c, resp, info, claude.RequestModeMessage)
+			}
+			return claude.ClaudeHandler(c, resp, info, claude.RequestModeMessage)
+		}
+	}
+
 	if info.RelayMode == constant.RelayModeAudioSpeech {
 		encoding := mapEncoding(c.GetString(contextKeyResponseFormat))
 		if info.IsStream {

relay/channel/volcengine/protocols.go

@@ -385,7 +385,7 @@ func (m *Message) writeSessionID(buf *bytes.Buffer) error {
 	}
 
 	size := len(m.SessionID)
-	if size > math.MaxUint32 {
+	if int64(size) > math.MaxUint32 {
 		return fmt.Errorf("session ID size (%d) exceeds max(uint32)", size)
 	}
 
@@ -407,7 +407,7 @@ func (m *Message) writeErrorCode(buf *bytes.Buffer) error {
 
 func (m *Message) writePayload(buf *bytes.Buffer) error {
 	size := len(m.Payload)
-	if size > math.MaxUint32 {
+	if int64(size) > math.MaxUint32 {
 		return fmt.Errorf("payload size (%d) exceeds max(uint32)", size)
 	}
 

relay/channel/volcengine/tts.go

@@ -184,9 +184,9 @@ func handleTTSResponse(c *gin.Context, resp *http.Response, info *relaycommon.Re
 	c.Data(http.StatusOK, contentType, audioData)
 
 	usage = &dto.Usage{
-		PromptTokens:     info.PromptTokens,
+		PromptTokens:     info.GetEstimatePromptTokens(),
 		CompletionTokens: 0,
-		TotalTokens:      info.PromptTokens,
+		TotalTokens:      info.GetEstimatePromptTokens(),
 	}
 
 	return usage, nil
@@ -284,9 +284,9 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 			if msg.Sequence < 0 {
 				c.Status(http.StatusOK)
 				usage = &dto.Usage{
-					PromptTokens:     info.PromptTokens,
+					PromptTokens:     info.GetEstimatePromptTokens(),
 					CompletionTokens: 0,
-					TotalTokens:      info.PromptTokens,
+					TotalTokens:      info.GetEstimatePromptTokens(),
 				}
 				return usage, nil
 			}
@@ -297,9 +297,9 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 
 	c.Status(http.StatusOK)
 	usage = &dto.Usage{
-		PromptTokens:     info.PromptTokens,
+		PromptTokens:     info.GetEstimatePromptTokens(),
 		CompletionTokens: 0,
-		TotalTokens:      info.PromptTokens,
+		TotalTokens:      info.GetEstimatePromptTokens(),
 	}
 	return usage, nil
 }

relay/channel/xai/text.go

@@ -70,7 +70,7 @@ func xAIStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 	})
 
 	if !containStreamUsage {
-		usage = service.ResponseText2Usage(responseTextBuilder.String(), info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(c, responseTextBuilder.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
 		usage.CompletionTokens += toolCount * 7
 	}
 

relay/channel/zhipu_4v/adaptor.go

@@ -6,6 +6,7 @@ import (
 	"io"
 	"net/http"
 
+	channelconstant "github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/relay/channel"
 	"github.com/QuantumNous/new-api/relay/channel/claude"
@@ -43,15 +44,30 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	baseURL := info.ChannelBaseUrl
+	if baseURL == "" {
+		baseURL = channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeZhipu_v4]
+	}
+	specialPlan, hasSpecialPlan := channelconstant.ChannelSpecialBases[baseURL]
+
 	switch info.RelayFormat {
 	case types.RelayFormatClaude:
-		return fmt.Sprintf("%s/api/anthropic/v1/messages", info.ChannelBaseUrl), nil
+		if hasSpecialPlan && specialPlan.ClaudeBaseURL != "" {
+			return fmt.Sprintf("%s/v1/messages", specialPlan.ClaudeBaseURL), nil
+		}
+		return fmt.Sprintf("%s/api/anthropic/v1/messages", baseURL), nil
 	default:
 		switch info.RelayMode {
 		case relayconstant.RelayModeEmbeddings:
-			return fmt.Sprintf("%s/api/paas/v4/embeddings", info.ChannelBaseUrl), nil
+			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {
+				return fmt.Sprintf("%s/embeddings", specialPlan.OpenAIBaseURL), nil
+			}
+			return fmt.Sprintf("%s/api/paas/v4/embeddings", baseURL), nil
 		default:
-			return fmt.Sprintf("%s/api/paas/v4/chat/completions", info.ChannelBaseUrl), nil
+			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {
+				return fmt.Sprintf("%s/chat/completions", specialPlan.OpenAIBaseURL), nil
+			}
+			return fmt.Sprintf("%s/api/paas/v4/chat/completions", baseURL), nil
 		}
 	}
 }

