diff --git a/.gitignore b/.gitignore
index 9769aef1d042..6e3b5b7ab521 100644
--- a/.gitignore
+++ b/.gitignore
@@ -16,4 +16,6 @@ script/
 *_test.go
 out.log
 out.log_2
-.env_3
\ No newline at end of file
+.env_3
+*.prof
+.env*
\ No newline at end of file
diff --git a/common/constants.go b/common/constants.go
index 669fd93b10b8..e01593b0cd54 100644
--- a/common/constants.go
+++ b/common/constants.go
@@ -239,6 +239,7 @@ const (
 	ChannelTypeVolcEngine     = 45
 	ChannelTypeBaiduV2        = 46
 	ChannelTypeXai            = 47
+	ChannelTypeDoubaoOffline  = 100
 	ChannelTypeDummy          // this one is only for count, do not add any channel after this
 
 )
@@ -293,4 +294,56 @@ var ChannelBaseURLs = []string{
 	"https://qianfan.baidubce.com",              //46
 	"",                                          //47
 	"https://api.x.ai",                          //48
+	"",                                          //49
+	"",                                          //50
+	"",                                          //51
+	"",                                          //52
+	"",                                          //53
+	"",                                          //54
+	"",                                          //55
+	"",                                          //56
+	"",                                          //57
+	"",                                          //58
+	"",                                          //59
+	"",                                          //60
+	"",                                          //61
+	"",                                          //62
+	"",                                          //63
+	"",                                          //64
+	"",                                          //65
+	"",                                          //66
+	"",                                          //67
+	"",                                          //68
+	"",                                          //69
+	"",                                          //70
+	"",                                          //71
+	"",                                          //72
+	"",                                          //73
+	"",                                          //74
+	"",                                          //75
+	"",                                          //76
+	"",                                          //77
+	"",                                          //78
+	"",                                          //79
+	"",                                          //80
+	"",                                          //81
+	"",                                          //82
+	"",                                          //83
+	"",                                          //84
+	"",                                          //85
+	"",                                          //86
+	"",                                          //87
+	"",                                          //88
+	"",                                          //89
+	"",                                          //90
+	"",                                          //91
+	"",                                          //92
+	"",                                          //93
+	"",                                          //94
+	"",                                          //95
+	"",                                          //96
+	"",                                          //97
+	"",                                          //98
+	"",                                          //99
+	"https://ark.cn-beijing.volces.com",         //100 - 豆包离线
 }
diff --git a/common/logger.go b/common/logger.go
index 123efb158dc1..735a3f8e38c6 100644
--- a/common/logger.go
+++ b/common/logger.go
@@ -10,9 +10,12 @@ import (
 	"path/filepath"
 	"runtime"
 	"sort"
+	"strings"
 	"sync"
 	"time"
 
+	"one-api/metrics"
+
 	"github.com/bytedance/gopkg/util/gopool"
 	"github.com/gin-gonic/gin"
 )
@@ -35,6 +38,20 @@ const (
 	maxLogCount = 1000000
 )
 
+// 错误类型常量
+const (
+	ErrorTypeOther              = "other"
+	ErrorTypeParameter          = "parameter_error"
+	ErrorTypeNoCandidates       = "no_candidates"
+	ErrorTypeRequestFailed      = "request_failed"
+	ErrorTypeBadGateway         = "bad_gateway"
+	ErrorTypeResponseFailed     = "response_failed"
+	ErrorTypeConnectionTimeout  = "connection_timeout"
+	ErrorTypeTokenUnavailable   = "token_unavailable"
+	ErrorTypeBadRequest         = "bad_request"
+	ErrorTypeNoAvailableChannel = "no_available_channel"
+)
+
 var logCount int
 var setupLogLock sync.Mutex
 var setupLogWorking bool
@@ -184,6 +201,42 @@ func LogError(ctx context.Context, msg string) {
 	logHelper(ctx, loggerError, msg)
 }
 
+// 获取错误类型
+func getErrorType(msg string) (string, string) {
+	// 提取错误码（如果有）
+	errorCode := "unknown"
+	if strings.Contains(msg, "status code:") {
+		parts := strings.Split(msg, "status code:")
+		if len(parts) > 1 {
+			errorCode = strings.TrimSpace(parts[1])
+		}
+	}
+
+	// 根据错误消息内容判断错误类型
+	switch {
+	case strings.Contains(msg, "One or more parameter"):
+		return ErrorTypeParameter, errorCode
+	case strings.Contains(msg, "No candidates"):
+		return ErrorTypeNoCandidates, errorCode
+	case strings.Contains(msg, "do request failed"):
+		return ErrorTypeRequestFailed, errorCode
+	case strings.Contains(msg, "status code: 502"):
+		return ErrorTypeBadGateway, errorCode
+	case strings.Contains(msg, "doResponse failed"):
+		return ErrorTypeResponseFailed, errorCode
+	case strings.Contains(msg, "write: connection timed out"):
+		return ErrorTypeConnectionTimeout, errorCode
+	case strings.Contains(msg, "该令牌状态不可用"):
+		return ErrorTypeTokenUnavailable, errorCode
+	case strings.Contains(msg, "bad response status code 400"):
+		return ErrorTypeBadRequest, errorCode
+	case strings.Contains(msg, "无可用渠道"):
+		return ErrorTypeNoAvailableChannel, errorCode
+	default:
+		return ErrorTypeOther, errorCode
+	}
+}
+
 func logHelper(ctx context.Context, level string, msg string) {
 	// 获取请求ID
 	var requestId string
@@ -209,6 +262,38 @@ func logHelper(ctx context.Context, level string, msg string) {
 	now := time.Now()
 	caller := getCallerInfo()
 	_, _ = fmt.Fprintf(writer, "[%s] %v | %s | %s | %s \n", level, now.Format("2006/01/02 - 15:04:05"), requestId, caller, msg)
+
+	// 如果是错误日志，增加错误计数
+	if level == loggerError {
+		errorType, errorCode := getErrorType(msg)
+		// 从上下文中获取相关信息
+		channel := "unknown"
+		channelName := "unknown"
+		model := "unknown"
+		group := "unknown"
+		tokenName := "unknown"
+
+		if ginCtx, ok := ctx.Value("gin_context").(*gin.Context); ok {
+			if ch := ginCtx.GetString("channel"); ch != "" {
+				channel = ch
+			}
+			if chName := ginCtx.GetString("channel_name"); chName != "" {
+				channelName = chName
+			}
+			if m := ginCtx.GetString("model"); m != "" {
+				model = m
+			}
+			if g := ginCtx.GetString("group"); g != "" {
+				group = g
+			}
+			if tn := ginCtx.GetString("token_name"); tn != "" {
+				tokenName = tn
+			}
+		}
+
+		metrics.IncrementErrorLog(channel, channelName, errorCode, errorType, model, group, tokenName, 1.0)
+	}
+
 	logCount++ // we don't need accurate count, so no lock here
 	if logCount > maxLogCount && !setupLogWorking {
 		logCount = 0
diff --git a/controller/misc.go b/controller/misc.go
index a451b5e3faa5..e5588edbe4ef 100644
--- a/controller/misc.go
+++ b/controller/misc.go
@@ -270,3 +270,13 @@ func ResetPassword(c *gin.Context) {
 	})
 	return
 }
+
+func Ping(c *gin.Context) {
+	c.Writer.Header().Set("Retry_request_id", "Retry_request_id")
+	c.JSON(http.StatusOK, gin.H{
+		"success":   true,
+		"message":   "pong",
+		"timestamp": common.GetTimestamp(),
+	})
+	return
+}
diff --git a/controller/relay.go b/controller/relay.go
index 6031ba4e2dae..4a4f88eab75a 100644
--- a/controller/relay.go
+++ b/controller/relay.go
@@ -4,8 +4,6 @@ import (
 	"bytes"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
-	"github.com/gorilla/websocket"
 	"io"
 	"log"
 	"net/http"
@@ -23,6 +21,9 @@ import (
 	"strconv"
 	"strings"
 	"time"
+
+	"github.com/gin-gonic/gin"
+	"github.com/gorilla/websocket"
 )
 
