feat: add remixGenerate const

.gitignore

@@ -16,7 +16,6 @@ new-api
 tiktoken_cache
 .eslintcache
 .gocache
-.gomodcache/
 .cache
 web/bun.lock
 

Dockerfile

@@ -14,7 +14,7 @@ ENV GO111MODULE=on CGO_ENABLED=0
 ARG TARGETOS
 ARG TARGETARCH
 ENV GOOS=${TARGETOS:-linux} GOARCH=${TARGETARCH:-amd64}
-ENV GOEXPERIMENT=greenteagc
+
 
 WORKDIR /build
 
@@ -25,11 +25,10 @@ COPY . .
 COPY --from=builder /build/dist ./web/dist
 RUN go build -ldflags "-s -w -X 'github.com/QuantumNous/new-api/common.Version=$(cat VERSION)'" -o new-api
 
-FROM debian:bookworm-slim
+FROM alpine
 
-RUN apt-get update \
-    && apt-get install -y --no-install-recommends ca-certificates tzdata libasan8 wget \
-    && rm -rf /var/lib/apt/lists/* \
+RUN apk upgrade --no-cache \
+    && apk add --no-cache ca-certificates tzdata \
     && update-ca-certificates
 
 COPY --from=builder2 /build/new-api /

README.en.md

@@ -305,7 +305,6 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis connection string | - |
 | `STREAMING_TIMEOUT` | Streaming timeout (seconds) | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | Max per-line buffer (MB) for the stream scanner; increase when upstream sends huge image/base64 payloads | `64` |
-| `MAX_REQUEST_BODY_MB` | Max request body size (MB, counted **after decompression**; prevents huge requests/zip bombs from exhausting memory). Exceeding it returns `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure API version | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Error log switch | `false` |
 

README.fr.md

@@ -301,7 +301,6 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Chaine de connexion Redis | - |
 | `STREAMING_TIMEOUT` | Délai d'expiration du streaming (secondes) | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | Taille max du buffer par ligne (Mo) pour le scanner SSE ; à augmenter quand les sorties image/base64 sont très volumineuses (ex. images 4K) | `64` |
-| `MAX_REQUEST_BODY_MB` | Taille maximale du corps de requête (Mo, comptée **après décompression** ; évite les requêtes énormes/zip bombs qui saturent la mémoire). Dépassement ⇒ `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Version de l'API Azure | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | Interrupteur du journal d'erreurs | `false` |
 

README.ja.md

@@ -310,7 +310,6 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis接続文字列 | - |
 | `STREAMING_TIMEOUT` | ストリーミング応答のタイムアウト時間（秒） | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | ストリームスキャナの1行あたりバッファ上限（MB）。4K画像など巨大なbase64 `data:` ペイロードを扱う場合は値を増加させてください | `64` |
-| `MAX_REQUEST_BODY_MB` | リクエストボディ最大サイズ（MB、**解凍後**に計測。巨大リクエスト/zip bomb によるメモリ枯渇を防止）。超過時は `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure APIバージョン | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | エラーログスイッチ | `false` |
 

README.md

@@ -306,7 +306,6 @@ docker run --name new-api -d --restart always \
 | `REDIS_CONN_STRING` | Redis 连接字符串                                                  | - |
 | `STREAMING_TIMEOUT` | 流式超时时间（秒）                                                    | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | 流式扫描器单行最大缓冲（MB），图像生成等超大 `data:` 片段（如 4K 图片 base64）需适当调大 | `64` |
-| `MAX_REQUEST_BODY_MB` | 请求体最大大小（MB，**解压后**计；防止超大请求/zip bomb 导致内存暴涨），超过将返回 `413` | `32` |
 | `AZURE_DEFAULT_API_VERSION` | Azure API 版本                                                 | `2025-04-01-preview` |
 | `ERROR_LOG_ENABLED` | 错误日志开关                                                       | `false` |
 

common/audio.go

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

common/email.go

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

common/gin.go

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

common/init.go

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

common/ip.go

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

common/json.go

@@ -23,11 +23,11 @@ func Marshal(v any) ([]byte, error) {
 }
 
 func GetJsonType(data json.RawMessage) string {
-	trimmed := bytes.TrimSpace(data)
-	if len(trimmed) == 0 {
+	data = bytes.TrimSpace(data)
+	if len(data) == 0 {
 		return "unknown"
 	}
-	firstChar := trimmed[0]
+	firstChar := bytes.TrimSpace(data)[0]
 	switch firstChar {
 	case '{':
 		return "object"

common/ssrf_protection.go

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

common/str.go

@@ -3,19 +3,12 @@ package common
 import (
 	"encoding/base64"
 	"encoding/json"
+	"math/rand"
 	"net/url"
 	"regexp"
 	"strconv"
 	"strings"
 	"unsafe"
-
-	"github.com/samber/lo"
-)
-
-var (
-	maskURLPattern    = regexp.MustCompile(`(http|https)://[^\s/$.?#].[^\s]*`)
-	maskDomainPattern = regexp.MustCompile(`\b(?:[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?\.)+[a-zA-Z]{2,}\b`)
-	maskIPPattern     = regexp.MustCompile(`\b(?:\d{1,3}\.){3}\d{1,3}\b`)
 )
 
 func GetStringIfEmpty(str string, defaultValue string) string {
@@ -26,10 +19,12 @@ func GetStringIfEmpty(str string, defaultValue string) string {
 }
 
 func GetRandomString(length int) string {
-	if length <= 0 {
-		return ""
+	//rand.Seed(time.Now().UnixNano())
+	key := make([]byte, length)
+	for i := 0; i < length; i++ {
+		key[i] = keyChars[rand.Intn(len(keyChars))]
 	}
-	return lo.RandomString(length, lo.AlphanumericCharset)
+	return string(key)
 }
 
 func MapToJsonStr(m map[string]interface{}) string {
@@ -175,7 +170,8 @@ func maskHostForPlainDomain(domain string) string {
 // api.openai.com -> ***.***.com
 func MaskSensitiveInfo(str string) string {
 	// Mask URLs
-	str = maskURLPattern.ReplaceAllStringFunc(str, func(urlStr string) string {
+	urlPattern := regexp.MustCompile(`(http|https)://[^\s/$.?#].[^\s]*`)
+	str = urlPattern.ReplaceAllStringFunc(str, func(urlStr string) string {
 		u, err := url.Parse(urlStr)
 		if err != nil {
 			return urlStr
@@ -228,12 +224,14 @@ func MaskSensitiveInfo(str string) string {
 	})
 
 	// Mask domain names without protocol (like openai.com, www.openai.com)
-	str = maskDomainPattern.ReplaceAllStringFunc(str, func(domain string) string {
+	domainPattern := regexp.MustCompile(`\b(?:[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?\.)+[a-zA-Z]{2,}\b`)
+	str = domainPattern.ReplaceAllStringFunc(str, func(domain string) string {
 		return maskHostForPlainDomain(domain)
 	})
 
 	// Mask IP addresses
-	str = maskIPPattern.ReplaceAllString(str, "***.***.***.***")
+	ipPattern := regexp.MustCompile(`\b(?:\d{1,3}\.){3}\d{1,3}\b`)
+	str = ipPattern.ReplaceAllString(str, "***.***.***.***")
 
 	return str
 }

common/utils.go

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

constant/context_key.go

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

constant/env.go

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

controller/billing.go

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

controller/channel.go

@@ -165,30 +165,6 @@ func GetAllChannels(c *gin.Context) {
 	return
 }
 
-func buildFetchModelsHeaders(channel *model.Channel, key string) (http.Header, error) {
-	var headers http.Header
-	switch channel.Type {
-	case constant.ChannelTypeAnthropic:
-		headers = GetClaudeAuthHeader(key)
-	default:
-		headers = GetAuthHeader(key)
-	}
-
-	headerOverride := channel.GetHeaderOverride()
-	for k, v := range headerOverride {
-		str, ok := v.(string)
-		if !ok {
-			return nil, fmt.Errorf("invalid header override for key %s", k)
-		}
-		if strings.Contains(str, "{api_key}") {
-			str = strings.ReplaceAll(str, "{api_key}", key)
-		}
-		headers.Set(k, str)
-	}
-
-	return headers, nil
-}
-
 func FetchUpstreamModels(c *gin.Context) {
 	id, err := strconv.Atoi(c.Param("id"))
 	if err != nil {
@@ -247,13 +223,14 @@ func FetchUpstreamModels(c *gin.Context) {
 	}
 	key = strings.TrimSpace(key)
 
-	headers, err := buildFetchModelsHeaders(channel, key)
-	if err != nil {
-		common.ApiError(c, err)
-		return
+	// 获取响应体 - 根据渠道类型决定是否添加 AuthHeader
+	var body []byte
+	switch channel.Type {
+	case constant.ChannelTypeAnthropic:
+		body, err = GetResponseBody("GET", url, channel, GetClaudeAuthHeader(key))
+	default:
+		body, err = GetResponseBody("GET", url, channel, GetAuthHeader(key))
 	}
-
-	body, err := GetResponseBody("GET", url, channel, headers)
 	if err != nil {
 		common.ApiError(c, err)
 		return

controller/discord.go

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

controller/model.go

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

controller/playground.go

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

controller/relay.go

@@ -2,7 +2,6 @@ package controller
 
 import (
 	"bytes"
-	"errors"
 	"fmt"
 	"io"
 	"log"
@@ -65,8 +64,8 @@ func geminiRelayHandler(c *gin.Context, info *relaycommon.RelayInfo) *types.NewA
 func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 	requestId := c.GetString(common.RequestIdKey)
-	//group := common.GetContextKeyString(c, constant.ContextKeyUsingGroup)
-	//originalModel := common.GetContextKeyString(c, constant.ContextKeyOriginalModel)
+	group := common.GetContextKeyString(c, constant.ContextKeyUsingGroup)
+	originalModel := common.GetContextKeyString(c, constant.ContextKeyOriginalModel)
 
 	var (
 		newAPIError *types.NewAPIError
@@ -105,12 +104,7 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 	request, err := helper.GetAndValidateRequest(c, relayFormat)
 	if err != nil {
-		// Map "request body too large" to 413 so clients can handle it correctly
-		if common.IsRequestBodyTooLargeError(err) || errors.Is(err, common.ErrRequestBodyTooLarge) {
-			newAPIError = types.NewErrorWithStatusCode(err, types.ErrorCodeReadRequestBodyFailed, http.StatusRequestEntityTooLarge, types.ErrOptionWithSkipRetry())
-		} else {
-			newAPIError = types.NewError(err, types.ErrorCodeInvalidRequest)
-		}
+		newAPIError = types.NewError(err, types.ErrorCodeInvalidRequest)
 		return
 	}
 
@@ -120,17 +114,9 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 		return
 	}
 
-	needSensitiveCheck := setting.ShouldCheckPromptSensitive()
-	needCountToken := constant.CountToken
-	// Avoid building huge CombineText (strings.Join) when token counting and sensitive check are both disabled.
-	var meta *types.TokenCountMeta
-	if needSensitiveCheck || needCountToken {
-		meta = request.GetTokenCountMeta()
-	} else {
-		meta = fastTokenCountMetaForPricing(request)
-	}
+	meta := request.GetTokenCountMeta()
 
-	if needSensitiveCheck && meta != nil {
+	if setting.ShouldCheckPromptSensitive() {
 		contains, words := service.CheckSensitiveText(meta.CombineText)
 		if contains {
 			logger.LogWarn(c, fmt.Sprintf("user sensitive words detected: %s", strings.Join(words, ", ")))
@@ -171,32 +157,16 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 		}
 	}()
 
-	retryParam := &service.RetryParam{
-		Ctx:        c,
-		TokenGroup: relayInfo.TokenGroup,
-		ModelName:  relayInfo.OriginModelName,
-		Retry:      common.GetPointer(0),
-	}
-
-	for ; retryParam.GetRetry() <= common.RetryTimes; retryParam.IncreaseRetry() {
-		channel, channelErr := getChannel(c, relayInfo, retryParam)
-		if channelErr != nil {
-			logger.LogError(c, channelErr.Error())
-			newAPIError = channelErr
+	for i := 0; i <= common.RetryTimes; i++ {
+		channel, err := getChannel(c, group, originalModel, i)
+		if err != nil {
+			logger.LogError(c, err.Error())
+			newAPIError = err
 			break
 		}
 
 		addUsedChannel(c, channel.Id)
-		requestBody, bodyErr := common.GetRequestBody(c)
-		if bodyErr != nil {
-			// Ensure consistent 413 for oversized bodies even when error occurs later (e.g., retry path)
-			if common.IsRequestBodyTooLargeError(bodyErr) || errors.Is(bodyErr, common.ErrRequestBodyTooLarge) {
-				newAPIError = types.NewErrorWithStatusCode(bodyErr, types.ErrorCodeReadRequestBodyFailed, http.StatusRequestEntityTooLarge, types.ErrOptionWithSkipRetry())
-			} else {
-				newAPIError = types.NewErrorWithStatusCode(bodyErr, types.ErrorCodeReadRequestBodyFailed, http.StatusBadRequest, types.ErrOptionWithSkipRetry())
-			}
-			break
-		}
+		requestBody, _ := common.GetRequestBody(c)
 		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 
 		switch relayFormat {
@@ -216,7 +186,7 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 		processChannelError(c, *types.NewChannelError(channel.Id, channel.Type, channel.Name, channel.ChannelInfo.IsMultiKey, common.GetContextKeyString(c, constant.ContextKeyChannelKey), channel.GetAutoBan()), newAPIError)
 
-		if !shouldRetry(c, newAPIError, common.RetryTimes-retryParam.GetRetry()) {
+		if !shouldRetry(c, newAPIError, common.RetryTimes-i) {
 			break
 		}
 	}
@@ -241,35 +211,8 @@ func addUsedChannel(c *gin.Context, channelId int) {
 	c.Set("use_channel", useChannel)
 }
 