relay/channel/zhipu_4v/constants.go

@@ -1,7 +1,7 @@
 package zhipu_4v
 
 var ModelList = []string{
-	"glm-4", "glm-4v", "glm-3-turbo", "glm-4-alltools", "glm-4-plus", "glm-4-0520", "glm-4-air", "glm-4-airx", "glm-4-long", "glm-4-flash", "glm-4v-plus",
+	"glm-4", "glm-4v", "glm-3-turbo", "glm-4-alltools", "glm-4-plus", "glm-4-0520", "glm-4-air", "glm-4-airx", "glm-4-long", "glm-4-flash", "glm-4v-plus", "glm-4.6",
 }
 
 var ChannelName = "zhipu_4v"

relay/claude_handler.go

@@ -123,7 +123,7 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 
 		// apply param override
 		if len(info.ParamOverride) > 0 {
-			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 			if err != nil {
 				return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 			}

relay/common/override.go

@@ -1,12 +1,12 @@
 package common
 
 import (
-	"encoding/json"
 	"fmt"
 	"regexp"
 	"strconv"
 	"strings"
 
+	"github.com/QuantumNous/new-api/common"
 	"github.com/tidwall/gjson"
 	"github.com/tidwall/sjson"
 )
@@ -30,7 +30,7 @@ type ParamOperation struct {
 	Logic      string               `json:"logic,omitempty"`      // AND, OR (默认OR)
 }
 
-func ApplyParamOverride(jsonData []byte, paramOverride map[string]interface{}) ([]byte, error) {
+func ApplyParamOverride(jsonData []byte, paramOverride map[string]interface{}, conditionContext map[string]interface{}) ([]byte, error) {
 	if len(paramOverride) == 0 {
 		return jsonData, nil
 	}
@@ -38,7 +38,7 @@ func ApplyParamOverride(jsonData []byte, paramOverride map[string]interface{}) (
 	// 尝试断言为操作格式
 	if operations, ok := tryParseOperations(paramOverride); ok {
 		// 使用新方法
-		result, err := applyOperations(string(jsonData), operations)
+		result, err := applyOperations(string(jsonData), operations, conditionContext)
 		return []byte(result), err
 	}
 
@@ -123,13 +123,13 @@ func tryParseOperations(paramOverride map[string]interface{}) ([]ParamOperation,
 	return nil, false
 }
 
-func checkConditions(jsonStr string, conditions []ConditionOperation, logic string) (bool, error) {
+func checkConditions(jsonStr, contextJSON string, conditions []ConditionOperation, logic string) (bool, error) {
 	if len(conditions) == 0 {
 		return true, nil // 没有条件，直接通过
 	}
 	results := make([]bool, len(conditions))
 	for i, condition := range conditions {
-		result, err := checkSingleCondition(jsonStr, condition)
+		result, err := checkSingleCondition(jsonStr, contextJSON, condition)
 		if err != nil {
 			return false, err
 		}
@@ -153,10 +153,13 @@ func checkConditions(jsonStr string, conditions []ConditionOperation, logic stri
 	}
 }
 