 func relayInfoHandler(c *gin.Context, relayMode int) (*relaycommon.RelayInfo, interface{}, string, *dto.OpenAIErrorWithStatusCode) {
@@ -121,6 +122,9 @@ func Relay(c *gin.Context) {
 			openaiErr = service.OpenAIErrorWrapperLocal(err, "get_channel_failed", http.StatusInternalServerError)
 			break
 		}
+		// 设置 channel 信息到上下文
+		c.Set("channel", strconv.Itoa(channel.Id))
+		c.Set("channel_name", channel.Name)
 		fillRelayRequest(c, channel)
 		var (
 			relayInfo    *relaycommon.RelayInfo
@@ -130,20 +134,20 @@ func Relay(c *gin.Context) {
 		relayInfo, request, requestModel, openaiErr = relayInfoHandler(c, relayMode)
 		if i == 0 {
 			// e2e 用户请求计数
-			metrics.IncrementRelayRequestE2ETotalCounter(strconv.Itoa(channel.Id), requestModel, group, tokenKey, tokenName, 1)
+			metrics.IncrementRelayRequestE2ETotalCounter(strconv.Itoa(channel.Id), channel.Name, requestModel, group, tokenKey, tokenName, 1)
 		} else {
 			// 重试计数
 			channelTag := ""
 			if channel.Tag != nil {
 				channelTag = *channel.Tag
 			}
-			metrics.IncrementRelayRetryCounter(strconv.Itoa(channel.Id), channelTag, channel.GetBaseURL(), requestModel, group, 1)
+			metrics.IncrementRelayRetryCounter(strconv.Itoa(channel.Id), channel.Name, channelTag, channel.GetBaseURL(), requestModel, group, 1)
 		}
 		if openaiErr == nil {
 			openaiErr = executeRelayRequest(c, relayMode, relayInfo, request)
 			if openaiErr == nil {
-				metrics.IncrementRelayRequestE2ESuccessCounter(strconv.Itoa(channel.Id), requestModel, group, tokenKey, tokenName, 1)
-				metrics.ObserveRelayRequestE2EDuration(strconv.Itoa(channel.Id), requestModel, group, tokenKey, tokenName, time.Since(startTime).Seconds())
+				metrics.IncrementRelayRequestE2ESuccessCounter(strconv.Itoa(channel.Id), channel.Name, requestModel, group, tokenKey, tokenName, 1)
+				metrics.ObserveRelayRequestE2EDuration(strconv.Itoa(channel.Id), channel.Name, requestModel, group, tokenKey, tokenName, time.Since(startTime).Seconds())
 				return
 			}
 		}
@@ -152,7 +156,7 @@ func Relay(c *gin.Context) {
 