-func fastTokenCountMetaForPricing(request dto.Request) *types.TokenCountMeta {
-	if request == nil {
-		return &types.TokenCountMeta{}
-	}
-	meta := &types.TokenCountMeta{
-		TokenType: types.TokenTypeTokenizer,
-	}
-	switch r := request.(type) {
-	case *dto.GeneralOpenAIRequest:
-		if r.MaxCompletionTokens > r.MaxTokens {
-			meta.MaxTokens = int(r.MaxCompletionTokens)
-		} else {
-			meta.MaxTokens = int(r.MaxTokens)
-		}
-	case *dto.OpenAIResponsesRequest:
-		meta.MaxTokens = int(r.MaxOutputTokens)
-	case *dto.ClaudeRequest:
-		meta.MaxTokens = int(r.MaxTokens)
-	case *dto.ImageRequest:
-		// Pricing for image requests depends on ImagePriceRatio; safe to compute even when CountToken is disabled.
-		return r.GetTokenCountMeta()
-	default:
-		// Best-effort: leave CombineText empty to avoid large allocations.
-	}
-	return meta
-}
-
-func getChannel(c *gin.Context, info *relaycommon.RelayInfo, retryParam *service.RetryParam) (*model.Channel, *types.NewAPIError) {
-	if info.ChannelMeta == nil {
+func getChannel(c *gin.Context, group, originalModel string, retryCount int) (*model.Channel, *types.NewAPIError) {
+	if retryCount == 0 {
 		autoBan := c.GetBool("auto_ban")
 		autoBanInt := 1
 		if !autoBan {
@@ -282,18 +225,14 @@ func getChannel(c *gin.Context, info *relaycommon.RelayInfo, retryParam *service
 			AutoBan: &autoBanInt,
 		}, nil
 	}
-	channel, selectGroup, err := service.CacheGetRandomSatisfiedChannel(retryParam)
-
-	info.PriceData.GroupRatioInfo = helper.HandleGroupRatio(c, info)
-
+	channel, selectGroup, err := service.CacheGetRandomSatisfiedChannel(c, group, originalModel, retryCount)
 	if err != nil {
-		return nil, types.NewError(fmt.Errorf("获取分组 %s 下模型 %s 的可用渠道失败（retry）: %s", selectGroup, info.OriginModelName, err.Error()), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
+		return nil, types.NewError(fmt.Errorf("获取分组 %s 下模型 %s 的可用渠道失败（retry）: %s", selectGroup, originalModel, err.Error()), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
 	}
 	if channel == nil {
-		return nil, types.NewError(fmt.Errorf("分组 %s 下模型 %s 的可用渠道不存在（retry）", selectGroup, info.OriginModelName), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
+		return nil, types.NewError(fmt.Errorf("分组 %s 下模型 %s 的可用渠道不存在（retry）", selectGroup, originalModel), types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
 	}
-
-	newAPIError := middleware.SetupContextForSelectedChannel(c, channel, info.OriginModelName)
+	newAPIError := middleware.SetupContextForSelectedChannel(c, channel, originalModel)
 	if newAPIError != nil {
 		return nil, newAPIError
 	}
@@ -346,7 +285,7 @@ func processChannelError(c *gin.Context, channelError types.ChannelError, err *t
 	logger.LogError(c, fmt.Sprintf("channel error (channel #%d, status code: %d): %s", channelError.ChannelId, err.StatusCode, err.Error()))
 	// 不要使用context获取渠道信息，异步处理时可能会出现渠道信息不一致的情况
 	// do not use context to get channel info, there may be inconsistent channel info when processing asynchronously
-	if service.ShouldDisableChannel(channelError.ChannelType, err) && channelError.AutoBan {
+	if service.ShouldDisableChannel(channelError.ChannelId, err) && channelError.AutoBan {
 		gopool.Go(func() {
 			service.DisableChannel(channelError, err.Error())
 		})
@@ -427,7 +366,7 @@ func RelayMidjourney(c *gin.Context) {
 }
 
 func RelayNotImplemented(c *gin.Context) {
-	err := types.OpenAIError{
+	err := dto.OpenAIError{
 		Message: "API not implemented",
 		Type:    "new_api_error",
 		Param:   "",
@@ -439,7 +378,7 @@ func RelayNotImplemented(c *gin.Context) {
 }
 
 func RelayNotFound(c *gin.Context) {
-	err := types.OpenAIError{
+	err := dto.OpenAIError{
 		Message: fmt.Sprintf("Invalid URL (%s %s)", c.Request.Method, c.Request.URL.Path),
 		Type:    "invalid_request_error",
 		Param:   "",
@@ -453,6 +392,8 @@ func RelayNotFound(c *gin.Context) {
 func RelayTask(c *gin.Context) {
 	retryTimes := common.RetryTimes
 	channelId := c.GetInt("channel_id")
+	group := c.GetString("group")
+	originalModel := c.GetString("original_model")
 	c.Set("use_channel", []string{fmt.Sprintf("%d", channelId)})
 	relayInfo, err := relaycommon.GenRelayInfo(c, types.RelayFormatTask, nil, nil)
 	if err != nil {
@@ -462,14 +403,8 @@ func RelayTask(c *gin.Context) {
 	if taskErr == nil {
 		retryTimes = 0
 	}
-	retryParam := &service.RetryParam{
-		Ctx:        c,
-		TokenGroup: relayInfo.TokenGroup,
-		ModelName:  relayInfo.OriginModelName,
-		Retry:      common.GetPointer(0),
-	}
-	for ; shouldRetryTaskRelay(c, channelId, taskErr, retryTimes) && retryParam.GetRetry() < retryTimes; retryParam.IncreaseRetry() {
-		channel, newAPIError := getChannel(c, relayInfo, retryParam)
+	for i := 0; shouldRetryTaskRelay(c, channelId, taskErr, retryTimes) && i < retryTimes; i++ {
+		channel, newAPIError := getChannel(c, group, originalModel, i)
 		if newAPIError != nil {
 			logger.LogError(c, fmt.Sprintf("CacheGetRandomSatisfiedChannel failed: %s", newAPIError.Error()))
 			taskErr = service.TaskErrorWrapperLocal(newAPIError.Err, "get_channel_failed", http.StatusInternalServerError)
@@ -479,18 +414,10 @@ func RelayTask(c *gin.Context) {
 		useChannel := c.GetStringSlice("use_channel")
 		useChannel = append(useChannel, fmt.Sprintf("%d", channelId))
 		c.Set("use_channel", useChannel)
-		logger.LogInfo(c, fmt.Sprintf("using channel #%d to retry (remain times %d)", channel.Id, retryParam.GetRetry()))
+		logger.LogInfo(c, fmt.Sprintf("using channel #%d to retry (remain times %d)", channel.Id, i))
 		//middleware.SetupContextForSelectedChannel(c, channel, originalModel)
 
-		requestBody, err := common.GetRequestBody(c)
-		if err != nil {
-			if common.IsRequestBodyTooLargeError(err) || errors.Is(err, common.ErrRequestBodyTooLarge) {
-				taskErr = service.TaskErrorWrapperLocal(err, "read_request_body_failed", http.StatusRequestEntityTooLarge)
-			} else {
-				taskErr = service.TaskErrorWrapperLocal(err, "read_request_body_failed", http.StatusBadRequest)
-			}
-			break
-		}
+		requestBody, _ := common.GetRequestBody(c)
 		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 		taskErr = taskRelayHandler(c, relayInfo)
 	}

controller/task.go

@@ -88,7 +88,7 @@ func UpdateSunoTaskAll(ctx context.Context, taskChannelM map[int][]string, taskM
 	for channelId, taskIds := range taskChannelM {
 		err := updateSunoTaskAll(ctx, channelId, taskIds, taskM)
 		if err != nil {
-			logger.LogError(ctx, fmt.Sprintf("渠道 #%d 更新异步任务失败: %s", channelId, err.Error()))
+			logger.LogError(ctx, fmt.Sprintf("渠道 #%d 更新异步任务失败: %d", channelId, err.Error()))
 		}
 	}
 	return nil
@@ -116,10 +116,9 @@ func updateSunoTaskAll(ctx context.Context, channelId int, taskIds []string, tas
 	if adaptor == nil {
 		return errors.New("adaptor not found")
 	}
-	proxy := channel.GetSetting().Proxy
 	resp, err := adaptor.FetchTask(*channel.BaseURL, channel.Key, map[string]any{
 		"ids": taskIds,
-	}, proxy)
+	})
 	if err != nil {
 		common.SysLog(fmt.Sprintf("Get Task Do req error: %v", err))
 		return err
@@ -141,7 +140,7 @@ func updateSunoTaskAll(ctx context.Context, channelId int, taskIds []string, tas
 		return err
 	}
 	if !responseItems.IsSuccess() {
-		common.SysLog(fmt.Sprintf("渠道 #%d 未完成的任务有: %d, 成功获取到任务数: %s", channelId, len(taskIds), string(responseBody)))
+		common.SysLog(fmt.Sprintf("渠道 #%d 未完成的任务有: %d, 成功获取到任务数: %d", channelId, len(taskIds), string(responseBody)))
 		return err
 	}
 

controller/task_video.go

@@ -67,7 +67,6 @@ func updateVideoSingleTask(ctx context.Context, adaptor channel.TaskAdaptor, cha
 	if channel.GetBaseURL() != "" {
 		baseURL = channel.GetBaseURL()
 	}
-	proxy := channel.GetSetting().Proxy
 
 	task := taskM[taskId]
 	if task == nil {
@@ -77,7 +76,7 @@ func updateVideoSingleTask(ctx context.Context, adaptor channel.TaskAdaptor, cha
 	resp, err := adaptor.FetchTask(baseURL, channel.Key, map[string]any{
 		"task_id": taskId,
 		"action":  task.Action,
-	}, proxy)
+	})
 	if err != nil {
 		return fmt.Errorf("fetchTask failed for task %s: %w", taskId, err)
 	}

controller/token.go

@@ -142,7 +142,7 @@ func AddToken(c *gin.Context) {
 		common.ApiError(c, err)
 		return
 	}
-	if len(token.Name) > 50 {
+	if len(token.Name) > 30 {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
 			"message": "令牌名称过长",
@@ -171,7 +171,6 @@ func AddToken(c *gin.Context) {
 		ModelLimits:        token.ModelLimits,
 		AllowIps:           token.AllowIps,
 		Group:              token.Group,
-		CrossGroupRetry:    token.CrossGroupRetry,
 	}
 	err = cleanToken.Insert()
 	if err != nil {
@@ -209,7 +208,7 @@ func UpdateToken(c *gin.Context) {
 		common.ApiError(c, err)
 		return
 	}
-	if len(token.Name) > 50 {
+	if len(token.Name) > 30 {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
 			"message": "令牌名称过长",
@@ -249,7 +248,6 @@ func UpdateToken(c *gin.Context) {
 		cleanToken.ModelLimits = token.ModelLimits
 		cleanToken.AllowIps = token.AllowIps
 		cleanToken.Group = token.Group
-		cleanToken.CrossGroupRetry = token.CrossGroupRetry
 	}
 	err = cleanToken.Update()
 	if err != nil {

controller/topup_creem.go

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

controller/video_proxy.go

@@ -1,7 +1,6 @@
 package controller
 
 import (
-	"context"
 	"fmt"
 	"io"
 	"net/http"
@@ -11,7 +10,6 @@ import (
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/model"
-	"github.com/QuantumNous/new-api/service"
 
 	"github.com/gin-gonic/gin"
 )
@@ -77,22 +75,11 @@ func VideoProxy(c *gin.Context) {
 	}
 
 	var videoURL string
-	proxy := channel.GetSetting().Proxy
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		logger.LogError(c.Request.Context(), fmt.Sprintf("Failed to create proxy client for task %s: %s", taskID, err.Error()))
-		c.JSON(http.StatusInternalServerError, gin.H{
-			"error": gin.H{
-				"message": "Failed to create proxy client",
-				"type":    "server_error",
-			},
-		})
-		return
+	client := &http.Client{
+		Timeout: 60 * time.Second,
 	}
 
-	ctx, cancel := context.WithTimeout(c.Request.Context(), 60*time.Second)
-	defer cancel()
-	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "", nil)
+	req, err := http.NewRequestWithContext(c.Request.Context(), http.MethodGet, "", nil)
 	if err != nil {
 		logger.LogError(c.Request.Context(), fmt.Sprintf("Failed to create request: %s", err.Error()))
 		c.JSON(http.StatusInternalServerError, gin.H{

controller/video_proxy_gemini.go

@@ -35,11 +35,10 @@ func getGeminiVideoURL(channel *model.Channel, task *model.Task, apiKey string)
 		return "", fmt.Errorf("api key not available for task")
 	}
 
-	proxy := channel.GetSetting().Proxy
 	resp, err := adaptor.FetchTask(baseURL, apiKey, map[string]any{
 		"task_id": task.TaskID,
 		"action":  task.Action,
-	}, proxy)
+	})
 	if err != nil {
 		return "", fmt.Errorf("fetch task failed: %w", err)
 	}

dto/audio.go

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

dto/error.go

@@ -1,31 +1,26 @@
 package dto
 
-import (
-	"encoding/json"
+import "github.com/QuantumNous/new-api/types"
 
-	"github.com/QuantumNous/new-api/common"
-	"github.com/QuantumNous/new-api/types"
-)
-
-//type OpenAIError struct {
-//	Message string `json:"message"`
-//	Type    string `json:"type"`
-//	Param   string `json:"param"`
-//	Code    any    `json:"code"`
-//}
+type OpenAIError struct {
+	Message string `json:"message"`
+	Type    string `json:"type"`
+	Param   string `json:"param"`
+	Code    any    `json:"code"`
+}
 
 type OpenAIErrorWithStatusCode struct {
-	Error      types.OpenAIError `json:"error"`
-	StatusCode int               `json:"status_code"`
+	Error      OpenAIError `json:"error"`
+	StatusCode int         `json:"status_code"`
 	LocalError bool
 }
 
 type GeneralErrorResponse struct {
-	Error    json.RawMessage `json:"error"`
-	Message  string          `json:"message"`
-	Msg      string          `json:"msg"`
-	Err      string          `json:"err"`
-	ErrorMsg string          `json:"error_msg"`
+	Error    types.OpenAIError `json:"error"`
+	Message  string            `json:"message"`
+	Msg      string            `json:"msg"`
+	Err      string            `json:"err"`
+	ErrorMsg string            `json:"error_msg"`
 	Header   struct {
 		Message string `json:"message"`
 	} `json:"header"`
@@ -36,35 +31,9 @@ type GeneralErrorResponse struct {
 	} `json:"response"`
 }
 