-func checkSingleCondition(jsonStr string, condition ConditionOperation) (bool, error) {
+func checkSingleCondition(jsonStr, contextJSON string, condition ConditionOperation) (bool, error) {
 	// 处理负数索引
 	path := processNegativeIndex(jsonStr, condition.Path)
 	value := gjson.Get(jsonStr, path)
+	if !value.Exists() && contextJSON != "" {
+		value = gjson.Get(contextJSON, condition.Path)
+	}
 	if !value.Exists() {
 		if condition.PassMissingKey {
 			return true, nil
@@ -165,7 +168,7 @@ func checkSingleCondition(jsonStr string, condition ConditionOperation) (bool, e
 	}
 
 	// 利用gjson的类型解析
-	targetBytes, err := json.Marshal(condition.Value)
+	targetBytes, err := common.Marshal(condition.Value)
 	if err != nil {
 		return false, fmt.Errorf("failed to marshal condition value: %v", err)
 	}
@@ -292,7 +295,7 @@ func compareNumeric(jsonValue, targetValue gjson.Result, operator string) (bool,
 // applyOperationsLegacy 原参数覆盖方法
 func applyOperationsLegacy(jsonData []byte, paramOverride map[string]interface{}) ([]byte, error) {
 	reqMap := make(map[string]interface{})
-	err := json.Unmarshal(jsonData, &reqMap)
+	err := common.Unmarshal(jsonData, &reqMap)
 	if err != nil {
 		return nil, err
 	}
@@ -301,14 +304,23 @@ func applyOperationsLegacy(jsonData []byte, paramOverride map[string]interface{}
 		reqMap[key] = value
 	}
 
-	return json.Marshal(reqMap)
+	return common.Marshal(reqMap)
 }
 
-func applyOperations(jsonStr string, operations []ParamOperation) (string, error) {
+func applyOperations(jsonStr string, operations []ParamOperation, conditionContext map[string]interface{}) (string, error) {
+	var contextJSON string
+	if conditionContext != nil && len(conditionContext) > 0 {
+		ctxBytes, err := common.Marshal(conditionContext)
+		if err != nil {
+			return "", fmt.Errorf("failed to marshal condition context: %v", err)
+		}
+		contextJSON = string(ctxBytes)
+	}
+
 	result := jsonStr
 	for _, op := range operations {
 		// 检查条件是否满足
-		ok, err := checkConditions(result, op.Conditions, op.Logic)
+		ok, err := checkConditions(result, contextJSON, op.Conditions, op.Logic)
 		if err != nil {
 			return "", err
 		}
@@ -414,7 +426,7 @@ func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (str
 	var currentMap, newMap map[string]interface{}
 
 	// 解析当前值
-	if err := json.Unmarshal([]byte(current.Raw), &currentMap); err != nil {
+	if err := common.Unmarshal([]byte(current.Raw), &currentMap); err != nil {
 		return "", err
 	}
 	// 解析新值
@@ -422,8 +434,8 @@ func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (str
 	case map[string]interface{}:
 		newMap = v
 	default:
-		jsonBytes, _ := json.Marshal(v)
-		if err := json.Unmarshal(jsonBytes, &newMap); err != nil {
+		jsonBytes, _ := common.Marshal(v)
+		if err := common.Unmarshal(jsonBytes, &newMap); err != nil {
 			return "", err
 		}
 	}
@@ -439,3 +451,31 @@ func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (str
 	}
 	return sjson.Set(jsonStr, path, result)
 }
+
+// BuildParamOverrideContext 提供 ApplyParamOverride 可用的上下文信息。
+// 目前内置以下字段：
+//   - model：优先使用上游模型名（UpstreamModelName），若不存在则回落到原始模型名（OriginModelName）。
+//   - upstream_model：始终为通道映射后的上游模型名。
+//   - original_model：请求最初指定的模型名。
+func BuildParamOverrideContext(info *RelayInfo) map[string]interface{} {
+	if info == nil || info.ChannelMeta == nil {
+		return nil
+	}
+
+	ctx := make(map[string]interface{})
+	if info.UpstreamModelName != "" {
+		ctx["model"] = info.UpstreamModelName
+		ctx["upstream_model"] = info.UpstreamModelName
+	}
+	if info.OriginModelName != "" {
+		ctx["original_model"] = info.OriginModelName
+		if _, exists := ctx["model"]; !exists {
+			ctx["model"] = info.OriginModelName
+		}
+	}
+
+	if len(ctx) == 0 {
+		return nil
+	}
+	return ctx
+}

relay/common/relay_info.go

@@ -73,6 +73,11 @@ type ChannelMeta struct {
 	SupportStreamOptions bool // 是否支持流式选项
 }
 