 		if !shouldRetry(c, openaiErr, common.RetryTimes-i) {
 			// e2e 失败计数
-			metrics.IncrementRelayRequestE2EFailedCounter(strconv.Itoa(channel.Id), requestModel, group, strconv.Itoa(openaiErr.StatusCode), tokenKey, tokenName, 1)
+			metrics.IncrementRelayRequestE2EFailedCounter(strconv.Itoa(channel.Id), channel.Name, requestModel, group, strconv.Itoa(openaiErr.StatusCode), tokenKey, tokenName, 1)
 			break
 		}
 	}
@@ -167,6 +171,33 @@ func Relay(c *gin.Context) {
 			common.LogError(c, fmt.Sprintf("origin 429 error: %s", openaiErr.Error.Message))
 			openaiErr.Error.Message = "当前分组上游负载已饱和，请稍后再试"
 		}
+		
+		// 处理自定义的 NewAPI batch 错误码
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchRateLimitExceeded {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "当前服务端限速已满，请稍后再试"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchTimeout {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "未等待到结果，请稍后使用Retry_request_id再次查询"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchInternal {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "服务内部错误，请稍后再试"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchSubmitted {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "批量请求已提交，但是结果还未出来，请使用Retry_request_id查询结果"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchAccepted {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "批量请求已接受，正在处理中，请稍后使用Retry_request_id查询结果"
+		}
+		if openaiErr.StatusCode == dto.StatusRequestConflict {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "请求冲突，有其他请求使用了这个Retry_request_id，请稍后再试"
+		}
+
 		openaiErr.Error.Message = common.MessageWithRequestId(openaiErr.Error.Message, requestId)
 		c.JSON(openaiErr.StatusCode, gin.H{
 			"error": openaiErr.Error,
@@ -230,6 +261,31 @@ func WssRelay(c *gin.Context) {
 		if openaiErr.StatusCode == http.StatusTooManyRequests {
 			openaiErr.Error.Message = "当前分组上游负载已饱和，请稍后再试"
 		}
+		// 处理自定义的 NewAPI batch 错误码
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchRateLimitExceeded {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "当前服务端限速已满，请稍后再试"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchTimeout {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "未等待到结果，请稍后使用Retry_request_id再次查询"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchInternal {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "服务内部错误，请稍后再试"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchSubmitted {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "批量请求已提交，但是结果还未出来，请使用Retry_request_id查询结果"
+		}
+		if openaiErr.StatusCode == dto.StatusNewAPIBatchAccepted {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "批量请求已接受，正在处理中，请稍后使用Retry_request_id查询结果"
+		}
+		if openaiErr.StatusCode == dto.StatusRequestConflict {
+			common.LogError(c, fmt.Sprintf("origin %d error: %s", openaiErr.StatusCode, openaiErr.Error.Message))
+			openaiErr.Error.Message = "请求冲突，有其他请求使用了这个Retry_request_id，请稍后再试"
+		}
 		openaiErr.Error.Message = common.MessageWithRequestId(openaiErr.Error.Message, requestId)
 		helper.WssError(c, ws, openaiErr.Error)
 	}
@@ -298,6 +354,26 @@ func shouldRetry(c *gin.Context, openaiErr *dto.OpenAIErrorWithStatusCode, retry
 	if openaiErr.StatusCode == http.StatusTooManyRequests {
 		return true
 	}
+	// 处理自定义的 NewAPI batch 错误码
+	if openaiErr.StatusCode == dto.StatusNewAPIBatchRateLimitExceeded {
+		return false
+	}
+	if openaiErr.StatusCode == dto.StatusNewAPIBatchTimeout {
+		return false
+	}
+	if openaiErr.StatusCode == dto.StatusNewAPIBatchInternal {
+		return false
+	}
+	if openaiErr.StatusCode == dto.StatusNewAPIBatchSubmitted {
+		return false
+	}
+	if openaiErr.StatusCode == dto.StatusNewAPIBatchAccepted {
+		return false
+	}
+	if openaiErr.StatusCode == dto.StatusRequestConflict {
+		return false
+	}
+
 	if openaiErr.StatusCode == 307 {
 		return true
 	}
diff --git a/dto/error.go b/dto/error.go
index b347f6a159ef..eaee8ac52097 100644
--- a/dto/error.go
+++ b/dto/error.go
@@ -53,3 +53,13 @@ func (e GeneralErrorResponse) ToMessage() string {
 	}
 	return ""
 }
+
+// 自定义HTTP状态码 (使用非标准状态码范围)
+const (
+	StatusNewAPIBatchRateLimitExceeded = 499 // 自定义限流状态码
+	StatusNewAPIBatchTimeout           = 598 // 自定义超时状态码
+	StatusNewAPIBatchInternal          = 599 // 自定义内部错误状态码
+	StatusNewAPIBatchSubmitted         = 203 // 批量请求已提交，需要重试获取结果
+	StatusNewAPIBatchAccepted          = 202 // 批量请求已接受，正在处理中
+	StatusRequestConflict              = 409 // 请求冲突，如分布式锁获取失败
+)
diff --git a/dto/openai_request.go b/dto/openai_request.go
index 2170e55d7c61..71dd9ad1833e 100644
--- a/dto/openai_request.go
+++ b/dto/openai_request.go
@@ -2,7 +2,10 @@ package dto
 
 import (
 	"encoding/json"
+	"fmt"
 	"strings"
+
+	"one-api/common"
 )
 