-func (e GeneralErrorResponse) TryToOpenAIError() *types.OpenAIError {
-	var openAIError types.OpenAIError
-	if len(e.Error) > 0 {
-		err := common.Unmarshal(e.Error, &openAIError)
-		if err == nil && openAIError.Message != "" {
-			return &openAIError
-		}
-	}
-	return nil
-}
-
 func (e GeneralErrorResponse) ToMessage() string {
-	if len(e.Error) > 0 {
-		switch common.GetJsonType(e.Error) {
-		case "object":
-			var openAIError types.OpenAIError
-			err := common.Unmarshal(e.Error, &openAIError)
-			if err == nil && openAIError.Message != "" {
-				return openAIError.Message
-			}
-		case "string":
-			var msg string
-			err := common.Unmarshal(e.Error, &msg)
-			if err == nil && msg != "" {
-				return msg
-			}
-		default:
-			return string(e.Error)
-		}
+	if e.Error.Message != "" {
+		return e.Error.Message
 	}
 	if e.Message != "" {
 		return e.Message

dto/openai_image.go

@@ -27,11 +27,8 @@ type ImageRequest struct {
 	OutputCompression json.RawMessage `json:"output_compression,omitempty"`
 	PartialImages     json.RawMessage `json:"partial_images,omitempty"`
 	// Stream            bool            `json:"stream,omitempty"`
-	Watermark *bool `json:"watermark,omitempty"`
-	// zhipu 4v
-	WatermarkEnabled json.RawMessage `json:"watermark_enabled,omitempty"`
-	UserId           json.RawMessage `json:"user_id,omitempty"`
-	Image            json.RawMessage `json:"image,omitempty"`
+	Watermark *bool           `json:"watermark,omitempty"`
+	Image     json.RawMessage `json:"image,omitempty"`
 	// 用匿名参数接收额外参数
 	Extra map[string]json.RawMessage `json:"-"`
 }

dto/openai_request.go

@@ -83,7 +83,6 @@ type GeneralOpenAIRequest struct {
 	// Ali Qwen Params
 	VlHighResolutionImages json.RawMessage `json:"vl_high_resolution_images,omitempty"`
 	EnableThinking         any             `json:"enable_thinking,omitempty"`
-	ChatTemplateKwargs     json.RawMessage `json:"chat_template_kwargs,omitempty"`
 	// ollama Params
 	Think json.RawMessage `json:"think,omitempty"`
 	// baidu v2

go.mod

@@ -33,7 +33,7 @@ require (
 	github.com/mewkiz/flac v1.0.13
 	github.com/pkg/errors v0.9.1
 	github.com/pquerna/otp v1.5.0
-	github.com/samber/lo v1.52.0
+	github.com/samber/lo v1.39.0
 	github.com/shirou/gopsutil v3.21.11+incompatible
 	github.com/shopspring/decimal v1.4.0
 	github.com/stripe/stripe-go/v81 v81.4.0
@@ -99,7 +99,6 @@ require (
 	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
-	github.com/ncruces/go-strftime v0.1.9 // indirect
 	github.com/pelletier/go-toml/v2 v2.2.1 // indirect
 	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
 	github.com/tidwall/match v1.1.1 // indirect
@@ -111,13 +110,13 @@ require (
 	github.com/x448/float16 v0.8.4 // indirect
 	github.com/yusufpapurcu/wmi v1.2.3 // indirect
 	golang.org/x/arch v0.21.0 // indirect
-	golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b // indirect
+	golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 // 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 v1.66.10 // indirect
-	modernc.org/mathutil v1.7.1 // indirect
-	modernc.org/memory v1.11.0 // indirect
-	modernc.org/sqlite v1.40.1 // indirect
+	modernc.org/libc v1.22.5 // indirect
+	modernc.org/mathutil v1.5.0 // indirect
+	modernc.org/memory v1.5.0 // indirect
+	modernc.org/sqlite v1.23.1 // indirect
 )

go.sum

@@ -120,7 +120,6 @@ github.com/google/go-tpm v0.9.5/go.mod h1:h9jEsEECg7gtLis0upRBQU+GhYVH6jMjrFxI8u
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
 github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26 h1:Xim43kblpZXfIBQsbuBVKCudVG457BR2GZFIz3uw3hQ=
 github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26/go.mod h1:dDKJzRmX4S37WGHujM7tX//fmj1uioxKzKxz3lo4HJo=
-github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e h1:ijClszYn+mADRFY17kjQEVQ1XRhq2/JR1M3sGqeJoxs=
 github.com/google/uuid v1.1.2/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
 github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
@@ -194,8 +193,6 @@ github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJ
 github.com/modern-go/reflect2 v0.0.0-20180701023420-4b7aa43c6742/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/modern-go/reflect2 v1.0.2 h1:xBagoLtFs94CBntxluKeaWgTMpvLxC4ur3nMaC9Gz0M=
 github.com/modern-go/reflect2 v1.0.2/go.mod h1:yWuevngMOJpCy52FWWMvUC8ws7m/LJsjYzDa0/r8luk=
-github.com/ncruces/go-strftime v0.1.9 h1:bY0MQC28UADQmHmaF5dgpLmImcShSi2kHU9XLdhx/f4=
-github.com/ncruces/go-strftime v0.1.9/go.mod h1:Fwc5htZGVVkseilnfgOVb9mKy6w1naJmn9CehxcKcls=
 github.com/nxadm/tail v1.4.8 h1:nPr65rt6Y5JFSKQO7qToXr7pePgD6Gwiw05lkbyAQTE=
 github.com/nxadm/tail v1.4.8/go.mod h1:+ncqLTQzXmGhMZNUePPaPqPvBxHAIsmXswZKocGu+AU=
 github.com/onsi/ginkgo v1.16.5 h1:8xi0RTUf59SOSfEtZMvwTvXYMzG4gV23XVHOZiXNtnE=
@@ -222,8 +219,6 @@ github.com/rogpeppe/go-internal v1.8.0 h1:FCbCCtXNOY3UtUuHUYaghJg4y7Fd14rXifAYUA
 github.com/rogpeppe/go-internal v1.8.0/go.mod h1:WmiCO8CzOY8rg0OYDC4/i/2WRWAB6poM+XZ2dLUbcbE=
 github.com/samber/lo v1.39.0 h1:4gTz1wUhNYLhFSKl6O+8peW0v2F4BCY034GRpU9WnuA=
 github.com/samber/lo v1.39.0/go.mod h1:+m/ZKRl6ClXCE2Lgf3MsQlWfh4bn1bz6CXEOxnEXnEA=
-github.com/samber/lo v1.52.0 h1:Rvi+3BFHES3A8meP33VPAxiBZX/Aws5RxrschYGjomw=
-github.com/samber/lo v1.52.0/go.mod h1:4+MXEGsJzbKGaUEQFKBq2xtfuznW9oz/WrgyzMzRoM0=
 github.com/shirou/gopsutil v3.21.11+incompatible h1:+1+c1VGhc88SSonWP6foOcLhvnKlUeu/erjjvaPEYiI=
 github.com/shirou/gopsutil v3.21.11+incompatible/go.mod h1:5b4v6he4MtMOwMlS0TUMTu2PcXUg8+E1lC7eC3UO/RA=
 github.com/shopspring/decimal v1.4.0 h1:bxl37RwXBklmTi0C79JfXCEBD1cqqHt0bbgBAGFp81k=
@@ -290,8 +285,6 @@ golang.org/x/crypto v0.45.0 h1:jMBrvKuj23MTlT0bQEOBcAE0mjg8mK9RXFhRH6nyF3Q=
 golang.org/x/crypto v0.45.0/go.mod h1:XTGrrkGJve7CYK7J8PEww4aY7gM3qMCElcJQ8n8JdX4=
 golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 h1:985EYyeCOxTpcgOTJpflJUwOeEz0CQOdPt73OzpE9F8=
 golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0/go.mod h1:/lliqkxwWAhPjf5oSOIJup2XcqJaw8RGS6k3TGEc7GI=
-golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b h1:M2rDM6z3Fhozi9O7NWsxAkg/yqS/lQJ6PmkyIV3YP+o=
-golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b/go.mod h1:3//PLf8L/X+8b4vuAfHzxeRUl04Adcb341+IGKfnqS8=
 golang.org/x/image v0.23.0 h1:HseQ7c2OpPKTPVzNjG5fwJsOTCiiwS4QdsYi5XU6H68=
 golang.org/x/image v0.23.0/go.mod h1:wJJBTdLfCCf3tiHa1fNxpZmUI4mmoZvwMCPP0ddoNKY=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
@@ -352,17 +345,9 @@ gorm.io/gorm v1.25.2 h1:gs1o6Vsa+oVKG/a9ElL3XgyGfghFfkKA2SInQaCyMho=
 gorm.io/gorm v1.25.2/go.mod h1:L4uxeKpfBml98NYqVqwAdmV1a2nBtAec/cf3fpucW/k=
 modernc.org/libc v1.22.5 h1:91BNch/e5B0uPbJFgqbxXuOnxBQjlS//icfQEGmvyjE=
 modernc.org/libc v1.22.5/go.mod h1:jj+Z7dTNX8fBScMVNRAYZ/jF91K8fdT2hYMThc3YjBY=
-modernc.org/libc v1.66.10 h1:yZkb3YeLx4oynyR+iUsXsybsX4Ubx7MQlSYEw4yj59A=
-modernc.org/libc v1.66.10/go.mod h1:8vGSEwvoUoltr4dlywvHqjtAqHBaw0j1jI7iFBTAr2I=
 modernc.org/mathutil v1.5.0 h1:rV0Ko/6SfM+8G+yKiyI830l3Wuz1zRutdslNoQ0kfiQ=
 modernc.org/mathutil v1.5.0/go.mod h1:mZW8CKdRPY1v87qxC/wUdX5O1qDzXMP5TH3wjfpga6E=
-modernc.org/mathutil v1.7.1 h1:GCZVGXdaN8gTqB1Mf/usp1Y/hSqgI2vAGGP4jZMCxOU=
-modernc.org/mathutil v1.7.1/go.mod h1:4p5IwJITfppl0G4sUEDtCr4DthTaT47/N3aT6MhfgJg=
 modernc.org/memory v1.5.0 h1:N+/8c5rE6EqugZwHii4IFsaJ7MUhoWX07J5tC/iI5Ds=
 modernc.org/memory v1.5.0/go.mod h1:PkUhL0Mugw21sHPeskwZW4D6VscE/GQJOnIpCnW6pSU=
-modernc.org/memory v1.11.0 h1:o4QC8aMQzmcwCK3t3Ux/ZHmwFPzE6hf2Y5LbkRs+hbI=
-modernc.org/memory v1.11.0/go.mod h1:/JP4VbVC+K5sU2wZi9bHoq2MAkCnrt2r98UGeSK7Mjw=
 modernc.org/sqlite v1.23.1 h1:nrSBg4aRQQwq59JpvGEQ15tNxoO5pX/kUjcRNwSAGQM=
 modernc.org/sqlite v1.23.1/go.mod h1:OrDj17Mggn6MhE+iPbBNf7RGKODDE9NFT0f3EwDzJqk=
-modernc.org/sqlite v1.40.1 h1:VfuXcxcUWWKRBuP8+BR9L7VnmusMgBNNnBYGEe9w/iY=
-modernc.org/sqlite v1.40.1/go.mod h1:9fjQZ0mB1LLP0GYrp39oOJXx/I2sxEnZtzCmEQIKvGE=

middleware/auth.go

@@ -2,14 +2,12 @@ package middleware
 
 import (
 	"fmt"
-	"net"
 	"net/http"
 	"strconv"
 	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
-	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
@@ -242,20 +240,13 @@ func TokenAuth() func(c *gin.Context) {
 			return
 		}
 
-		allowIps := token.GetIpLimits()
-		if len(allowIps) > 0 {
+		allowIpsMap := token.GetIpLimitsMap()
+		if len(allowIpsMap) != 0 {
 			clientIp := c.ClientIP()
-			logger.LogDebug(c, "Token has IP restrictions, checking client IP %s", clientIp)
-			ip := net.ParseIP(clientIp)
-			if ip == nil {
-				abortWithOpenAiMessage(c, http.StatusForbidden, "无法解析客户端 IP 地址")
-				return
-			}
-			if common.IsIpInCIDRList(ip, allowIps) == false {
+			if _, ok := allowIpsMap[clientIp]; !ok {
 				abortWithOpenAiMessage(c, http.StatusForbidden, "您的 IP 不在令牌允许访问的列表中")
 				return
 			}
-			logger.LogDebug(c, "Client IP %s passed the token IP restrictions check", clientIp)
 		}
 
 		userCache, err := model.GetUserCache(token.UserId)
@@ -316,8 +307,7 @@ func SetupContextForToken(c *gin.Context, token *model.Token, parts ...string) e
 	} else {
 		c.Set("token_model_limit_enabled", false)
 	}
-	common.SetContextKey(c, constant.ContextKeyTokenGroup, token.Group)
-	common.SetContextKey(c, constant.ContextKeyTokenCrossGroupRetry, token.CrossGroupRetry)
+	c.Set("token_group", token.Group)
 	if len(parts) > 1 {
 		if model.IsAdmin(token.UserId) {
 			c.Set("specific_channel_id", parts[1])

middleware/distributor.go

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

middleware/gzip.go

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

model/channel.go

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

model/token.go

@@ -6,6 +6,7 @@ import (
 	"strings"
 