+type TokenCountMeta struct {
+	//promptTokens int
+	estimatePromptTokens int
+}
+
 type RelayInfo struct {
 	TokenId           int
 	TokenKey          string
@@ -91,7 +96,6 @@ type RelayInfo struct {
 	RelayMode              int
 	OriginModelName        string
 	RequestURLPath         string
-	PromptTokens           int
 	ShouldIncludeUsage     bool
 	DisablePing            bool // 是否禁止向下游发送自定义 Ping
 	ClientWs               *websocket.Conn
@@ -115,6 +119,7 @@ type RelayInfo struct {
 	Request dto.Request
 
 	ThinkingContentInfo
+	TokenCountMeta
 	*ClaudeConvertInfo
 	*RerankerInfo
 	*ResponsesUsageInfo
@@ -189,7 +194,7 @@ func (info *RelayInfo) ToString() string {
 	fmt.Fprintf(b, "IsPlayground: %t, ", info.IsPlayground)
 	fmt.Fprintf(b, "RequestURLPath: %q, ", info.RequestURLPath)
 	fmt.Fprintf(b, "OriginModelName: %q, ", info.OriginModelName)
-	fmt.Fprintf(b, "PromptTokens: %d, ", info.PromptTokens)
+	fmt.Fprintf(b, "EstimatePromptTokens: %d, ", info.estimatePromptTokens)
 	fmt.Fprintf(b, "ShouldIncludeUsage: %t, ", info.ShouldIncludeUsage)
 	fmt.Fprintf(b, "DisablePing: %t, ", info.DisablePing)
 	fmt.Fprintf(b, "SendResponseCount: %d, ", info.SendResponseCount)
@@ -391,7 +396,6 @@ func genBaseRelayInfo(c *gin.Context, request dto.Request) *RelayInfo {
 		UserEmail:  common.GetContextKeyString(c, constant.ContextKeyUserEmail),
 
 		OriginModelName: common.GetContextKeyString(c, constant.ContextKeyOriginalModel),
-		PromptTokens:    common.GetContextKeyInt(c, constant.ContextKeyPromptTokens),
 
 		TokenId:        common.GetContextKeyInt(c, constant.ContextKeyTokenId),
 		TokenKey:       common.GetContextKeyString(c, constant.ContextKeyTokenKey),
@@ -408,6 +412,10 @@ func genBaseRelayInfo(c *gin.Context, request dto.Request) *RelayInfo {
 			IsFirstThinkingContent:  true,
 			SendLastThinkingContent: false,
 		},
+		TokenCountMeta: TokenCountMeta{
+			//promptTokens: common.GetContextKeyInt(c, constant.ContextKeyPromptTokens),
+			estimatePromptTokens: common.GetContextKeyInt(c, constant.ContextKeyEstimatedTokens),
+		},
 	}
 
 	if info.RelayMode == relayconstant.RelayModeUnknown {
@@ -463,8 +471,16 @@ func GenRelayInfo(c *gin.Context, relayFormat types.RelayFormat, request dto.Req
 	}
 }
 
-func (info *RelayInfo) SetPromptTokens(promptTokens int) {
-	info.PromptTokens = promptTokens
+//func (info *RelayInfo) SetPromptTokens(promptTokens int) {
+//	info.promptTokens = promptTokens
+//}
+
+func (info *RelayInfo) SetEstimatePromptTokens(promptTokens int) {
+	info.estimatePromptTokens = promptTokens
+}
+
+func (info *RelayInfo) GetEstimatePromptTokens() int {
+	return info.estimatePromptTokens
 }
 
 func (info *RelayInfo) SetFirstResponseTime() {
@@ -498,11 +514,11 @@ type TaskSubmitReq struct {
 	Metadata       map[string]interface{} `json:"metadata,omitempty"`
 }
 