 type ResponseFormat struct {
@@ -42,6 +45,7 @@ type GeneralOpenAIRequest struct {
 	ResponseFormat      *ResponseFormat   `json:"response_format,omitempty"`
 	EncodingFormat      any               `json:"encoding_format,omitempty"`
 	Seed                float64           `json:"seed,omitempty"`
+	LogitBias           map[string]int    `json:"logit_bias,omitempty"`
 	Tools               []ToolCallRequest `json:"tools,omitempty"`
 	ToolChoice          any               `json:"tool_choice,omitempty"`
 	User                string            `json:"user,omitempty"`
@@ -286,6 +290,13 @@ func (m *Message) ParseContent() []MediaContent {
 				if audioData, ok := contentItem["input_audio"].(map[string]interface{}); ok {
 					data, ok1 := audioData["data"].(string)
 					format, ok2 := audioData["format"].(string)
+					if !ok2 {
+						if mimeType, ok3 := audioData["mime_type"].(string); ok3 {
+							format = mimeType
+							ok2 = true
+						}
+					}
+					common.SysLog(fmt.Sprintf("Parsing audio content: data_ok=%v, format_ok=%v, format=%s, data_length=%d", ok1, ok2, format, len(data)))
 					if ok1 && ok2 {
 						contentList = append(contentList, MediaContent{
 							Type: ContentTypeInputAudio,
diff --git a/go.mod b/go.mod
index 83fe5a067654..8c09f403784a 100644
--- a/go.mod
+++ b/go.mod
@@ -28,6 +28,7 @@ require (
 	github.com/prometheus/client_golang v1.21.1
 	github.com/samber/lo v1.39.0
 	github.com/shirou/gopsutil v3.21.11+incompatible
+	github.com/volcengine/volcengine-go-sdk v1.1.17
 	github.com/xuri/excelize/v2 v2.9.0
 	golang.org/x/crypto v0.31.0
 	golang.org/x/image v0.23.0
@@ -69,6 +70,7 @@ require (
 	github.com/jackc/puddle/v2 v2.2.2 // indirect
 	github.com/jinzhu/inflection v1.0.0 // indirect
 	github.com/jinzhu/now v1.1.5 // indirect
+	github.com/jmespath/go-jmespath v0.4.0 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
 	github.com/klauspost/compress v1.17.11 // indirect
 	github.com/klauspost/cpuid/v2 v2.2.9 // indirect
@@ -90,6 +92,7 @@ require (
 	github.com/tklauser/numcpus v0.6.1 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	github.com/ugorji/go/codec v1.2.12 // indirect
+	github.com/volcengine/volc-sdk-golang v1.0.23 // indirect
 	github.com/xuri/efp v0.0.0-20240408161823-9ad904a10d6d // indirect
 	github.com/xuri/nfp v0.0.0-20240318013403-ab9948c2c4a7 // indirect
 	github.com/yusufpapurcu/wmi v1.2.3 // indirect
@@ -99,6 +102,7 @@ require (
 	golang.org/x/sys v0.28.0 // indirect
 	golang.org/x/text v0.21.0 // indirect
 	google.golang.org/protobuf v1.36.1 // indirect
+	gopkg.in/yaml.v2 v2.4.0 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 	modernc.org/libc v1.22.5 // indirect
 	modernc.org/mathutil v1.5.0 // indirect
diff --git a/go.sum b/go.sum
index bb2c72adc401..f8016b2ddcf3 100644
--- a/go.sum
+++ b/go.sum
@@ -1,3 +1,5 @@
+cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
+github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
 github.com/Calcium-Ion/go-epay v0.0.4 h1:C96M7WfRLadcIVscWzwLiYs8etI1wrDmtFMuK2zP22A=
 github.com/Calcium-Ion/go-epay v0.0.4/go.mod h1:cxo/ZOg8ClvE3VAnCmEzbuyAZINSq7kFEN9oHj5WQ2U=
 github.com/andybalholm/brotli v1.1.1 h1:PR2pgnyFznKEugtsUo0xLdDop5SKXd5Qf5ysW+7XdTA=
@@ -6,6 +8,7 @@ github.com/anknown/ahocorasick v0.0.0-20190904063843-d75dbd5169c0 h1:onfun1RA+Kc
 github.com/anknown/ahocorasick v0.0.0-20190904063843-d75dbd5169c0/go.mod h1:4yg+jNTYlDEzBjhGS96v+zjyA3lfXlFd5CiTLIkPBLI=
 github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6 h1:HblK3eJHq54yET63qPCTJnks3loDse5xRmmqHgHzwoI=
 github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6/go.mod h1:pbiaLIeYLUbgMY1kwEAdwO6UKD5ZNwdPGQlwokS9fe8=
+github.com/avast/retry-go v3.0.0+incompatible/go.mod h1:XtSnn+n/sHqQIpZ10K1qAevBhOOCWBLXXy3hyiqqBrY=
 github.com/aws/aws-sdk-go-v2 v1.26.1 h1:5554eUqIYVWpU0YmeeYZ0wU64H2VLBs8TlhRB2L+EkA=
 github.com/aws/aws-sdk-go-v2 v1.26.1/go.mod h1:ffIFB97e2yNsv4aTSGkqtHnppsIJzw7G7BReUZ3jCXM=
 github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2 h1:x6xsQXGSmW6frevwDA+vi/wqhp1ct18mVXYN08/93to=
@@ -28,8 +31,10 @@ github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc
 github.com/bytedance/sonic v1.11.6/go.mod h1:LysEHSvpvDySVdC2f87zGWf6CIKJcAvqab1ZaiQtds4=
 github.com/bytedance/sonic/loader v0.1.1 h1:c+e5Pt1k/cy5wMveRDyk2X4B9hF4g7an8N3zCYjJFNM=
 github.com/bytedance/sonic/loader v0.1.1/go.mod h1:ncP89zfokxS5LZrJxl5z0UJcsk4M4yY2JpfqGeCtNLU=
+github.com/census-instrumentation/opencensus-proto v0.2.1/go.mod h1:f6KPmirojxKA12rnyqOA5BBL4O983OfeGPqjHWSTneU=
 github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs=
 github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
+github.com/client9/misspell v0.3.4/go.mod h1:qj6jICC3Q7zFZvVWo7KLAzC3yx5G7kyvSDkc90ppPyw=
 github.com/cloudwego/base64x v0.1.4 h1:jwCgWpFanWmN8xoIUHa2rtzmkd5J2plF/dnLS6Xd/0Y=
 github.com/cloudwego/base64x v0.1.4/go.mod h1:0zlkT4Wn5C6NdauXdJRhSKRlJvmclQ1hhJgA0rcu/8w=
 github.com/cloudwego/iasm v0.2.0 h1:1KNIy1I1H9hNNFEEH3DVnI4UujN+1zjpuk6gwHLTssg=
@@ -44,6 +49,8 @@ github.com/dlclark/regexp2 v1.11.0 h1:G/nrcoOa7ZXlpoa/91N3X7mM3r8eIlMBBJZvsz/mxK
 github.com/dlclark/regexp2 v1.11.0/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
 github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkpeCY=
 github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
+github.com/envoyproxy/go-control-plane v0.9.1-0.20191026205805-5f8ba28d4473/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
+github.com/envoyproxy/protoc-gen-validate v0.1.0/go.mod h1:iSmxcyjqTsJpI2R4NaDN7+kN2VEUnK/pcBlmesArF7c=
 github.com/fsnotify/fsnotify v1.4.9 h1:hsms1Qyu0jgnwNXIxa+/V/PDsU6CfLf6CNO8H7IWoS4=
 github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ=
 github.com/gabriel-vasile/mimetype v1.4.3 h1:in2uUcidCuFcDKtdcBxlR0rJ1+fsokWf+uqxgUFjbI0=
@@ -93,14 +100,31 @@ github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU=
 github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
 github.com/golang-jwt/jwt v3.2.2+incompatible h1:IfV12K8xAKAnZqdXVzCZ+TOjboZ2keLg81eXfW3O+oY=