 	"github.com/QuantumNous/new-api/common"
+
 	"github.com/bytedance/gopkg/util/gopool"
 	"gorm.io/gorm"
 )
@@ -26,7 +27,6 @@ type Token struct {
 	AllowIps           *string        `json:"allow_ips" gorm:"default:''"`
 	UsedQuota          int            `json:"used_quota" gorm:"default:0"` // used quota
 	Group              string         `json:"group" gorm:"default:''"`
-	CrossGroupRetry    bool           `json:"cross_group_retry" gorm:"default:false"` // 跨分组重试，仅auto分组有效
 	DeletedAt          gorm.DeletedAt `gorm:"index"`
 }
 
@@ -34,26 +34,26 @@ func (token *Token) Clean() {
 	token.Key = ""
 }
 
-func (token *Token) GetIpLimits() []string {
+func (token *Token) GetIpLimitsMap() map[string]any {
 	// delete empty spaces
 	//split with \n
-	ipLimits := make([]string, 0)
+	ipLimitsMap := make(map[string]any)
 	if token.AllowIps == nil {
-		return ipLimits
+		return ipLimitsMap
 	}
 	cleanIps := strings.ReplaceAll(*token.AllowIps, " ", "")
 	if cleanIps == "" {
-		return ipLimits
+		return ipLimitsMap
 	}
 	ips := strings.Split(cleanIps, "\n")
 	for _, ip := range ips {
 		ip = strings.TrimSpace(ip)
 		ip = strings.ReplaceAll(ip, ",", "")
-		if ip != "" {
-			ipLimits = append(ipLimits, ip)
+		if common.IsIP(ip) {
+			ipLimitsMap[ip] = true
 		}
 	}
-	return ipLimits
+	return ipLimitsMap
 }
 
 func GetAllUserTokens(userId int, startIdx int, num int) ([]*Token, error) {
@@ -185,7 +185,7 @@ func (token *Token) Update() (err error) {
 		}
 	}()
 	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota",
-		"model_limits_enabled", "model_limits", "allow_ips", "group", "cross_group_retry").Updates(token).Error
+		"model_limits_enabled", "model_limits", "allow_ips", "group").Updates(token).Error
 	return err
 }
 

relay/audio_handler.go

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

relay/channel/adapter.go

@@ -47,7 +47,7 @@ type TaskAdaptor interface {
 	GetChannelName() string
 
 	// FetchTask
-	FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error)
+	FetchTask(baseUrl, key string, body map[string]any) (*http.Response, error)
 
 	ParseTaskResult(respBody []byte) (*relaycommon.TaskInfo, error)
 }

relay/channel/aws/constants.go

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

relay/channel/aws/relay-aws.go

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

relay/channel/baidu/relay-baidu.go

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

relay/channel/claude/constants.go

@@ -9,7 +9,6 @@ var ModelList = []string{
 	"claude-3-opus-20240229",
 	"claude-3-haiku-20240307",
 	"claude-3-5-haiku-20241022",
-	"claude-haiku-4-5-20251001",
 	"claude-3-5-sonnet-20240620",
 	"claude-3-5-sonnet-20241022",
 	"claude-3-7-sonnet-20250219",

relay/channel/coze/relay-coze.go

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

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

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

relay/channel/openai/adaptor.go

@@ -42,7 +42,7 @@ type Adaptor struct {
 // support OAI models: o1-mini/o3-mini/o4-mini/o1/o3 etc...
 // minimal effort only available in gpt-5
 func parseReasoningEffortFromModelSuffix(model string) (string, string) {
-	effortSuffixes := []string{"-high", "-minimal", "-low", "-medium", "-none", "-xhigh"}
+	effortSuffixes := []string{"-high", "-minimal", "-low", "-medium", "-none"}
 	for _, suffix := range effortSuffixes {
 		if strings.HasSuffix(model, suffix) {
 			effort := strings.TrimPrefix(suffix, "-")
@@ -306,11 +306,10 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 			request.Temperature = nil
 		}
 
-		// gpt-5系列模型适配 归零不再支持的参数
 		if strings.HasPrefix(info.UpstreamModelName, "gpt-5") {
-			request.Temperature = nil
-			request.TopP = 0 // oai 的 top_p 默认值是 1.0，但是为了 omitempty 属性直接不传，这里显式设置为 0
-			request.LogProbs = false
+			if info.UpstreamModelName != "gpt-5-chat-latest" {
+				request.Temperature = nil
+			}
 		}
 
 		// 转换模型推理力度后缀

relay/channel/openai/audio.go

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

relay/channel/openai/helper.go

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

relay/channel/openai/relay-openai.go

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

relay/channel/task/ali/adaptor.go

@@ -393,7 +393,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 }
 
 // FetchTask 查询任务状态
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -408,11 +408,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 
 	req.Header.Set("Authorization", "Bearer "+key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/doubao/adaptor.go

@@ -146,7 +146,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -163,11 +163,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Content-Type", "application/json")
 	req.Header.Set("Authorization", "Bearer "+key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/gemini/adaptor.go

@@ -200,7 +200,7 @@ func (a *TaskAdaptor) GetChannelName() string {
 }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -223,11 +223,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Accept", "application/json")
 	req.Header.Set("x-goog-api-key", key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) ParseTaskResult(respBody []byte) (*relaycommon.TaskInfo, error) {

relay/channel/task/hailuo/adaptor.go

@@ -110,7 +110,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 	return hResp.TaskID, responseBody, nil
 }
 
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -126,11 +126,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Accept", "application/json")
 	req.Header.Set("Authorization", "Bearer "+key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/jimeng/adaptor.go

@@ -196,7 +196,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 	}
 
 	if jResp.Code != 10000 {
-		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", jResp.Message), fmt.Sprintf("%d", jResp.Code), http.StatusInternalServerError)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf(jResp.Message), fmt.Sprintf("%d", jResp.Code), http.StatusInternalServerError)
 		return
 	}
 
@@ -210,7 +210,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -251,11 +251,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 			return nil, errors.Wrap(err, "sign request failed")
 		}
 	}
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/kling/adaptor.go

@@ -186,7 +186,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 		return
 	}
 	if kResp.Code != 0 {
-		taskErr = service.TaskErrorWrapperLocal(fmt.Errorf("%s", kResp.Message), "task_failed", http.StatusBadRequest)
+		taskErr = service.TaskErrorWrapperLocal(fmt.Errorf(kResp.Message), "task_failed", http.StatusBadRequest)
 		return
 	}
 	ov := dto.NewOpenAIVideo()
@@ -199,7 +199,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -228,11 +228,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Authorization", "Bearer "+token)
 	req.Header.Set("User-Agent", "kling-sdk/1.0")
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/sora/adaptor.go

@@ -146,7 +146,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, _ *relayco
 }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -161,11 +161,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 
 	req.Header.Set("Authorization", "Bearer "+key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/task/suno/adaptor.go

@@ -105,7 +105,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 		return
 	}
 	if !sunoResponse.IsSuccess() {
-		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", sunoResponse.Message), sunoResponse.Code, http.StatusInternalServerError)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf(sunoResponse.Message), sunoResponse.Code, http.StatusInternalServerError)
 		return
 	}
 
@@ -132,7 +132,7 @@ func (a *TaskAdaptor) GetChannelName() string {
 	return ChannelName
 }
 
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any) (*http.Response, error) {
 	requestUrl := fmt.Sprintf("%s/suno/fetch", baseUrl)
 	byteBody, err := json.Marshal(body)
 	if err != nil {
@@ -153,11 +153,11 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req = req.WithContext(ctx)
 	req.Header.Set("Content-Type", "application/json")
 	req.Header.Set("Authorization", "Bearer "+key)
-	client, err := service.GetHttpClientWithProxy(proxy)
+	resp, err := service.GetHttpClient().Do(req)
 	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
+		return nil, err
 	}
-	return client.Do(req)
+	return resp, nil
 }
 
 func actionValidate(c *gin.Context, sunoRequest *dto.SunoSubmitReq, action string) (err error) {

relay/channel/task/vertex/adaptor.go

@@ -120,11 +120,7 @@ func (a *TaskAdaptor) BuildRequestHeader(c *gin.Context, req *http.Request, info
 		return fmt.Errorf("failed to decode credentials: %w", err)
 	}
 
-	proxy := ""
-	if info != nil {
-		proxy = info.ChannelSetting.Proxy
-	}
-	token, err := vertexcore.AcquireAccessToken(*adc, proxy)
+	token, err := vertexcore.AcquireAccessToken(*adc, "")
 	if err != nil {
 		return fmt.Errorf("failed to acquire access token: %w", err)
 	}
@@ -220,7 +216,7 @@ func (a *TaskAdaptor) GetModelList() []string { return []string{"veo-3.0-generat
 func (a *TaskAdaptor) GetChannelName() string { return "vertex" }
 
 // FetchTask fetch task status
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -253,7 +249,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	if err := json.Unmarshal([]byte(key), adc); err != nil {
 		return nil, fmt.Errorf("failed to decode credentials: %w", err)
 	}
-	token, err := vertexcore.AcquireAccessToken(*adc, proxy)
+	token, err := vertexcore.AcquireAccessToken(*adc, "")
 	if err != nil {
 		return nil, fmt.Errorf("failed to acquire access token: %w", err)
 	}
@@ -265,11 +261,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Accept", "application/json")
 	req.Header.Set("Authorization", "Bearer "+token)
 	req.Header.Set("x-goog-user-project", adc.ProjectID)
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) ParseTaskResult(respBody []byte) (*relaycommon.TaskInfo, error) {

relay/channel/task/vidu/adaptor.go

@@ -188,7 +188,7 @@ func (a *TaskAdaptor) DoResponse(c *gin.Context, resp *http.Response, info *rela
 	return vResp.TaskId, responseBody, nil
 }
 
-func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy string) (*http.Response, error) {
+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")
@@ -204,11 +204,7 @@ func (a *TaskAdaptor) FetchTask(baseUrl, key string, body map[string]any, proxy
 	req.Header.Set("Accept", "application/json")
 	req.Header.Set("Authorization", "Token "+key)
 
-	client, err := service.GetHttpClientWithProxy(proxy)
-	if err != nil {
-		return nil, fmt.Errorf("new proxy http client failed: %w", err)
-	}
-	return client.Do(req)
+	return service.GetHttpClient().Do(req)
 }
 
 func (a *TaskAdaptor) GetModelList() []string {

relay/channel/vertex/adaptor.go

@@ -17,7 +17,6 @@ import (
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/relay/constant"
 	"github.com/QuantumNous/new-api/setting/model_setting"
-	"github.com/QuantumNous/new-api/setting/reasoning"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -51,43 +50,10 @@ type Adaptor struct {
 }
 
 func (a *Adaptor) ConvertGeminiRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeminiChatRequest) (any, error) {
-	// Vertex AI does not support functionResponse.id; keep it stripped here for consistency.
-	if model_setting.GetGeminiSettings().RemoveFunctionResponseIdEnabled {
-		removeFunctionResponseID(request)
-	}
 	geminiAdaptor := gemini.Adaptor{}
 	return geminiAdaptor.ConvertGeminiRequest(c, info, request)
 }
 
-func removeFunctionResponseID(request *dto.GeminiChatRequest) {
-	if request == nil {
-		return
-	}
-
-	if len(request.Contents) > 0 {
-		for i := range request.Contents {
-			if len(request.Contents[i].Parts) == 0 {
-				continue
-			}
-			for j := range request.Contents[i].Parts {
-				part := &request.Contents[i].Parts[j]
-				if part.FunctionResponse == nil {
-					continue
-				}
-				if len(part.FunctionResponse.ID) > 0 {
-					part.FunctionResponse.ID = nil
-				}
-			}
-		}
-	}
-
-	if len(request.Requests) > 0 {
-		for i := range request.Requests {
-			removeFunctionResponseID(&request.Requests[i])
-		}
-	}
-}
-
 func (a *Adaptor) ConvertClaudeRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.ClaudeRequest) (any, error) {
 	if v, ok := claudeModelMap[info.UpstreamModelName]; ok {
 		c.Set("request_model", v)
@@ -215,8 +181,6 @@ 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, ok := reasoning.TrimEffortSuffix(info.UpstreamModelName); ok && level != "" {
-				info.UpstreamModelName = baseModel
 			}
 		}
 

relay/channel/zhipu_4v/adaptor.go

@@ -36,7 +36,8 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 }
 
 func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
-	return request, nil
+	//TODO implement me
+	return nil, errors.New("not implemented")
 }
 
 func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
@@ -62,8 +63,6 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 				return fmt.Sprintf("%s/embeddings", specialPlan.OpenAIBaseURL), nil
 			}
 			return fmt.Sprintf("%s/api/paas/v4/embeddings", baseURL), nil
-		case relayconstant.RelayModeImagesGenerations:
-			return fmt.Sprintf("%s/api/paas/v4/images/generations", baseURL), nil
 		default:
 			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {
 				return fmt.Sprintf("%s/chat/completions", specialPlan.OpenAIBaseURL), nil
@@ -115,9 +114,6 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 			return claude.ClaudeHandler(c, resp, info, claude.RequestModeMessage)
 		}
 	default:
-		if info.RelayMode == relayconstant.RelayModeImagesGenerations {
-			return zhipu4vImageHandler(c, resp, info)
-		}
 		adaptor := openai.Adaptor{}
 		return adaptor.DoResponse(c, resp, info)
 	}