-func (t TaskSubmitReq) GetPrompt() string {
+func (t *TaskSubmitReq) GetPrompt() string {
 	return t.Prompt
 }
 
-func (t TaskSubmitReq) HasImage() bool {
+func (t *TaskSubmitReq) HasImage() bool {
 	return len(t.Images) > 0
 }
 
@@ -537,6 +553,20 @@ func (t *TaskSubmitReq) UnmarshalJSON(data []byte) error {
 
 	return nil
 }
+func (t *TaskSubmitReq) UnmarshalMetadata(v any) error {
+	metadata := t.Metadata
+	if metadata != nil {
+		metadataBytes, err := json.Marshal(metadata)
+		if err != nil {
+			return fmt.Errorf("marshal metadata failed: %w", err)
+		}
+		err = json.Unmarshal(metadataBytes, v)
+		if err != nil {
+			return fmt.Errorf("unmarshal metadata to target failed: %w", err)
+		}
+	}
+	return nil
+}
 
 type TaskInfo struct {
 	Code             int    `json:"code"`

relay/common_handler/rerank.go

@@ -57,8 +57,8 @@ func RerankHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 		jinaResp = dto.RerankResponse{
 			Results: jinaRespResults,
 			Usage: dto.Usage{
-				PromptTokens: info.PromptTokens,
-				TotalTokens:  info.PromptTokens,
+				PromptTokens: info.GetEstimatePromptTokens(),
+				TotalTokens:  info.GetEstimatePromptTokens(),
 			},
 		}
 	} else {

relay/compatible_handler.go

@@ -144,7 +144,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 
 		// apply param override
 		if len(info.ParamOverride) > 0 {
-			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 			if err != nil {
 				return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 			}
@@ -192,9 +192,9 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.Usage, extraContent string) {
 	if usage == nil {
 		usage = &dto.Usage{
-			PromptTokens:     relayInfo.PromptTokens,
+			PromptTokens:     relayInfo.GetEstimatePromptTokens(),
 			CompletionTokens: 0,
-			TotalTokens:      relayInfo.PromptTokens,
+			TotalTokens:      relayInfo.GetEstimatePromptTokens(),
 		}
 		extraContent += "（可能是请求出错）"
 	}

relay/embedding_handler.go

@@ -49,6 +49,14 @@ func EmbeddingHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 	if err != nil {
 		return types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
 	}
+
+	if len(info.ParamOverride) > 0 {
+		jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
+		if err != nil {
+			return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
+		}
+	}
+
 	logger.LogDebug(c, fmt.Sprintf("converted embedding request body: %s", string(jsonData)))
 	requestBody := bytes.NewBuffer(jsonData)
 	statusCodeMappingStr := c.GetString("status_code_mapping")

relay/gemini_handler.go

@@ -156,7 +156,7 @@ func GeminiHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 
 		// apply param override
 		if len(info.ParamOverride) > 0 {
-			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 			if err != nil {
 				return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 			}

relay/helper/price.go

@@ -99,7 +99,10 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 	// check if free model pre-consume is disabled
 	if !operation_setting.GetQuotaSetting().EnableFreeModelPreConsume {
 		// if model price or ratio is 0, do not pre-consume quota
-		if usePrice {
+		if groupRatioInfo.GroupRatio == 0 {
+			preConsumedQuota = 0
+			freeModel = true
+		} else if usePrice {
 			if modelPrice == 0 {
 				preConsumedQuota = 0
 				freeModel = true

relay/helper/stream_scanner.go

@@ -22,11 +22,18 @@ import (
 )
 
 const (
-	InitialScannerBufferSize = 64 << 10 // 64KB (64*1024)
-	MaxScannerBufferSize     = 10 << 20 // 10MB (10*1024*1024)
-	DefaultPingInterval      = 10 * time.Second
+	InitialScannerBufferSize    = 64 << 10 // 64KB (64*1024)
+	DefaultMaxScannerBufferSize = 64 << 20 // 64MB (64*1024*1024) default SSE buffer size
+	DefaultPingInterval         = 10 * time.Second
 )
 