 github.com/golang-jwt/jwt v3.2.2+incompatible/go.mod h1:8pz2t5EyA70fFQQSrl6XZXzqecmYZeUEB8OUGHkxJ+I=
+github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b/go.mod h1:SBH7ygxi8pfUlaOkMMuAQtPIUF8ecWP5IEl/CR7VP2Q=
+github.com/golang/mock v1.1.1/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
+github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/protobuf v1.3.2/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
 github.com/golang/protobuf v1.3.3/go.mod h1:vzj43D7+SQXF/4pzW/hwtAqwc6iTitCiVSaWz5lYuqw=
+github.com/golang/protobuf v1.4.0-rc.1/go.mod h1:ceaxUfeHdC40wWswd/P6IGgMaK3YpKi5j83Wpe3EHw8=
+github.com/golang/protobuf v1.4.0-rc.1.0.20200221234624-67d41d38c208/go.mod h1:xKAWHe0F5eneWXFV3EuXVDTCmh+JuBKY0li0aMyXATA=
+github.com/golang/protobuf v1.4.0-rc.2/go.mod h1:LlEzMj4AhA7rCAGe4KMBDvJI+AwstrUpVNzEA03Pprs=
+github.com/golang/protobuf v1.4.0-rc.4.0.20200313231945-b860323f09d0/go.mod h1:WU3c8KckQ9AFe+yFwt9sWVRKCVIyN9cPHBJSNnbL67w=
+github.com/golang/protobuf v1.4.0/go.mod h1:jodUvKwWbYaEsadDk5Fwe5c77LiNKVO9IDvqG2KuDX0=
+github.com/golang/protobuf v1.4.1/go.mod h1:U8fpvMrcmy5pZrNK1lt4xCsGvpyWQ/VVv6QDs8UjoX8=
+github.com/golang/protobuf v1.4.3/go.mod h1:oDoupMAO8OvCJWAcko0GGGIgR6R6ocIYbsSw735rRwI=
 github.com/golang/protobuf v1.5.0/go.mod h1:FsONVRAS9T7sI+LIUmWTfcYkHO4aIWwzhcaSAoJOfIk=
+github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
+github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
+github.com/google/go-cmp v0.3.1/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
+github.com/google/go-cmp v0.4.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
+github.com/google/go-cmp v0.5.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
 github.com/google/go-cmp v0.5.5/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
 github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
 github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 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/uuid v1.3.0/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=
 github.com/gorilla/context v1.1.1 h1:AWwleXJkX/nhcU9bZSnZoi3h/qGYqQAGhq6zZe/aQW8=
@@ -124,6 +148,10 @@ github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkr
 github.com/jinzhu/now v1.1.4/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8=
 github.com/jinzhu/now v1.1.5 h1:/o9tlHleP7gOFmsnYNz3RGnqzefHA47wQpKrrdTIwXQ=
 github.com/jinzhu/now v1.1.5/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8=
+github.com/jmespath/go-jmespath v0.4.0 h1:BEgLn5cpjn8UN1mAw4NjwDrS35OdebyEtFe+9YPoQUg=
+github.com/jmespath/go-jmespath v0.4.0/go.mod h1:T8mJZnbsbmF+m6zOOFylbeCJqk5+pHWvzYPziyZiYoo=
+github.com/jmespath/go-jmespath/internal/testify v1.5.1 h1:shLQSRRSCCPj3f2gpwzGwWFoC7ycTf1rcQZHOlsJ6N8=
+github.com/jmespath/go-jmespath/internal/testify v1.5.1/go.mod h1:L3OGu8Wl2/fWfCI6z80xFu9LTZmf1ZRjMHUOPmWr69U=
 github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
 github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4=
 github.com/json-iterator/go v1.1.9/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
@@ -136,6 +164,7 @@ github.com/klauspost/cpuid/v2 v2.2.9 h1:66ze0taIn2H33fBvCkXuv9BmCwDfafmiIVpKV9kK
 github.com/klauspost/cpuid/v2 v2.2.9/go.mod h1:rqkxqrZ1EhYM9G+hXH7YdowN5R5RGN6NK4QwQ3WMXF8=
 github.com/knz/go-libedit v1.10.1/go.mod h1:MZTVkCWyz0oBc7JOWP3wNAzd002ZbM/5hgShxwh4x8M=
 github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
+github.com/kr/pretty v0.2.0/go.mod h1:ipq/a2n7PKx3OHsz4KJII5eveXtPO4qwEXGdVfWzfnI=
 github.com/kr/pretty v0.2.1/go.mod h1:ipq/a2n7PKx3OHsz4KJII5eveXtPO4qwEXGdVfWzfnI=
 github.com/kr/pretty v0.3.0/go.mod h1:640gp4NfQd8pI5XOwp5fnNeVWj67G7CFk/SaSQn7NBk=
 github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
@@ -184,6 +213,7 @@ github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZb
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/prometheus/client_golang v1.21.1 h1:DOvXXTqVzvkIewV/CDPFdejpMCGeMcbGCQ8YOmu+Ibk=
 github.com/prometheus/client_golang v1.21.1/go.mod h1:U9NM32ykUErtVBxdvD3zfi+EuFkkaBvMb09mIfe0Zgg=
+github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/prometheus/client_model v0.6.1 h1:ZKSh/rekM+n3CeS952MLRAdFwIKqeY8b62p8ais2e9E=
 github.com/prometheus/client_model v0.6.1/go.mod h1:OrxVMOVHjw3lKMa8+x6HeMGkHMQyHDk9E3jmP2AmGiY=
 github.com/prometheus/common v0.62.0 h1:xasJaQlnWAeyHdUBeGjXmutelfJHWMRr+Fg4QszZ2Io=
@@ -212,6 +242,7 @@ github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpE
 github.com/stretchr/objx v0.5.2/go.mod h1:FRsXN1f5AsAjCGJKqEizvkpNtU+EGNCLh3NxZ/8L+MA=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
+github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA=
 github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
 github.com/stretchr/testify v1.7.0/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
 github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
@@ -233,6 +264,10 @@ github.com/ugorji/go/codec v1.1.7/go.mod h1:Ax+UKWsSmolVDwsd+7N3ZtXu+yMGCf907BLY
 github.com/ugorji/go/codec v1.2.7/go.mod h1:WGN1fab3R1fzQlVQTkfxVtIBhWDRqOviHU95kRgeqEY=
 github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65EE=
 github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
+github.com/volcengine/volc-sdk-golang v1.0.23 h1:anOslb2Qp6ywnsbyq9jqR0ljuO63kg9PY+4OehIk5R8=
+github.com/volcengine/volc-sdk-golang v1.0.23/go.mod h1:AfG/PZRUkHJ9inETvbjNifTDgut25Wbkm2QoYBTbvyU=
+github.com/volcengine/volcengine-go-sdk v1.1.17 h1:Izrcx/FERzGvpY3ufPjt4GR7Ak6y94aMVXbnLmeuw2g=
+github.com/volcengine/volcengine-go-sdk v1.1.17/go.mod h1:EyKoi6t6eZxoPNGr2GdFCZti2Skd7MO3eUzx7TtSvNo=
 github.com/xuri/efp v0.0.0-20240408161823-9ad904a10d6d h1:llb0neMWDQe87IzJLS4Ci7psK/lVsjIS2otl+1WyRyY=
 github.com/xuri/efp v0.0.0-20240408161823-9ad904a10d6d/go.mod h1:ybY/Jr0T0GTCnYjKqmdwxyxn2BQf2RcQIIvex5QldPI=
 github.com/xuri/excelize/v2 v2.9.0 h1:1tgOaEq92IOEumR1/JfYS/eR0KHOCsRv/rYXXh6YJQE=
@@ -246,19 +281,34 @@ github.com/yusufpapurcu/wmi v1.2.3/go.mod h1:SBZ9tNy3G9/m5Oi98Zks0QjeHVDvuK0qfxQ