relay/channel/zhipu_4v/dto.go

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

relay/channel/zhipu_4v/image.go

@@ -1,127 +0,0 @@
-package zhipu_4v
-
-import (
-	"io"
-	"net/http"
-
-	"github.com/QuantumNous/new-api/common"
-	"github.com/QuantumNous/new-api/dto"
-	"github.com/QuantumNous/new-api/logger"
-	relaycommon "github.com/QuantumNous/new-api/relay/common"
-	"github.com/QuantumNous/new-api/service"
-	"github.com/QuantumNous/new-api/types"
-
-	"github.com/gin-gonic/gin"
-)
-
-type zhipuImageRequest struct {
-	Model            string `json:"model"`
-	Prompt           string `json:"prompt"`
-	Quality          string `json:"quality,omitempty"`
-	Size             string `json:"size,omitempty"`
-	WatermarkEnabled *bool  `json:"watermark_enabled,omitempty"`
-	UserID           string `json:"user_id,omitempty"`
-}
-
-type zhipuImageResponse struct {
-	Created       *int64            `json:"created,omitempty"`
-	Data          []zhipuImageData  `json:"data,omitempty"`
-	ContentFilter any               `json:"content_filter,omitempty"`
-	Usage         *dto.Usage        `json:"usage,omitempty"`
-	Error         *zhipuImageError  `json:"error,omitempty"`
-	RequestID     string            `json:"request_id,omitempty"`
-	ExtendParam   map[string]string `json:"extendParam,omitempty"`
-}
-
-type zhipuImageError struct {
-	Code    string `json:"code"`
-	Message string `json:"message"`
-}
-
-type zhipuImageData struct {
-	Url      string `json:"url,omitempty"`
-	ImageUrl string `json:"image_url,omitempty"`
-	B64Json  string `json:"b64_json,omitempty"`
-	B64Image string `json:"b64_image,omitempty"`
-}
-
-type openAIImagePayload struct {
-	Created int64             `json:"created"`
-	Data    []openAIImageData `json:"data"`
-}
-
-type openAIImageData struct {
-	B64Json string `json:"b64_json"`
-}
-
-func zhipu4vImageHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.Usage, *types.NewAPIError) {
-	responseBody, err := io.ReadAll(resp.Body)
-	if err != nil {
-		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
-	}
-	service.CloseResponseBodyGracefully(resp)
-
-	var zhipuResp zhipuImageResponse
-	if err := common.Unmarshal(responseBody, &zhipuResp); err != nil {
-		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
-	}
-
-	if zhipuResp.Error != nil && zhipuResp.Error.Message != "" {
-		return nil, types.WithOpenAIError(types.OpenAIError{
-			Message: zhipuResp.Error.Message,
-			Type:    "zhipu_image_error",
-			Code:    zhipuResp.Error.Code,
-		}, resp.StatusCode)
-	}
-
-	payload := openAIImagePayload{}
-	if zhipuResp.Created != nil && *zhipuResp.Created != 0 {
-		payload.Created = *zhipuResp.Created
-	} else {
-		payload.Created = info.StartTime.Unix()
-	}
-	for _, data := range zhipuResp.Data {
-		url := data.Url
-		if url == "" {
-			url = data.ImageUrl
-		}
-		if url == "" {
-			logger.LogWarn(c, "zhipu_image_missing_url")
-			continue
-		}
-
-		var b64 string
-		switch {
-		case data.B64Json != "":
-			b64 = data.B64Json
-		case data.B64Image != "":
-			b64 = data.B64Image
-		default:
-			_, downloaded, err := service.GetImageFromUrl(url)
-			if err != nil {
-				logger.LogError(c, "zhipu_image_get_b64_failed: "+err.Error())
-				continue
-			}
-			b64 = downloaded
-		}
-
-		if b64 == "" {
-			logger.LogWarn(c, "zhipu_image_empty_b64")
-			continue
-		}
-
-		imageData := openAIImageData{
-			B64Json: b64,
-		}
-		payload.Data = append(payload.Data, imageData)
-	}
-
-	jsonResp, err := common.Marshal(payload)
-	if err != nil {
-		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
-	}
-
-	service.IOCopyBytesGracefully(c, resp, jsonResp)
-
-	return &dto.Usage{}, nil
-}

relay/common/override.go

@@ -11,8 +11,6 @@ import (
 	"github.com/tidwall/sjson"
 )
 
-var negativeIndexRegexp = regexp.MustCompile(`\.(-\d+)`)
-
 type ConditionOperation struct {
 	Path           string      `json:"path"`             // JSON路径
 	Mode           string      `json:"mode"`             // full, prefix, suffix, contains, gt, gte, lt, lte
@@ -188,7 +186,8 @@ func checkSingleCondition(jsonStr, contextJSON string, condition ConditionOperat
 }
 
 func processNegativeIndex(jsonStr string, path string) string {
-	matches := negativeIndexRegexp.FindAllStringSubmatch(path, -1)
+	re := regexp.MustCompile(`\.(-\d+)`)
+	matches := re.FindAllStringSubmatch(path, -1)
 
 	if len(matches) == 0 {
 		return path

relay/common/relay_info.go

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

relay/compatible_handler.go

@@ -181,7 +181,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 		return newApiErr
 	}
 
-	if usage.(*dto.Usage).CompletionTokenDetails.AudioTokens > 0 || usage.(*dto.Usage).PromptTokensDetails.AudioTokens > 0 {
+	if strings.HasPrefix(info.OriginModelName, "gpt-4o-audio") {
 		service.PostAudioConsumeQuota(c, info, usage.(*dto.Usage), "")
 	} else {
 		postConsumeQuota(c, info, usage.(*dto.Usage), "")

relay/helper/common.go

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

relay/relay_task.go

@@ -83,20 +83,12 @@ func RelayTaskSubmit(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.
 				taskErr = service.TaskErrorWrapperLocal(errors.New("the channel of the origin task is disabled"), "task_channel_disable", http.StatusBadRequest)
 				return
 			}
-			key, _, newAPIError := channel.GetNextEnabledKey()
-			if newAPIError != nil {
-				taskErr = service.TaskErrorWrapper(newAPIError, "channel_no_available_key", newAPIError.StatusCode)
-				return
-			}
-			common.SetContextKey(c, constant.ContextKeyChannelKey, key)
-			common.SetContextKey(c, constant.ContextKeyChannelType, channel.Type)
-			common.SetContextKey(c, constant.ContextKeyChannelBaseUrl, channel.GetBaseURL())
-			common.SetContextKey(c, constant.ContextKeyChannelId, originTask.ChannelId)
+			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
-			info.ChannelType = channel.Type
-			info.ApiKey = key
 			platform = originTask.Platform
 		}
 
@@ -196,7 +188,7 @@ func RelayTaskSubmit(c *gin.Context, info *relaycommon.RelayInfo) (taskErr *dto.
 	// handle response
 	if resp != nil && resp.StatusCode != http.StatusOK {
 		responseBody, _ := io.ReadAll(resp.Body)
-		taskErr = service.TaskErrorWrapper(fmt.Errorf("%s", string(responseBody)), "fail_to_fetch_task", resp.StatusCode)
+		taskErr = service.TaskErrorWrapper(fmt.Errorf(string(responseBody)), "fail_to_fetch_task", resp.StatusCode)
 		return
 	}
 
@@ -385,7 +377,6 @@ func videoFetchByIDRespBodyBuilder(c *gin.Context) (respBody []byte, taskResp *d
 		if channelModel.GetBaseURL() != "" {
 			baseURL = channelModel.GetBaseURL()
 		}
-		proxy := channelModel.GetSetting().Proxy
 		adaptor := GetTaskAdaptor(constant.TaskPlatform(strconv.Itoa(channelModel.Type)))
 		if adaptor == nil {
 			return
@@ -393,7 +384,7 @@ func videoFetchByIDRespBodyBuilder(c *gin.Context) (respBody []byte, taskResp *d
 		resp, err2 := adaptor.FetchTask(baseURL, channelModel.Key, map[string]any{
 			"task_id": originTask.TaskID,
 			"action":  originTask.Action,
-		}, proxy)
+		})
 		if err2 != nil || resp == nil {
 			return
 		}

service/channel_select.go

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

service/convert.go

@@ -201,10 +201,6 @@ func generateStopBlock(index int) *dto.ClaudeResponse {
 }
 
 func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamResponse, info *relaycommon.RelayInfo) []*dto.ClaudeResponse {
-	if info.ClaudeConvertInfo.Done {
-		return nil
-	}
-
 	var claudeResponses []*dto.ClaudeResponse
 	if info.SendResponseCount == 1 {
 		msg := &dto.ClaudeMediaMessage{
@@ -222,117 +218,45 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 			Type:    "message_start",
 			Message: msg,
 		})
+		claudeResponses = append(claudeResponses)
 		//claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
 		//	Type: "ping",
 		//})
 		if openAIResponse.IsToolCall() {
 			info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeTools
-			var toolCall dto.ToolCallResponse
-			if len(openAIResponse.Choices) > 0 && len(openAIResponse.Choices[0].Delta.ToolCalls) > 0 {
-				toolCall = openAIResponse.Choices[0].Delta.ToolCalls[0]
-			} else {
-				first := openAIResponse.GetFirstToolCall()
-				if first != nil {
-					toolCall = *first
-				} else {
-					toolCall = dto.ToolCallResponse{}
-				}
-			}
 			resp := &dto.ClaudeResponse{
 				Type: "content_block_start",
 				ContentBlock: &dto.ClaudeMediaMessage{
-					Id:    toolCall.ID,
+					Id:    openAIResponse.GetFirstToolCall().ID,
 					Type:  "tool_use",
-					Name:  toolCall.Function.Name,
+					Name:  openAIResponse.GetFirstToolCall().Function.Name,
 					Input: map[string]interface{}{},
 				},
 			}
 			resp.SetIndex(0)
 			claudeResponses = append(claudeResponses, resp)
-			// 首块包含工具 delta，则追加 input_json_delta
-			if toolCall.Function.Arguments != "" {
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &info.ClaudeConvertInfo.Index,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type:        "input_json_delta",
-						PartialJson: &toolCall.Function.Arguments,
-					},
-				})
-			}
 		} else {
 