+func getScannerBufferSize() int {
+	if constant.StreamScannerMaxBufferMB > 0 {
+		return constant.StreamScannerMaxBufferMB << 20
+	}
+	return DefaultMaxScannerBufferSize
+}
+
 func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, dataHandler func(data string) bool) {
 
 	if resp == nil || dataHandler == nil {
@@ -65,6 +72,8 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 	if common.DebugEnabled {
 		// print timeout and ping interval for debugging
 		println("relay timeout seconds:", common.RelayTimeout)
+		println("relay max idle conns:", common.RelayMaxIdleConns)
+		println("relay max idle conns per host:", common.RelayMaxIdleConnsPerHost)
 		println("streaming timeout seconds:", int64(streamingTimeout.Seconds()))
 		println("ping interval seconds:", int64(pingInterval.Seconds()))
 	}
@@ -95,7 +104,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 		close(stopChan)
 	}()
 
-	scanner.Buffer(make([]byte, InitialScannerBufferSize), MaxScannerBufferSize)
+	scanner.Buffer(make([]byte, InitialScannerBufferSize), getScannerBufferSize())
 	scanner.Split(bufio.ScanLines)
 	SetEventStreamHeaders(c)
 

relay/image_handler.go

@@ -69,7 +69,7 @@ func ImageHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 
 			// apply param override
 			if len(info.ParamOverride) > 0 {
-				jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+				jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 				if err != nil {
 					return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 				}

relay/relay_adaptor.go

@@ -32,6 +32,7 @@ import (
 	taskali "github.com/QuantumNous/new-api/relay/channel/task/ali"
 	taskdoubao "github.com/QuantumNous/new-api/relay/channel/task/doubao"
 	taskGemini "github.com/QuantumNous/new-api/relay/channel/task/gemini"
+	"github.com/QuantumNous/new-api/relay/channel/task/hailuo"
 	taskjimeng "github.com/QuantumNous/new-api/relay/channel/task/jimeng"
 	"github.com/QuantumNous/new-api/relay/channel/task/kling"
 	tasksora "github.com/QuantumNous/new-api/relay/channel/task/sora"
@@ -153,6 +154,8 @@ func GetTaskAdaptor(platform constant.TaskPlatform) channel.TaskAdaptor {
 			return &tasksora.TaskAdaptor{}
 		case constant.ChannelTypeGemini:
 			return &taskGemini.TaskAdaptor{}
+		case constant.ChannelTypeMiniMax:
+			return &hailuo.TaskAdaptor{}
 		}
 	}
 	return nil