 golang.org/x/arch v0.0.0-20210923205945-b76863e36670/go.mod h1:5om86z9Hs0C8fWVUuoMHwpExlXzs5Tkyp9hOrfG7pp8=
 golang.org/x/arch v0.12.0 h1:UsYJhbzPYGsT0HbEdmYcqtCv8UNGvnaL561NnIUvaKg=
 golang.org/x/arch v0.12.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys=
+golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20210711020723-a769d52b0f97/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
 golang.org/x/crypto v0.31.0 h1:ihbySMvVjLAeSH1IbfcRTkD/iNscyz8rGzjF/E5hV6U=
 golang.org/x/crypto v0.31.0/go.mod h1:kDsLvtWBEx7MV9tJOj9bnXsPbxwJQ6csT/x4KIN4Ssk=
+golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 h1:985EYyeCOxTpcgOTJpflJUwOeEz0CQOdPt73OzpE9F8=
 golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0/go.mod h1:/lliqkxwWAhPjf5oSOIJup2XcqJaw8RGS6k3TGEc7GI=
 golang.org/x/image v0.23.0 h1:HseQ7c2OpPKTPVzNjG5fwJsOTCiiwS4QdsYi5XU6H68=
 golang.org/x/image v0.23.0/go.mod h1:wJJBTdLfCCf3tiHa1fNxpZmUI4mmoZvwMCPP0ddoNKY=
+golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
+golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU=
+golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
+golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20190213061140-3a22650c66bd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
 golang.org/x/net v0.33.0 h1:74SYHlV8BIgHIFC/LrYkOGIwL19eTYXQ5wc6TBuO36I=
 golang.org/x/net v0.33.0/go.mod h1:HXLR5J+9DxmrqMwG9qjGCxZ+zKXxBru04zlTvWlWuN4=
+golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
+golang.org/x/sync v0.0.0-20180314180146-1d60e4601c6f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20210220032951-036812b2e83c/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.10.0 h1:3NQrjDixjgGwUOCaF8w2+VYHv0Ve/vGYSbdkTa98gmQ=
 golang.org/x/sync v0.10.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
+golang.org/x/sys v0.0.0-20180830151530-49385e6e1522/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20201119102817-f84b799fce68/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
@@ -272,19 +322,42 @@ golang.org/x/sys v0.11.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.28.0 h1:Fksou7UEQUWlKvIdsqzJmUmCX3cZuD2+P3XyyzwMhlA=
 golang.org/x/sys v0.28.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
+golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.21.0 h1:zyQAAkrwaneQ066sspRyJaG9VNi/YJ1NfzcGB3hZ/qo=
 golang.org/x/text v0.21.0/go.mod h1:4IBbMaMmOPCJ8SecivzSH54+73PCFmPWxNTLm+vZkEQ=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20190114222345-bf090417da8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20190226205152-f727befe758c/go.mod h1:9Yl7xja0Znq3iFh3HoIrodX9oNMXvdceNzlUR8zjMvY=
+golang.org/x/tools v0.0.0-20190311212946-11955173bddd/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
+golang.org/x/tools v0.0.0-20190524140312-2c0ae7006135/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+google.golang.org/appengine v1.1.0/go.mod h1:EbEs0AVv82hx2wNQdGPgUI5lhzA/G0D9YwlJXL52JkM=
+google.golang.org/appengine v1.4.0/go.mod h1:xpcJRLb0r/rnEns0DIKYYv+WjYCduHsrkT7/EB5XEv4=
+google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
+google.golang.org/genproto v0.0.0-20190819201941-24fa4b261c55/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
+google.golang.org/genproto v0.0.0-20200526211855-cb27e3aa2013/go.mod h1:NbSheEEYHJ7i3ixzK3sjbqSGDJWnxyFXZblF3eUsNvo=
+google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
+google.golang.org/grpc v1.23.0/go.mod h1:Y5yQAOtifL1yxbo5wqy6BxZv8vAUGQwXBOALyacEbxg=
+google.golang.org/grpc v1.27.0/go.mod h1:qbnxyOmOxrQa7FizSgH+ReBfzJrCY1pSN7KXBS8abTk=
+google.golang.org/protobuf v0.0.0-20200109180630-ec00e32a8dfd/go.mod h1:DFci5gLYBciE7Vtevhsrf46CRTquxDuWsQurQQe4oz8=
+google.golang.org/protobuf v0.0.0-20200221191635-4d8936d0db64/go.mod h1:kwYJMbMJ01Woi6D6+Kah6886xMZcty6N08ah7+eCXa0=
+google.golang.org/protobuf v0.0.0-20200228230310-ab0ca4ff8a60/go.mod h1:cfTl7dwQJ+fmap5saPgwCLgHXTUD7jkjRqWcaiX5VyM=
+google.golang.org/protobuf v1.20.1-0.20200309200217-e05f789c0967/go.mod h1:A+miEFZTKqfCUM6K7xSMQL9OKL/b6hQv+e19PK+JZNE=
+google.golang.org/protobuf v1.21.0/go.mod h1:47Nbq4nVaFHyn7ilMalzfO3qCViNmqZ2kzikPIcrTAo=
+google.golang.org/protobuf v1.22.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
+google.golang.org/protobuf v1.23.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
+google.golang.org/protobuf v1.23.1-0.20200526195155-81db48ad09cc/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
+google.golang.org/protobuf v1.25.0/go.mod h1:9JNX74DMeImyA3h4bdi1ymwjUzf21/xIlbajtzgsN7c=
 google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
 google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
 google.golang.org/protobuf v1.36.1 h1:yBPeRvTftaleIgM3PZ/WBIZ7XM/eEYAaEyCwvyjq/gk=
 google.golang.org/protobuf v1.36.1/go.mod h1:9fA7Ob0pmnwhb644+1+CVWFRbNajQ6iRojtC/QF5bRE=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c h1:Hei/4ADfdWqJk1ZMxUNpqntNwaWcugrBjAiHlqqRiVk=
 gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c/go.mod h1:JHkPIbrfpd72SG/EVd6muEfDQjcINNoR0C8j2r3qZ4Q=
 gopkg.in/errgo.v2 v2.1.0/go.mod h1:hNsd1EY+bozCKY1Ytp96fpM3vjJbqLJn88ws8XvfDNI=
@@ -305,6 +378,8 @@ gorm.io/driver/postgres v1.5.2/go.mod h1:fmpX0m2I1PKuR7mKZiEluwrP3hbs+ps7JIGMUBp
 gorm.io/gorm v1.23.8/go.mod h1:l2lP/RyAtc1ynaTjFksBde/O8v9oOGIApu2/xRitmZk=
 gorm.io/gorm v1.25.2 h1:gs1o6Vsa+oVKG/a9ElL3XgyGfghFfkKA2SInQaCyMho=
 gorm.io/gorm v1.25.2/go.mod h1:L4uxeKpfBml98NYqVqwAdmV1a2nBtAec/cf3fpucW/k=
+honnef.co/go/tools v0.0.0-20190102054323-c2f93a96b099/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
+honnef.co/go/tools v0.0.0-20190523083050-ea95bdfd59fc/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
 modernc.org/libc v1.22.5 h1:91BNch/e5B0uPbJFgqbxXuOnxBQjlS//icfQEGmvyjE=
 modernc.org/libc v1.22.5/go.mod h1:jj+Z7dTNX8fBScMVNRAYZ/jF91K8fdT2hYMThc3YjBY=
 modernc.org/mathutil v1.5.0 h1:rV0Ko/6SfM+8G+yKiyI830l3Wuz1zRutdslNoQ0kfiQ=
diff --git a/main.go b/main.go
index eda3799c0cc7..8a31c6c0108e 100644
--- a/main.go
+++ b/main.go
@@ -28,6 +28,8 @@ import (
 	"github.com/joho/godotenv"
 