 		}
 		// 判断首个响应是否存在内容（非标准的 OpenAI 响应）
-		if len(openAIResponse.Choices) > 0 {
-			reasoning := openAIResponse.Choices[0].Delta.GetReasoningContent()
-			content := openAIResponse.Choices[0].Delta.GetContentString()
-
-			if reasoning != "" {
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &info.ClaudeConvertInfo.Index,
-					Type:  "content_block_start",
-					ContentBlock: &dto.ClaudeMediaMessage{
-						Type:     "thinking",
-						Thinking: common.GetPointer[string](""),
-					},
-				})
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &info.ClaudeConvertInfo.Index,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type:     "thinking_delta",
-						Thinking: &reasoning,
-					},
-				})
-				info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeThinking
-			} else if content != "" {
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &info.ClaudeConvertInfo.Index,
-					Type:  "content_block_start",
-					ContentBlock: &dto.ClaudeMediaMessage{
-						Type: "text",
-						Text: common.GetPointer[string](""),
-					},
-				})
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &info.ClaudeConvertInfo.Index,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type: "text_delta",
-						Text: common.GetPointer[string](content),
-					},
-				})
-				info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeText
-			}
-		}
-
-		// 如果首块就带 finish_reason，需要立即发送停止块
-		if len(openAIResponse.Choices) > 0 && openAIResponse.Choices[0].FinishReason != nil && *openAIResponse.Choices[0].FinishReason != "" {
-			info.FinishReason = *openAIResponse.Choices[0].FinishReason
-			claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
-			oaiUsage := openAIResponse.Usage
-			if oaiUsage == nil {
-				oaiUsage = info.ClaudeConvertInfo.Usage
-			}
-			if oaiUsage != nil {
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Type: "message_delta",
-					Usage: &dto.ClaudeUsage{
-						InputTokens:              oaiUsage.PromptTokens,
-						OutputTokens:             oaiUsage.CompletionTokens,
-						CacheCreationInputTokens: oaiUsage.PromptTokensDetails.CachedCreationTokens,
-						CacheReadInputTokens:     oaiUsage.PromptTokensDetails.CachedTokens,
-					},
-					Delta: &dto.ClaudeMediaMessage{
-						StopReason: common.GetPointer[string](stopReasonOpenAI2Claude(info.FinishReason)),
-					},
-				})
-			}
+		if len(openAIResponse.Choices) > 0 && len(openAIResponse.Choices[0].Delta.GetContentString()) > 0 {
 			claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-				Type: "message_stop",
+				Index: &info.ClaudeConvertInfo.Index,
+				Type:  "content_block_start",
+				ContentBlock: &dto.ClaudeMediaMessage{
+					Type: "text",
+					Text: common.GetPointer[string](""),
+				},
 			})
-			info.ClaudeConvertInfo.Done = true
+			claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+				Index: &info.ClaudeConvertInfo.Index,
+				Type:  "content_block_delta",
+				Delta: &dto.ClaudeMediaMessage{
+					Type: "text_delta",
+					Text: common.GetPointer[string](openAIResponse.Choices[0].Delta.GetContentString()),
+				},
+			})
+			info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeText
 		}
 		return claudeResponses
 	}
@@ -340,7 +264,7 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 	if len(openAIResponse.Choices) == 0 {
 		// no choices
 		// 可能为非标准的 OpenAI 响应，判断是否已经完成
-		if info.ClaudeConvertInfo.Done {
+		if info.Done {
 			claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
 			oaiUsage := info.ClaudeConvertInfo.Usage
 			if oaiUsage != nil {
@@ -364,110 +288,16 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 		return claudeResponses
 	} else {
 		chosenChoice := openAIResponse.Choices[0]
-		doneChunk := chosenChoice.FinishReason != nil && *chosenChoice.FinishReason != ""
-		if doneChunk {
+		if chosenChoice.FinishReason != nil && *chosenChoice.FinishReason != "" {
+			// should be done
 			info.FinishReason = *chosenChoice.FinishReason
-		}
-
-		var claudeResponse dto.ClaudeResponse
-		var isEmpty bool
-		claudeResponse.Type = "content_block_delta"
-		if len(chosenChoice.Delta.ToolCalls) > 0 {
-			toolCalls := chosenChoice.Delta.ToolCalls
-			if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeTools {
-				claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
-				info.ClaudeConvertInfo.Index++
-			}
-			info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeTools
-
-			for i, toolCall := range toolCalls {
-				blockIndex := info.ClaudeConvertInfo.Index
-				if toolCall.Index != nil {
-					blockIndex = *toolCall.Index
-				} else if len(toolCalls) > 1 {
-					blockIndex = info.ClaudeConvertInfo.Index + i
-				}
-
-				idx := blockIndex
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_start",
-					ContentBlock: &dto.ClaudeMediaMessage{
-						Id:    toolCall.ID,
-						Type:  "tool_use",
-						Name:  toolCall.Function.Name,
-						Input: map[string]interface{}{},
-					},
-				})
-
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type:        "input_json_delta",
-						PartialJson: &toolCall.Function.Arguments,
-					},
-				})
-
-				info.ClaudeConvertInfo.Index = blockIndex
-			}
-		} else {
-			reasoning := chosenChoice.Delta.GetReasoningContent()
-			textContent := chosenChoice.Delta.GetContentString()
-			if reasoning != "" || textContent != "" {
-				if reasoning != "" {
-					if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeThinking {
-						claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-							Index: &info.ClaudeConvertInfo.Index,
-							Type:  "content_block_start",
-							ContentBlock: &dto.ClaudeMediaMessage{
-								Type:     "thinking",
-								Thinking: common.GetPointer[string](""),
-							},
-						})
-					}
-					info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeThinking
-					claudeResponse.Delta = &dto.ClaudeMediaMessage{
-						Type:     "thinking_delta",
-						Thinking: &reasoning,
-					}
-				} else {
-					if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeText {
-						if info.ClaudeConvertInfo.LastMessagesType == relaycommon.LastMessageTypeThinking || info.ClaudeConvertInfo.LastMessagesType == relaycommon.LastMessageTypeTools {
-							claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
-							info.ClaudeConvertInfo.Index++
-						}
-						claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-							Index: &info.ClaudeConvertInfo.Index,
-							Type:  "content_block_start",
-							ContentBlock: &dto.ClaudeMediaMessage{
-								Type: "text",
-								Text: common.GetPointer[string](""),
-							},
-						})
-					}
-					info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeText
-					claudeResponse.Delta = &dto.ClaudeMediaMessage{
-						Type: "text_delta",
-						Text: common.GetPointer[string](textContent),
-					}
-				}
-			} else {
-				isEmpty = true
+			if !info.Done {
+				return claudeResponses
 			}
 		}
-
-		claudeResponse.Index = &info.ClaudeConvertInfo.Index
-		if !isEmpty && claudeResponse.Delta != nil {
-			claudeResponses = append(claudeResponses, &claudeResponse)
-		}
-
-		if doneChunk || info.ClaudeConvertInfo.Done {
+		if info.Done {
 			claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
-			oaiUsage := openAIResponse.Usage
-			if oaiUsage == nil {
-				oaiUsage = info.ClaudeConvertInfo.Usage
-			}
+			oaiUsage := info.ClaudeConvertInfo.Usage
 			if oaiUsage != nil {
 				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
 					Type: "message_delta",
@@ -485,8 +315,83 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 			claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
 				Type: "message_stop",
 			})
-			info.ClaudeConvertInfo.Done = true
-			return claudeResponses
+		} else {
+			var claudeResponse dto.ClaudeResponse
+			var isEmpty bool
+			claudeResponse.Type = "content_block_delta"
+			if len(chosenChoice.Delta.ToolCalls) > 0 {
+				if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeTools {
+					claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
+					info.ClaudeConvertInfo.Index++
+					claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+						Index: &info.ClaudeConvertInfo.Index,
+						Type:  "content_block_start",
+						ContentBlock: &dto.ClaudeMediaMessage{
+							Id:    openAIResponse.GetFirstToolCall().ID,
+							Type:  "tool_use",
+							Name:  openAIResponse.GetFirstToolCall().Function.Name,
+							Input: map[string]interface{}{},
+						},
+					})
+				}
+				info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeTools
+				// tools delta
+				claudeResponse.Delta = &dto.ClaudeMediaMessage{
+					Type:        "input_json_delta",
+					PartialJson: &chosenChoice.Delta.ToolCalls[0].Function.Arguments,
+				}
+			} else {
+				reasoning := chosenChoice.Delta.GetReasoningContent()
+				textContent := chosenChoice.Delta.GetContentString()
+				if reasoning != "" || textContent != "" {
+					if reasoning != "" {
+						if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeThinking {
+							//info.ClaudeConvertInfo.Index++
+							claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+								Index: &info.ClaudeConvertInfo.Index,
+								Type:  "content_block_start",
+								ContentBlock: &dto.ClaudeMediaMessage{
+									Type:     "thinking",
+									Thinking: common.GetPointer[string](""),
+								},
+							})
+						}
+						info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeThinking
+						// text delta
+						claudeResponse.Delta = &dto.ClaudeMediaMessage{
+							Type:     "thinking_delta",
+							Thinking: &reasoning,
+						}
+					} else {
+						if info.ClaudeConvertInfo.LastMessagesType != relaycommon.LastMessageTypeText {
+							if info.LastMessagesType == relaycommon.LastMessageTypeThinking || info.LastMessagesType == relaycommon.LastMessageTypeTools {
+								claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
+								info.ClaudeConvertInfo.Index++
+							}
+							claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+								Index: &info.ClaudeConvertInfo.Index,
+								Type:  "content_block_start",
+								ContentBlock: &dto.ClaudeMediaMessage{
+									Type: "text",
+									Text: common.GetPointer[string](""),
+								},
+							})
+						}
+						info.ClaudeConvertInfo.LastMessagesType = relaycommon.LastMessageTypeText
+						// text delta
+						claudeResponse.Delta = &dto.ClaudeMediaMessage{
+							Type: "text_delta",
+							Text: common.GetPointer[string](textContent),
+						}
+					}
+				} else {
+					isEmpty = true
+				}
+			}
+			claudeResponse.Index = &info.ClaudeConvertInfo.Index
+			if !isEmpty {
+				claudeResponses = append(claudeResponses, &claudeResponse)
+			}
 		}
 	}
 

service/error.go

@@ -96,21 +96,19 @@ func RelayErrorHandler(ctx context.Context, resp *http.Response, showBodyWhenFai
 		if showBodyWhenFail {
 			newApiErr.Err = fmt.Errorf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody))
 		} else {
-			logger.LogError(ctx, fmt.Sprintf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody)))
+			if common.DebugEnabled {
+				logger.LogInfo(ctx, fmt.Sprintf("bad response status code %d, body: %s", resp.StatusCode, string(responseBody)))
+			}
 			newApiErr.Err = fmt.Errorf("bad response status code %d", resp.StatusCode)
 		}
 		return
 	}
-
-	if common.GetJsonType(errResponse.Error) == "object" {
+	if errResponse.Error.Message != "" {
 		// General format error (OpenAI, Anthropic, Gemini, etc.)
-		oaiError := errResponse.TryToOpenAIError()
-		if oaiError != nil {
-			newApiErr = types.WithOpenAIError(*oaiError, resp.StatusCode)
-			return
-		}
+		newApiErr = types.WithOpenAIError(errResponse.Error, resp.StatusCode)
+	} else {
+		newApiErr = types.NewOpenAIError(errors.New(errResponse.ToMessage()), types.ErrorCodeBadResponseStatusCode, resp.StatusCode)
 	}
-	newApiErr = types.NewOpenAIError(errors.New(errResponse.ToMessage()), types.ErrorCodeBadResponseStatusCode, resp.StatusCode)
 	return
 }
 

service/http_client.go

@@ -35,9 +35,9 @@ func checkRedirect(req *http.Request, via []*http.Request) error {
 
 func InitHttpClient() {
 	transport := &http.Transport{
-		MaxIdleConns:        common.RelayMaxIdleConns,
-		MaxIdleConnsPerHost: common.RelayMaxIdleConnsPerHost,
-		ForceAttemptHTTP2:   true,
+		MaxIdleConns:          common.RelayMaxIdleConns,
+		MaxIdleConnsPerHost:   common.RelayMaxIdleConnsPerHost,
+		ForceAttemptHTTP2:     true,
 	}
 
 	if common.RelayTimeout == 0 {
@@ -58,14 +58,6 @@ func GetHttpClient() *http.Client {
 	return httpClient
 }
 
-// GetHttpClientWithProxy returns the default client or a proxy-enabled one when proxyURL is provided.
-func GetHttpClientWithProxy(proxyURL string) (*http.Client, error) {
-	if proxyURL == "" {
-		return GetHttpClient(), nil
-	}
-	return NewProxyHttpClient(proxyURL)
-}
-
 // ResetProxyClientCache 清空代理客户端缓存，确保下次使用时重新初始化
 func ResetProxyClientCache() {
 	proxyClientLock.Lock()
@@ -100,10 +92,10 @@ func NewProxyHttpClient(proxyURL string) (*http.Client, error) {
 	case "http", "https":
 		client := &http.Client{
 			Transport: &http.Transport{
-				MaxIdleConns:        common.RelayMaxIdleConns,
-				MaxIdleConnsPerHost: common.RelayMaxIdleConnsPerHost,
-				ForceAttemptHTTP2:   true,
-				Proxy:               http.ProxyURL(parsedURL),
+				MaxIdleConns:          common.RelayMaxIdleConns,
+				MaxIdleConnsPerHost:   common.RelayMaxIdleConnsPerHost,
+				ForceAttemptHTTP2:     true,
+				Proxy: http.ProxyURL(parsedURL),
 			},
 			CheckRedirect: checkRedirect,
 		}
@@ -135,9 +127,9 @@ func NewProxyHttpClient(proxyURL string) (*http.Client, error) {
 
 		client := &http.Client{
 			Transport: &http.Transport{
-				MaxIdleConns:        common.RelayMaxIdleConns,
-				MaxIdleConnsPerHost: common.RelayMaxIdleConnsPerHost,
-				ForceAttemptHTTP2:   true,
+				MaxIdleConns:          common.RelayMaxIdleConns,
+				MaxIdleConnsPerHost:   common.RelayMaxIdleConnsPerHost,
+				ForceAttemptHTTP2:     true,
 				DialContext: func(ctx context.Context, network, addr string) (net.Conn, error) {
 					return dialer.Dial(network, addr)
 				},

service/quota.go

@@ -108,7 +108,7 @@ func PreWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usag
 	groupRatio := ratio_setting.GetGroupRatio(relayInfo.UsingGroup)
 	modelRatio, _, _ := ratio_setting.GetModelRatio(modelName)
 
-	autoGroup, exists := common.GetContextKey(ctx, constant.ContextKeyAutoGroup)
+	autoGroup, exists := ctx.Get("auto_group")
 	if exists {
 		groupRatio = ratio_setting.GetGroupRatio(autoGroup.(string))
 		log.Printf("final group ratio: %f", groupRatio)

service/token_counter.go

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

setting/model_setting/gemini.go

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

setting/ratio_setting/cache_ratio.go

@@ -43,7 +43,6 @@ var defaultCacheRatio = map[string]float64{
 	"claude-3-opus-20240229":              0.1,
 	"claude-3-haiku-20240307":             0.1,
 	"claude-3-5-haiku-20241022":           0.1,
-	"claude-haiku-4-5-20251001":           0.1,
 	"claude-3-5-sonnet-20240620":          0.1,
 	"claude-3-5-sonnet-20241022":          0.1,
 	"claude-3-7-sonnet-20250219":          0.1,
@@ -65,7 +64,6 @@ var defaultCreateCacheRatio = map[string]float64{
 	"claude-3-opus-20240229":              1.25,
 	"claude-3-haiku-20240307":             1.25,
 	"claude-3-5-haiku-20241022":           1.25,
-	"claude-haiku-4-5-20251001":           1.25,
 	"claude-3-5-sonnet-20240620":          1.25,
 	"claude-3-5-sonnet-20241022":          1.25,
 	"claude-3-7-sonnet-20250219":          1.25,

setting/ratio_setting/model_ratio.go

@@ -137,7 +137,6 @@ var defaultModelRatio = map[string]float64{
 	"claude-2.1":                                4,     // $8 / 1M tokens
 	"claude-3-haiku-20240307":                   0.125, // $0.25 / 1M tokens
 	"claude-3-5-haiku-20241022":                 0.5,   // $1 / 1M tokens
-	"claude-haiku-4-5-20251001":                 0.5,   // $1 / 1M tokens
 	"claude-3-sonnet-20240229":                  1.5,   // $3 / 1M tokens
 	"claude-3-5-sonnet-20240620":                1.5,
 	"claude-3-5-sonnet-20241022":                1.5,
@@ -297,7 +296,6 @@ var defaultModelPrice = map[string]float64{
 	"mj_upload":                      0.05,
 	"sora-2":                         0.3,
 	"sora-2-pro":                     0.5,
-	"gpt-4o-mini-tts":                0.3,
 }
 
 var defaultAudioRatio = map[string]float64{
@@ -305,13 +303,11 @@ var defaultAudioRatio = map[string]float64{
 	"gpt-4o-mini-audio-preview":    66.67,
 	"gpt-4o-realtime-preview":      8,
 	"gpt-4o-mini-realtime-preview": 16.67,
-	"gpt-4o-mini-tts":              25,
 }
 
 var defaultAudioCompletionRatio = map[string]float64{
 	"gpt-4o-realtime":      2,
 	"gpt-4o-mini-realtime": 2,
-	"gpt-4o-mini-tts":      1,
 }
 
 var (
@@ -539,10 +535,7 @@ func getHardcodedCompletionModelRatio(name string) (float64, bool) {
 			if name == "gpt-4o-2024-05-13" {
 				return 3, true
 			}
-			if strings.HasPrefix(name, "gpt-4o-mini-tts") {
-				return 20, false
-			}
-			return 4, false
+			return 4, true
 		}
 		// gpt-5 匹配
 		if strings.HasPrefix(name, "gpt-5") {
@@ -567,7 +560,7 @@ func getHardcodedCompletionModelRatio(name string) (float64, bool) {
 
 	if strings.Contains(name, "claude-3") {
 		return 5, true
-	} else if strings.Contains(name, "claude-sonnet-4") || strings.Contains(name, "claude-opus-4") || strings.Contains(name, "claude-haiku-4") {
+	} else if strings.Contains(name, "claude-sonnet-4") || strings.Contains(name, "claude-opus-4") {
 		return 5, true
 	} else if strings.Contains(name, "claude-instant-1") || strings.Contains(name, "claude-2") {
 		return 3, true