relay/relay_task.go

@@ -32,7 +32,86 @@ func RelayTaskSubmit(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.
 	if info.TaskRelayInfo == nil {
 		info.TaskRelayInfo = &relaycommon.TaskRelayInfo{}
 	}
+	path := c.Request.URL.Path
+	if strings.Contains(path, "/v1/videos/") && strings.HasSuffix(path, "/remix") {
+		info.Action = constant.TaskActionRemix
+	}
+
+	// 提取 remix 任务的 video_id
+	if info.Action == constant.TaskActionRemix {
+		videoID := c.Param("video_id")
+		if strings.TrimSpace(videoID) == "" {
+			return service.TaskErrorWrapperLocal(fmt.Errorf("video_id is required"), "invalid_request", http.StatusBadRequest)
+		}
+		info.OriginTaskID = videoID
+	}
+
 	platform := constant.TaskPlatform(c.GetString("platform"))
+
+	// 获取原始任务信息
+	if info.OriginTaskID != "" {
+		originTask, exist, err := model.GetByTaskId(info.UserId, info.OriginTaskID)
+		if err != nil {
+			taskErr = service.TaskErrorWrapper(err, "get_origin_task_failed", http.StatusInternalServerError)
+			return
+		}
+		if !exist {
+			taskErr = service.TaskErrorWrapperLocal(errors.New("task_origin_not_exist"), "task_not_exist", http.StatusBadRequest)
+			return
+		}
+		if info.OriginModelName == "" {
+			if originTask.Properties.OriginModelName != "" {
+				info.OriginModelName = originTask.Properties.OriginModelName
+			} else if originTask.Properties.UpstreamModelName != "" {
+				info.OriginModelName = originTask.Properties.UpstreamModelName
+			} else {
+				var taskData map[string]interface{}
+				_ = json.Unmarshal(originTask.Data, &taskData)
+				if m, ok := taskData["model"].(string); ok && m != "" {
+					info.OriginModelName = m
+					platform = originTask.Platform
+				}
+			}
+		}
+		if originTask.ChannelId != info.ChannelId {
+			channel, err := model.GetChannelById(originTask.ChannelId, true)
+			if err != nil {
+				taskErr = service.TaskErrorWrapperLocal(err, "channel_not_found", http.StatusBadRequest)
+				return
+			}
+			if channel.Status != common.ChannelStatusEnabled {
+				taskErr = service.TaskErrorWrapperLocal(errors.New("the channel of the origin task is disabled"), "task_channel_disable", http.StatusBadRequest)
+				return
+			}
+			c.Set("base_url", channel.GetBaseURL())
+			c.Set("channel_id", originTask.ChannelId)
+			c.Request.Header.Set("Authorization", fmt.Sprintf("Bearer %s", channel.Key))
+
+			info.ChannelBaseUrl = channel.GetBaseURL()
+			info.ChannelId = originTask.ChannelId
+			platform = originTask.Platform
+		}
+
+		// 使用原始任务的参数
+		if info.Action == constant.TaskActionRemix {
+			var taskData map[string]interface{}
+			_ = json.Unmarshal(originTask.Data, &taskData)
+			secondsStr, _ := taskData["seconds"].(string)
+			seconds, _ := strconv.Atoi(secondsStr)
+			if seconds <= 0 {
+				seconds = 4
+			}
+			sizeStr, _ := taskData["size"].(string)
+			if info.PriceData.OtherRatios == nil {
+				info.PriceData.OtherRatios = map[string]float64{}
+			}
+			info.PriceData.OtherRatios["seconds"] = float64(seconds)
+			info.PriceData.OtherRatios["size"] = 1
+			if sizeStr == "1792x1024" || sizeStr == "1024x1792" {
+				info.PriceData.OtherRatios["size"] = 1.666667
+			}
+		}
+	}
 	if platform == "" {
 		platform = GetTaskPlatform(c)
 	}
@@ -94,34 +173,6 @@ func RelayTaskSubmit(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.
 		return
 	}
 
-	if info.OriginTaskID != "" {
-		originTask, exist, err := model.GetByTaskId(info.UserId, info.OriginTaskID)
-		if err != nil {
-			taskErr = service.TaskErrorWrapper(err, "get_origin_task_failed", http.StatusInternalServerError)
-			return
-		}
-		if !exist {
-			taskErr = service.TaskErrorWrapperLocal(errors.New("task_origin_not_exist"), "task_not_exist", http.StatusBadRequest)
-			return
-		}
-		if originTask.ChannelId != info.ChannelId {
-			channel, err := model.GetChannelById(originTask.ChannelId, true)
-			if err != nil {
-				taskErr = service.TaskErrorWrapperLocal(err, "channel_not_found", http.StatusBadRequest)
-				return
-			}
-			if channel.Status != common.ChannelStatusEnabled {
-				return service.TaskErrorWrapperLocal(errors.New("该任务所属渠道已被禁用"), "task_channel_disable", http.StatusBadRequest)
-			}
-			c.Set("base_url", channel.GetBaseURL())
-			c.Set("channel_id", originTask.ChannelId)
-			c.Request.Header.Set("Authorization", fmt.Sprintf("Bearer %s", channel.Key))
-
-			info.ChannelBaseUrl = channel.GetBaseURL()
-			info.ChannelId = originTask.ChannelId
-		}
-	}
-
 	// build body
 	requestBody, err := adaptor.BuildRequestBody(c, info)
 	if err != nil {