 	_ "net/http/pprof"
+
+	"one-api/relay/channel/volcengine"
 )
 
 //go:embed web/dist
@@ -134,11 +136,26 @@ func main() {
 		common.FatalLog("failed to initialize Redis: " + err.Error())
 	}
 
+	// Initialize Keep-Alive Manager for Redis keys
+	if err := volcengine.InitKeepAliveManager(); err != nil {
+		common.FatalLog("failed to initialize keep-alive manager: " + err.Error())
+	}
+	// 在应用关闭时清理保活管理器
+	defer func() {
+		if err := volcengine.ShutdownKeepAliveManager(); err != nil {
+			common.SysError("failed to shutdown keep-alive manager: " + err.Error())
+		}
+	}()
+
 	// Initialize constants
 	constant.InitEnv()
 	// Initialize options
 	model.InitOptionMap()
 	model.InitGroups()
+
+	// 初始化batch请求平均耗时
+	volcengine.InitBatchRequestAverageDuration()
+
 	if common.RedisEnabled {
 		// for compatibility with old versions
 		common.MemoryCacheEnabled = true
diff --git a/metrics/metrics.go b/metrics/metrics.go
index 348ac3951637..f0b174fe1e1b 100644
--- a/metrics/metrics.go
+++ b/metrics/metrics.go
@@ -21,11 +21,16 @@ func RegisterMetrics(registry prometheus.Registerer) {
 	registry.MustRegister(relayRequestE2ESuccessCounter)
 	registry.MustRegister(relayRequestE2EFailedCounter)
 	registry.MustRegister(relayRequestE2EDurationObsever)
+	// batch
+	registry.MustRegister(batchRequestCounter)
+	registry.MustRegister(batchRequestDurationObsever)
 	// token metrics
 	registry.MustRegister(inputTokensCounter)
 	registry.MustRegister(outputTokensCounter)
 	registry.MustRegister(cacheHitTokensCounter)
 	registry.MustRegister(inferenceTokensCounter)
+	// error log metrics
+	registry.MustRegister(errorLogCounter)
 }
 
 var (
@@ -34,25 +39,25 @@ var (
 			Subsystem: Namespace,
 			Name:      "relay_request_total",
 			Help:      "Total number of relay request total",
-		}, []string{"channel", "tag", "base_url", "model", "group"})
+		}, []string{"channel", "channel_name", "tag", "base_url", "model", "group"})
 	relayRequestSuccessCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_success",
 			Help:      "Total number of relay request success",
-		}, []string{"channel", "tag", "base_url", "model", "group"})
+		}, []string{"channel", "channel_name", "tag", "base_url", "model", "group", "code"})
 	relayRequestFailedCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_failed",
 			Help:      "Total number of relay request failed",
-		}, []string{"channel", "tag", "base_url", "model", "group", "code"})
+		}, []string{"channel", "channel_name", "tag", "base_url", "model", "group", "code"})
 	relayRequestRetryCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_retry",
 			Help:      "Total number of relay request retry",
-		}, []string{"channel", "tag", "base_url", "model", "group"})
+		}, []string{"channel", "channel_name", "tag", "base_url", "model", "group"})
 	relayRequestDurationObsever = promauto.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: Namespace,
@@ -60,26 +65,26 @@ var (
 			Help:      "Duration of relay request",
 			Buckets:   prometheus.ExponentialBuckets(1, 2, 12),
 		},
-		[]string{"channel", "tag", "base_url", "model", "group"},
+		[]string{"channel", "channel_name", "tag", "base_url", "model", "group"},
 	)
 	relayRequestE2ETotalCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_e2e_total",
 			Help:      "Total number of relay request e2e total",
-		}, []string{"channel", "model", "group", "token_key", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "token_key", "token_name"})
 	relayRequestE2ESuccessCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_e2e_success",
 			Help:      "Total number of relay request e2e success",
-		}, []string{"channel", "model", "group", "token_key", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "token_key", "token_name"})
 	relayRequestE2EFailedCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "relay_request_e2e_failed",
 			Help:      "Total number of relay request e2e failed",
-		}, []string{"channel", "model", "group", "code", "token_key", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "code", "token_key", "token_name"})
 	relayRequestE2EDurationObsever = promauto.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: Namespace,
@@ -87,7 +92,23 @@ var (
 			Help:      "Duration of relay request e2e",
 			Buckets:   prometheus.ExponentialBuckets(1, 2, 12),
 		},
-		[]string{"channel", "model", "group", "token_key", "token_name"},
+		[]string{"channel", "channel_name", "model", "group", "token_key", "token_name"},
+	)
+	// Batch request metrics
+	batchRequestCounter = prometheus.NewCounterVec(
+		prometheus.CounterOpts{
+			Subsystem: Namespace,
+			Name:      "batch_request_total",
+			Help:      "Total number of batch requests by status code",
+		}, []string{"channel", "channel_name", "tag", "base_url", "model", "group", "code"})
+	batchRequestDurationObsever = promauto.NewHistogramVec(
+		prometheus.HistogramOpts{
+			Subsystem: Namespace,
+			Name:      "batch_request_duration",
+			Help:      "Duration of batch request",
+			Buckets:   prometheus.ExponentialBuckets(1, 2, 12),
+		},
+		[]string{"channel", "channel_name", "tag", "base_url", "model", "group", "code"},
 	)
 	// Token metrics
 	inputTokensCounter = prometheus.NewCounterVec(
@@ -95,79 +116,100 @@ var (
 			Subsystem: Namespace,
 			Name:      "input_tokens_total",
 			Help:      "Total number of input tokens processed",
-		}, []string{"channel", "model", "group", "user_id", "user_name", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "user_id", "user_name", "token_name"})
 