setting/system_setting/discord.go

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

types/error.go

@@ -94,14 +94,6 @@ type NewAPIError struct {
 	StatusCode     int
 }
 
-// Unwrap enables errors.Is / errors.As to work with NewAPIError by exposing the underlying error.
-func (e *NewAPIError) Unwrap() error {
-	if e == nil {
-		return nil
-	}
-	return e.Err
-}
-
 func (e *NewAPIError) GetErrorCode() ErrorCode {
 	if e == nil {
 		return ""

web/bun.lock

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

web/package.json

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

web/src/components/auth/LoginForm.jsx

@@ -294,7 +294,7 @@ const LoginForm = () => {
       setGithubButtonDisabled(true);
     }, 20000);
     try {
-      onGitHubOAuthClicked(status.github_client_id, { shouldLogout: true });
+      onGitHubOAuthClicked(status.github_client_id);
     } finally {
       // 由于重定向，这里不会执行到，但为了完整性添加
       setTimeout(() => setGithubLoading(false), 3000);
@@ -309,7 +309,7 @@ const LoginForm = () => {
     }
     setDiscordLoading(true);
     try {
-      onDiscordOAuthClicked(status.discord_client_id, { shouldLogout: true });
+      onDiscordOAuthClicked(status.discord_client_id);
     } finally {
       // 由于重定向，这里不会执行到，但为了完整性添加
       setTimeout(() => setDiscordLoading(false), 3000);
@@ -324,12 +324,7 @@ const LoginForm = () => {
     }
     setOidcLoading(true);
     try {
-      onOIDCClicked(
-        status.oidc_authorization_endpoint,
-        status.oidc_client_id,
-        false,
-        { shouldLogout: true },
-      );
+      onOIDCClicked(status.oidc_authorization_endpoint, status.oidc_client_id);
     } finally {
       // 由于重定向，这里不会执行到，但为了完整性添加
       setTimeout(() => setOidcLoading(false), 3000);
@@ -344,7 +339,7 @@ const LoginForm = () => {
     }
     setLinuxdoLoading(true);
     try {
-      onLinuxDOOAuthClicked(status.linuxdo_client_id, { shouldLogout: true });
+      onLinuxDOOAuthClicked(status.linuxdo_client_id);
     } finally {
       // 由于重定向，这里不会执行到，但为了完整性添加
       setTimeout(() => setLinuxdoLoading(false), 3000);

web/src/components/auth/RegisterForm.jsx

@@ -261,7 +261,7 @@ const RegisterForm = () => {
       setGithubButtonDisabled(true);
     }, 20000);
     try {
-      onGitHubOAuthClicked(status.github_client_id, { shouldLogout: true });
+      onGitHubOAuthClicked(status.github_client_id);
     } finally {
       setTimeout(() => setGithubLoading(false), 3000);
     }
@@ -270,7 +270,7 @@ const RegisterForm = () => {
   const handleDiscordClick = () => {
     setDiscordLoading(true);
     try {
-      onDiscordOAuthClicked(status.discord_client_id, { shouldLogout: true });
+      onDiscordOAuthClicked(status.discord_client_id);
     } finally {
       setTimeout(() => setDiscordLoading(false), 3000);
     }
@@ -279,12 +279,7 @@ const RegisterForm = () => {
   const handleOIDCClick = () => {
     setOidcLoading(true);
     try {
-      onOIDCClicked(
-        status.oidc_authorization_endpoint,
-        status.oidc_client_id,
-        false,
-        { shouldLogout: true },
-      );
+      onOIDCClicked(status.oidc_authorization_endpoint, status.oidc_client_id);
     } finally {
       setTimeout(() => setOidcLoading(false), 3000);
     }
@@ -293,7 +288,7 @@ const RegisterForm = () => {
   const handleLinuxDOClick = () => {
     setLinuxdoLoading(true);
     try {
-      onLinuxDOOAuthClicked(status.linuxdo_client_id, { shouldLogout: true });
+      onLinuxDOOAuthClicked(status.linuxdo_client_id);
     } finally {
       setTimeout(() => setLinuxdoLoading(false), 3000);
     }

web/src/components/layout/SiderBar.jsx

@@ -377,6 +377,7 @@ const SiderBar = ({ onNavigate = () => {} }) => {
       className='sidebar-container'
       style={{
         width: 'var(--sidebar-current-width)',
+        background: 'var(--semi-color-bg-0)',
       }}
     >
       <SkeletonWrapper

web/src/components/settings/ModelSetting.jsx

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

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

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

web/src/components/table/task-logs/TaskLogsColumnDefs.jsx

@@ -129,7 +129,7 @@ const renderType = (type, t) => {
     case TASK_ACTION_REMIX_GENERATE:
       return (
         <Tag color='blue' shape='circle' prefixIcon={<Sparkles size={14} />}>
-          {t('视频Remix')}
+          {t('remix生视频')}
         </Tag>
       );
     default:

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

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

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

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

web/src/helpers/api.js

@@ -231,22 +231,8 @@ export async function getOAuthState() {
   }
 }
 
-async function prepareOAuthState(options = {}) {
-  const { shouldLogout = false } = options;
-  if (shouldLogout) {
-    try {
-      await API.get('/api/user/logout', { skipErrorHandler: true });
-    } catch (err) {
-
-    }
-    localStorage.removeItem('user');
-    updateAPI();
-  }
-  return await getOAuthState();
-}
-
-export async function onDiscordOAuthClicked(client_id, options = {}) {
-  const state = await prepareOAuthState(options);
+export async function onDiscordOAuthClicked(client_id) {
+  const state = await getOAuthState();
   if (!state) return;
   const redirect_uri = `${window.location.origin}/oauth/discord`;
   const response_type = 'code';
@@ -256,13 +242,8 @@ export async function onDiscordOAuthClicked(client_id, options = {}) {
   );
 }
 
-export async function onOIDCClicked(
-  auth_url,
-  client_id,
-  openInNewTab = false,
-  options = {},
-) {
-  const state = await prepareOAuthState(options);
+export async function onOIDCClicked(auth_url, client_id, openInNewTab = false) {
+  const state = await getOAuthState();
   if (!state) return;
   const url = new URL(auth_url);
   url.searchParams.set('client_id', client_id);
@@ -277,19 +258,16 @@ export async function onOIDCClicked(
   }
 }
 
-export async function onGitHubOAuthClicked(github_client_id, options = {}) {
-  const state = await prepareOAuthState(options);
+export async function onGitHubOAuthClicked(github_client_id) {
+  const state = await getOAuthState();
   if (!state) return;
   window.open(
     `https://github.com/login/oauth/authorize?client_id=${github_client_id}&state=${state}&scope=user:email`,
   );
 }
 
-export async function onLinuxDOOAuthClicked(
-  linuxdo_client_id,
-  options = { shouldLogout: false },
-) {
-  const state = await prepareOAuthState(options);
+export async function onLinuxDOOAuthClicked(linuxdo_client_id) {
+  const state = await getOAuthState();
   if (!state) return;
   window.open(
     `https://connect.linux.do/oauth2/authorize?response_type=code&client_id=${linuxdo_client_id}&state=${state}`,

web/src/helpers/render.jsx

@@ -1086,12 +1086,9 @@ function renderPriceSimpleCore({
   );
   const finalGroupRatio = effectiveGroupRatio;
 
-  const { symbol, rate } = getCurrencyConfig();
   if (modelPrice !== -1) {
-    const displayPrice = (modelPrice * rate).toFixed(6);
-    return i18next.t('价格：{{symbol}}{{price}} * {{ratioType}}：{{ratio}}', {
-      symbol: symbol,
-      price: displayPrice,
+    return i18next.t('价格：${{price}} * {{ratioType}}：{{ratio}}', {
+      price: modelPrice,
       ratioType: ratioLabel,
       ratio: finalGroupRatio,
     });

web/src/i18n/locales/en.json

@@ -97,7 +97,7 @@
     "Homepage URL 填": "Fill in the Homepage URL",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "IP whitelist (supports CIDR expressions)",
+    "IP白名单": "IP whitelist",
     "IP限制": "IP restrictions",
     "IP黑名单": "IP blacklist",
     "JSON": "JSON",
@@ -153,7 +153,6 @@
     "URL链接": "URL Link",
     "USD (美元)": "USD (US Dollar)",
     "User Info Endpoint": "User Info Endpoint",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI does not support the functionResponse.id field. When enabled, this field will be automatically removed",
     "Webhook 密钥": "Webhook Secret",
     "Webhook 签名密钥": "Webhook Signature Key",
     "Webhook地址": "Webhook URL",
@@ -549,7 +548,6 @@
     "参数值": "Parameter value",
     "参数覆盖": "Parameters override",
     "参照生视频": "Reference video generation",
-    "视频Remix": "Video remix",
     "友情链接": "Friendly links",
     "发布日期": "Publish Date",
     "发布时间": "Publish Time",
@@ -1511,7 +1509,6 @@
     "私有IP访问详细说明": "⚠️ Security Warning: Enabling this allows access to internal network resources (localhost, private networks). Only enable if you need to access internal services and understand the security implications.",
     "私有部署地址": "Private Deployment Address",
     "秒": "Second",
-    "移除 functionResponse.id 字段": "Remove functionResponse.id Field",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Removal of One API copyright mark must first be authorized. Project maintenance requires a lot of effort. If this project is meaningful to you, please actively support it.",
     "窗口处理": "window handling",
     "窗口等待": "window wait",
@@ -1754,7 +1751,7 @@
     "请先阅读并同意用户协议和隐私政策": "Please read and agree to the user agreement and privacy policy first",
     "请再次输入新密码": "Please enter the new password again",
     "请前往个人设置 → 安全设置进行配置。": "Please go to Personal Settings → Security Settings to configure.",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Do not over-trust this feature, IP can be spoofed, please use it in conjunction with gateways such as nginx and CDN",
+    "请勿过度信任此功能，IP可能被伪造": "Do not over-trust this feature, IP can be spoofed",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Please edit Group ratios in system settings to add new groups:",
     "请填写完整的产品信息": "Please fill in complete product information",
     "请填写完整的管理员账号信息": "Please fill in the complete administrator account information",
@@ -2179,9 +2176,6 @@
     "默认区域，如: us-central1": "Default region, e.g.: us-central1",
     "默认折叠侧边栏": "Default collapse sidebar",
     "默认测试模型": "Default Test Model",
-    "默认补全倍率": "Default completion ratio",
-    "跨分组重试": "Cross-group retry",
-    "跨分组": "Cross-group",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "After enabling, when the current group channel fails, it will try the next group's channel in order"
+    "默认补全倍率": "Default completion ratio"
   }
 }

web/src/i18n/locales/fr.json

@@ -99,7 +99,7 @@
     "Homepage URL 填": "Remplir l'URL de la page d'accueil",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "Liste blanche d'adresses IP (prise en charge des expressions CIDR)",
+    "IP白名单": "Liste blanche d'adresses IP",
     "IP限制": "Restrictions d'IP",
     "IP黑名单": "Liste noire d'adresses IP",
     "JSON": "JSON",
@@ -154,7 +154,6 @@
     "URL链接": "Lien URL",
     "USD (美元)": "USD (Dollar US)",
     "User Info Endpoint": "Point de terminaison des informations utilisateur",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI ne prend pas en charge le champ functionResponse.id. Lorsqu'il est activé, ce champ sera automatiquement supprimé",
     "Webhook 密钥": "Clé Webhook",
     "Webhook 签名密钥": "Clé de signature Webhook",
     "Webhook地址": "URL du Webhook",
@@ -552,7 +551,6 @@
     "参数值": "Valeur du paramètre",
     "参数覆盖": "Remplacement des paramètres",
     "参照生视频": "Générer une vidéo par référence",
-    "视频Remix": "Remix vidéo",
     "友情链接": "Liens amicaux",
     "发布日期": "Date de publication",
     "发布时间": "Heure de publication",
@@ -1521,7 +1519,6 @@
     "私有IP访问详细说明": "⚠️ Avertissement de sécurité : l'activation de cette option autorise l'accès aux ressources du réseau interne (localhost, réseaux privés). N'activez cette option que si vous devez accéder à des services internes et que vous comprenez les implications en matière de sécurité.",
     "私有部署地址": "Adresse de déploiement privée",
     "秒": "Seconde",
-    "移除 functionResponse.id 字段": "Supprimer le champ functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "La suppression de la marque de copyright de One API doit d'abord être autorisée. La maintenance du projet demande beaucoup d'efforts. Si ce projet a du sens pour vous, veuillez le soutenir activement.",
     "窗口处理": "gestion des fenêtres",
     "窗口等待": "attente de la fenêtre",
@@ -1764,7 +1761,7 @@
     "请先阅读并同意用户协议和隐私政策": "Veuillez d'abord lire et accepter l'accord utilisateur et la politique de confidentialité",
     "请再次输入新密码": "Veuillez saisir à nouveau le nouveau mot de passe",
     "请前往个人设置 → 安全设置进行配置。": "Veuillez aller dans Paramètres personnels → Paramètres de sécurité pour configurer.",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Ne faites pas trop confiance à cette fonctionnalité, l'IP peut être usurpée, veuillez l'utiliser en conjonction avec des passerelles telles que nginx et cdn",
+    "请勿过度信任此功能，IP可能被伪造": "Ne faites pas trop confiance à cette fonctionnalité, l'IP peut être usurpée",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Veuillez modifier les ratios de groupe dans les paramètres système pour ajouter de nouveaux groupes :",
     "请填写完整的产品信息": "Veuillez renseigner l'ensemble des informations produit",
     "请填写完整的管理员账号信息": "Veuillez remplir les informations complètes du compte administrateur",
@@ -2228,9 +2225,6 @@
     "默认助手消息": "Bonjour ! Comment puis-je vous aider aujourd'hui ?",
     "可选，用于复现结果": "Optionnel, pour des résultats reproductibles",
     "随机种子 (留空为随机)": "Graine aléatoire (laisser vide pour aléatoire)",
-    "默认补全倍率": "Taux de complétion par défaut",
-    "跨分组重试": "Nouvelle tentative inter-groupes",
-    "跨分组": "Inter-groupes",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "Après activation, lorsque le canal du groupe actuel échoue, il essaiera le canal du groupe suivant dans l'ordre"
+    "默认补全倍率": "Taux de complétion par défaut"
   }
 }