relay/rerank_handler.go

@@ -60,7 +60,7 @@ func RerankHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 
 		// apply param override
 		if len(info.ParamOverride) > 0 {
-			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 			if err != nil {
 				return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 			}

relay/responses_handler.go

@@ -66,7 +66,7 @@ func ResponsesHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 
 		// apply param override
 		if len(info.ParamOverride) > 0 {
-			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride)
+			jsonData, err = relaycommon.ApplyParamOverride(jsonData, info.ParamOverride, relaycommon.BuildParamOverrideContext(info))
 			if err != nil {
 				return types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
 			}

router/api-router.go

@@ -30,6 +30,7 @@ func SetApiRouter(router *gin.Engine) {
 		apiRouter.GET("/reset_password", middleware.CriticalRateLimit(), middleware.TurnstileCheck(), controller.SendPasswordResetEmail)
 		apiRouter.POST("/user/reset", middleware.CriticalRateLimit(), controller.ResetPassword)
 		apiRouter.GET("/oauth/github", middleware.CriticalRateLimit(), controller.GitHubOAuth)
+		apiRouter.GET("/oauth/discord", middleware.CriticalRateLimit(), controller.DiscordOAuth)
 		apiRouter.GET("/oauth/oidc", middleware.CriticalRateLimit(), controller.OidcAuth)
 		apiRouter.GET("/oauth/linuxdo", middleware.CriticalRateLimit(), controller.LinuxdoOAuth)
 		apiRouter.GET("/oauth/state", middleware.CriticalRateLimit(), controller.GenerateOAuthCode)

router/video-router.go

@@ -9,11 +9,12 @@ import (
 
 func SetVideoRouter(router *gin.Engine) {
 	videoV1Router := router.Group("/v1")
-	videoV1Router.GET("/videos/:task_id/content", controller.VideoProxy)
 	videoV1Router.Use(middleware.TokenAuth(), middleware.Distribute())
 	{
+		videoV1Router.GET("/videos/:task_id/content", controller.VideoProxy)
 		videoV1Router.POST("/video/generations", controller.RelayTask)
 		videoV1Router.GET("/video/generations/:task_id", controller.RelayTask)
+		videoV1Router.POST("/videos/:video_id/remix", controller.RelayTask)
 	}
 	// openai compatible API video routes
 	// docs: https://platform.openai.com/docs/api-reference/videos/create

service/convert.go

@@ -209,7 +209,7 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 			Type:  "message",
 			Role:  "assistant",
 			Usage: &dto.ClaudeUsage{
-				InputTokens:  info.PromptTokens,
+				InputTokens:  info.GetEstimatePromptTokens(),
 				OutputTokens: 0,
 			},
 		}
@@ -734,12 +734,18 @@ func StreamResponseOpenAI2Gemini(openAIResponse *dto.ChatCompletionsStreamRespon
 	geminiResponse := &dto.GeminiChatResponse{
 		Candidates: make([]dto.GeminiChatCandidate, 0, len(openAIResponse.Choices)),
 		UsageMetadata: dto.GeminiUsageMetadata{
-			PromptTokenCount:     info.PromptTokens,
+			PromptTokenCount:     info.GetEstimatePromptTokens(),
 			CandidatesTokenCount: 0, // 流式响应中可能没有完整的 usage 信息
-			TotalTokenCount:      info.PromptTokens,
+			TotalTokenCount:      info.GetEstimatePromptTokens(),
 		},
 	}
 
+	if openAIResponse.Usage != nil {
+		geminiResponse.UsageMetadata.PromptTokenCount = openAIResponse.Usage.PromptTokens
+		geminiResponse.UsageMetadata.CandidatesTokenCount = openAIResponse.Usage.CompletionTokens
+		geminiResponse.UsageMetadata.TotalTokenCount = openAIResponse.Usage.TotalTokens
+	}
+
 	for _, choice := range openAIResponse.Choices {
 		candidate := dto.GeminiChatCandidate{
 			Index:         int64(choice.Index),