web/src/i18n/locales/ja.json

@@ -82,7 +82,7 @@
     "Homepage URL 填": "ホームページURLを入力してください",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "IPホワイトリスト（CIDR表記に対応）",
+    "IP白名单": "IPホワイトリスト",
     "IP限制": "IP制限",
     "IP黑名单": "IPブラックリスト",
     "JSON": "JSON",
@@ -136,7 +136,6 @@
     "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Uptime Kumaの監視分類管理：サービスステータス表示用に、複数の監視分類を設定できます（最大20個）",
     "URL链接": "URL",
     "User Info Endpoint": "User Info Endpoint",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AIはfunctionResponse.idフィールドをサポートしていません。有効にすると、このフィールドは自動的に削除されます",
     "Webhook 签名密钥": "Webhook署名シークレット",
     "Webhook地址": "Webhook URL",
     "Webhook地址必须以https://开头": "Webhook URLは、https://で始まることが必須です",
@@ -511,7 +510,6 @@
     "参数值": "パラメータ値",
     "参数覆盖": "パラメータの上書き",
     "参照生视频": "参照動画生成",
-    "视频Remix": "動画リミックス",
     "友情链接": "関連リンク",
     "发布日期": "公開日",
     "发布时间": "公開日時",
@@ -1441,7 +1439,6 @@
     "私有IP访问详细说明": "プライベートIPアクセスの詳細説明",
     "私有部署地址": "プライベートデプロイ先URL",
     "秒": "秒",
-    "移除 functionResponse.id 字段": "functionResponse.idフィールドを削除",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "One APIの著作権表示を削除するには、事前の許可が必要です。プロジェクトの維持には多大な労力がかかります。もしこのプロジェクトがあなたにとって有意義でしたら、積極的なご支援をお願いいたします",
     "窗口处理": "ウィンドウ処理",
     "窗口等待": "ウィンドウ待機中",
@@ -1671,7 +1668,7 @@
     "请先阅读并同意用户协议和隐私政策": "まずユーザー利用規約とプライバシーポリシーをご確認の上、同意してください",
     "请再次输入新密码": "新しいパスワードを再入力してください",
     "请前往个人设置 → 安全设置进行配置。": "アカウント設定 → セキュリティ設定 にて設定してください。",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "IPは偽装される可能性があるため、この機能を過信しないでください。nginxやCDNなどのゲートウェイと組み合わせて使用してください。",
+    "请勿过度信任此功能，IP可能被伪造": "IPは偽装される可能性があるため、この機能を過信しないでください",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "新規グループを追加するには、システム設定ページでグループ倍率を編集してください：",
     "请填写完整的管理员账号信息": "管理者アカウント情報をすべて入力してください",
     "请填写密钥": "APIキーを入力してください",
@@ -2127,9 +2124,6 @@
     "默认用户消息": "こんにちは",
     "默认助手消息": "こんにちは！何かお手伝いできることはありますか？",
     "可选，用于复现结果": "オプション、結果の再現用",
-    "随机种子 (留空为随机)": "ランダムシード（空欄でランダム）",
-    "跨分组重试": "グループ間リトライ",
-    "跨分组": "グループ間",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "有効にすると、現在のグループチャネルが失敗した場合、次のグループのチャネルを順番に試行します"
+    "随机种子 (留空为随机)": "ランダムシード（空欄でランダム）"
   }
 }

web/src/i18n/locales/ru.json

@@ -101,7 +101,7 @@
     "Homepage URL 填": "URL домашней страницы:",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "Белый список IP (поддерживает выражения CIDR)",
+    "IP白名单": "Белый список IP",
     "IP限制": "Ограничения IP",
     "IP黑名单": "Черный список IP",
     "JSON": "JSON",
@@ -156,7 +156,6 @@
     "URL链接": "URL ссылка",
     "USD (美元)": "USD (доллар США)",
     "User Info Endpoint": "Конечная точка информации о пользователе",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI не поддерживает поле functionResponse.id. При включении это поле будет автоматически удалено",
     "Webhook 密钥": "Секрет вебхука",
     "Webhook 签名密钥": "Ключ подписи Webhook",
     "Webhook地址": "Адрес Webhook",
@@ -556,7 +555,6 @@
     "参数值": "Значение параметра",
     "参数覆盖": "Переопределение параметров",
     "参照生视频": "Ссылка на генерацию видео",
-    "视频Remix": "Видео ремикс",
     "友情链接": "Дружественные ссылки",
     "发布日期": "Дата публикации",
     "发布时间": "Время публикации",
@@ -1532,7 +1530,6 @@
     "私有IP访问详细说明": "⚠️ Предупреждение безопасности: включение этой опции позволит доступ к ресурсам внутренней сети (localhost, частные сети). Включайте только при необходимости доступа к внутренним службам и понимании рисков безопасности.",
     "私有部署地址": "Адрес частного развёртывания",
     "秒": "секунда",
-    "移除 functionResponse.id 字段": "Удалить поле functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Удаление авторских знаков One API требует предварительного разрешения, поддержка проекта требует больших усилий, если этот проект важен для вас, пожалуйста, поддержите его",
     "窗口处理": "Обработка окна",
     "窗口等待": "Ожидание окна",
@@ -1775,7 +1772,7 @@
     "请先阅读并同意用户协议和隐私政策": "Пожалуйста, сначала прочтите и согласитесь с пользовательским соглашением и политикой конфиденциальности",
     "请再次输入新密码": "Пожалуйста, введите новый пароль ещё раз",
     "请前往个人设置 → 安全设置进行配置。": "Пожалуйста, перейдите в Личные настройки → Настройки безопасности для конфигурации.",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Не доверяйте этой функции чрезмерно, IP может быть подделан, используйте её вместе с nginx и CDN и другими шлюзами",
+    "请勿过度信任此功能，IP可能被伪造": "Не доверяйте этой функции чрезмерно, IP может быть подделан",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Пожалуйста, отредактируйте коэффициенты групп на странице системных настроек для добавления новой группы:",
     "请填写完整的产品信息": "Пожалуйста, заполните всю информацию о продукте",
     "请填写完整的管理员账号信息": "Пожалуйста, заполните полную информацию об учётной записи администратора",
@@ -2238,9 +2235,6 @@
     "默认用户消息": "Здравствуйте",
     "默认助手消息": "Здравствуйте! Чем я могу вам помочь?",
     "可选，用于复现结果": "Необязательно, для воспроизводимых результатов",
-    "随机种子 (留空为随机)": "Случайное зерно (оставьте пустым для случайного)",
-    "跨分组重试": "Повторная попытка между группами",
-    "跨分组": "Межгрупповой",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "После включения, когда канал текущей группы не работает, он будет пытаться использовать канал следующей группы по порядку"
+    "随机种子 (留空为随机)": "Случайное зерно (оставьте пустым для случайного)"
   }
 }

web/src/i18n/locales/vi.json

@@ -82,7 +82,7 @@
     "Homepage URL 填": "Điền URL trang chủ",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "Danh sách trắng IP (hỗ trợ biểu thức CIDR)",
+    "IP白名单": "Danh sách trắng IP",
     "IP限制": "Hạn chế IP",
     "IP黑名单": "Danh sách đen IP",
     "JSON": "JSON",
@@ -136,7 +136,6 @@
     "Uptime Kuma监控分类管理，可以配置多个监控分类用于服务状态展示（最多20个）": "Quản lý danh mục giám sát Uptime Kuma, bạn có thể cấu hình nhiều danh mục giám sát để hiển thị trạng thái dịch vụ (tối đa 20)",
     "URL链接": "Liên kết URL",
     "User Info Endpoint": "User Info Endpoint",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI không hỗ trợ trường functionResponse.id. Khi bật, trường này sẽ tự động bị xóa",
     "Webhook 签名密钥": "Khóa chữ ký Webhook",
     "Webhook地址": "URL Webhook",
     "Webhook地址必须以https://开头": "URL Webhook phải bắt đầu bằng https://",
@@ -511,7 +510,6 @@
     "参数值": "Giá trị tham số",
     "参数覆盖": "Ghi đè tham số",
     "参照生视频": "Tạo video tham chiếu",
-    "视频Remix": "Remix video",
     "友情链接": "Liên kết thân thiện",
     "发布日期": "Ngày xuất bản",
     "发布时间": "Thời gian xuất bản",
@@ -1988,7 +1986,7 @@
     "请先阅读并同意用户协议和隐私政策": "Vui lòng đọc và đồng ý với thỏa thuận người dùng và chính sách bảo mật trước",
     "请再次输入新密码": "Vui lòng nhập lại mật khẩu mới",
     "请前往个人设置 → 安全设置进行配置。": "Vui lòng truy cập Cài đặt cá nhân → Cài đặt bảo mật để cấu hình.",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "Đừng quá tin tưởng tính năng này, IP có thể bị giả mạo, vui lòng sử dụng cùng với nginx và các cổng khác như cdn",
+    "请勿过度信任此功能，IP可能被伪造": "Đừng quá tin tưởng tính năng này, IP có thể bị giả mạo",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "Vui lòng chỉnh sửa tỷ lệ nhóm trên trang cài đặt hệ thống để thêm nhóm mới:",
     "请填写完整的管理员账号信息": "Vui lòng điền đầy đủ thông tin tài khoản quản trị viên",
     "请填写密钥": "Vui lòng điền khóa",
@@ -2649,7 +2647,6 @@
     "私有IP访问详细说明": "⚠️ Cảnh báo bảo mật: Bật tính năng này cho phép truy cập vào tài nguyên mạng nội bộ (localhost, mạng riêng). Chỉ bật nếu bạn cần truy cập các dịch vụ nội bộ và hiểu rõ các rủi ro bảo mật.",
     "私有部署地址": "Địa chỉ triển khai riêng",
     "秒": "Giây",
-    "移除 functionResponse.id 字段": "Xóa trường functionResponse.id",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "Việc xóa dấu bản quyền One API trước tiên phải được ủy quyền. Việc bảo trì dự án đòi hỏi rất nhiều nỗ lực. Nếu dự án này có ý nghĩa với bạn, vui lòng chủ động ủng hộ dự án này.",
     "窗口处理": "xử lý cửa sổ",
     "窗口等待": "chờ cửa sổ",
@@ -2738,9 +2735,6 @@
     "默认用户消息": "Xin chào",
     "默认助手消息": "Xin chào! Tôi có thể giúp gì cho bạn?",
     "可选，用于复现结果": "Tùy chọn, để tái tạo kết quả",
-    "随机种子 (留空为随机)": "Hạt giống ngẫu nhiên (để trống cho ngẫu nhiên)",
-    "跨分组重试": "Thử lại giữa các nhóm",
-    "跨分组": "Giữa các nhóm",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "Sau khi bật, khi kênh nhóm hiện tại thất bại, nó sẽ thử kênh của nhóm tiếp theo theo thứ tự"
+    "随机种子 (留空为随机)": "Hạt giống ngẫu nhiên (để trống cho ngẫu nhiên)"
   }
 }

web/src/i18n/locales/zh.json

@@ -95,7 +95,7 @@
     "Homepage URL 填": "Homepage URL 填",
     "ID": "ID",
     "IP": "IP",
-    "IP白名单（支持CIDR表达式）": "IP白名单（支持CIDR表达式）",
+    "IP白名单": "IP白名单",
     "IP限制": "IP限制",
     "IP黑名单": "IP黑名单",
     "JSON": "JSON",
@@ -150,7 +150,6 @@
     "URL链接": "URL链接",
     "USD (美元)": "USD (美元)",
     "User Info Endpoint": "User Info Endpoint",
-    "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段": "Vertex AI 不支持 functionResponse.id 字段，开启后将自动移除该字段",
     "Webhook 密钥": "Webhook 密钥",
     "Webhook 签名密钥": "Webhook 签名密钥",
     "Webhook地址": "Webhook地址",
@@ -544,7 +543,6 @@
     "参数值": "参数值",
     "参数覆盖": "参数覆盖",
     "参照生视频": "参照生视频",
-    "视频Remix": "视频 Remix",
     "友情链接": "友情链接",
     "发布日期": "发布日期",
     "发布时间": "发布时间",
@@ -1499,7 +1497,6 @@
     "私有IP访问详细说明": "⚠️ 安全警告：启用此选项将允许访问内网资源（本地主机、私有网络）。仅在需要访问内部服务且了解安全风险的情况下启用。",
     "私有部署地址": "私有部署地址",
     "秒": "秒",
-    "移除 functionResponse.id 字段": "移除 functionResponse.id 字段",
     "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目": "移除 One API 的版权标识必须首先获得授权，项目维护需要花费大量精力，如果本项目对你有意义，请主动支持本项目",
     "窗口处理": "窗口处理",
     "窗口等待": "窗口等待",
@@ -1742,7 +1739,7 @@
     "请先阅读并同意用户协议和隐私政策": "请先阅读并同意用户协议和隐私政策",
     "请再次输入新密码": "请再次输入新密码",
     "请前往个人设置 → 安全设置进行配置。": "请前往个人设置 → 安全设置进行配置。",
-    "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用": "请勿过度信任此功能，IP可能被伪造，请配合nginx和cdn等网关使用",
+    "请勿过度信任此功能，IP可能被伪造": "请勿过度信任此功能，IP可能被伪造",
     "请在系统设置页面编辑分组倍率以添加新的分组：": "请在系统设置页面编辑分组倍率以添加新的分组：",
     "请填写完整的产品信息": "请填写完整的产品信息",
     "请填写完整的管理员账号信息": "请填写完整的管理员账号信息",
@@ -2205,9 +2202,6 @@
     "默认用户消息": "你好",
     "默认助手消息": "你好！有什么我可以帮助你的吗？",
     "可选，用于复现结果": "可选，用于复现结果",
-    "随机种子 (留空为随机)": "随机种子 (留空为随机)",
-    "跨分组重试": "跨分组重试",
-    "跨分组": "跨分组",
-    "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道": "开启后，当前分组渠道失败时会按顺序尝试下一个分组的渠道"
+    "随机种子 (留空为随机)": "随机种子 (留空为随机)"
   }
 }