 	outputTokensCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "output_tokens_total",
 			Help:      "Total number of output tokens generated",
-		}, []string{"channel", "model", "group", "user_id", "user_name", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "user_id", "user_name", "token_name"})
 
 	cacheHitTokensCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "cache_hit_tokens_total",
 			Help:      "Total number of tokens served from cache",
-		}, []string{"channel", "model", "group", "user_id", "user_name", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "user_id", "user_name", "token_name"})
 
 	inferenceTokensCounter = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: Namespace,
 			Name:      "inference_tokens_total",
 			Help:      "Total number of tokens processed during inference",
-		}, []string{"channel", "model", "group", "user_id", "user_name", "token_name"})
+		}, []string{"channel", "channel_name", "model", "group", "user_id", "user_name", "token_name"})
+
+	errorLogCounter = prometheus.NewCounterVec(
+		prometheus.CounterOpts{
+			Subsystem: Namespace,
+			Name:      "error_log_total",
+			Help:      "Total number of error logs",
+		}, []string{"channel", "channel_name", "error_code", "error_type", "model", "group", "token_name"})
 )
 
-func IncrementRelayRequestTotalCounter(channel, tag, baseURL, model, group string, add float64) {
-	relayRequestTotalCounter.WithLabelValues(channel, tag, baseURL, model, group).Add(add)
+func IncrementRelayRequestTotalCounter(channel, channelName, tag, baseURL, model, group string, add float64) {
+	relayRequestTotalCounter.WithLabelValues(channel, channelName, tag, baseURL, model, group).Add(add)
+}
+
+func IncrementRelayRequestSuccessCounter(channel, channelName, tag, baseURL, model, group, statusCode string, add float64) {
+	relayRequestSuccessCounter.WithLabelValues(channel, channelName, tag, baseURL, model, group, statusCode).Add(add)
 }
 
-func IncrementRelayRequestSuccessCounter(channel, tag, baseURL, model, group string, add float64) {
-	relayRequestSuccessCounter.WithLabelValues(channel, tag, baseURL, model, group).Add(add)
+func IncrementRelayRequestFailedCounter(channel, channelName, tag, baseURL, model, group, code string, add float64) {
+	relayRequestFailedCounter.WithLabelValues(channel, channelName, tag, baseURL, model, group, code).Add(add)
 }
 
-func IncrementRelayRequestFailedCounter(channel, tag, baseURL, model, group, code string, add float64) {
-	relayRequestFailedCounter.WithLabelValues(channel, tag, baseURL, model, group, code).Add(add)
+func IncrementRelayRetryCounter(channel, channelName, tag, baseURL, model, group string, add float64) {
+	relayRequestRetryCounter.WithLabelValues(channel, channelName, tag, baseURL, model, group).Add(add)
 }
 
-func IncrementRelayRetryCounter(channel, tag, baseURL, model, group string, add float64) {
-	relayRequestRetryCounter.WithLabelValues(channel, tag, baseURL, model, group).Add(add)
+func ObserveRelayRequestDuration(channel, channelName, tag, baseURL, model, group string, duration float64) {
+	relayRequestDurationObsever.WithLabelValues(channel, channelName, tag, baseURL, model, group).Observe(duration)
 }
 
-func ObserveRelayRequestDuration(channel, tag, baseURL, model, group string, duration float64) {
-	relayRequestDurationObsever.WithLabelValues(channel, tag, baseURL, model, group).Observe(duration)
+func IncrementRelayRequestE2ETotalCounter(channel, channelName, model, group, tokenKey, tokenName string, add float64) {
+	relayRequestE2ETotalCounter.WithLabelValues(channel, channelName, model, group, tokenKey, tokenName).Add(add)
 }
 
-func IncrementRelayRequestE2ETotalCounter(channel, model, group, tokenKey, tokenName string, add float64) {
-	relayRequestE2ETotalCounter.WithLabelValues(channel, model, group, tokenKey, tokenName).Add(add)
+func IncrementRelayRequestE2ESuccessCounter(channel, channelName, model, group, tokenKey, tokenName string, add float64) {
+	relayRequestE2ESuccessCounter.WithLabelValues(channel, channelName, model, group, tokenKey, tokenName).Add(add)
 }
 
-func IncrementRelayRequestE2ESuccessCounter(channel, model, group, tokenKey, tokenName string, add float64) {
-	relayRequestE2ESuccessCounter.WithLabelValues(channel, model, group, tokenKey, tokenName).Add(add)
+func IncrementRelayRequestE2EFailedCounter(channel, channelName, model, group, code, tokenKey, tokenName string, add float64) {
+	relayRequestE2EFailedCounter.WithLabelValues(channel, channelName, model, group, code, tokenKey, tokenName).Add(add)
 }
 
-func IncrementRelayRequestE2EFailedCounter(channel, model, group, code, tokenKey, tokenName string, add float64) {
-	relayRequestE2EFailedCounter.WithLabelValues(channel, model, group, code, tokenKey, tokenName).Add(add)
+func ObserveRelayRequestE2EDuration(channel, channelName, model, group, tokenKey, tokenName string, duration float64) {
+	relayRequestE2EDurationObsever.WithLabelValues(channel, channelName, model, group, tokenKey, tokenName).Observe(duration)
 }
 
-func ObserveRelayRequestE2EDuration(channel, model, group, tokenKey, tokenName string, duration float64) {
-	relayRequestE2EDurationObsever.WithLabelValues(channel, model, group, tokenKey, tokenName).Observe(duration)
+// Batch request metrics functions
+func IncrementBatchRequestCounter(channel, channelName, tag, baseURL, model, group, code string, add float64) {
+	batchRequestCounter.WithLabelValues(channel, channelName, tag, baseURL, model, group, code).Add(add)
+}
+
+func ObserveBatchRequestDuration(channel, channelName, tag, baseURL, model, group, code string, duration float64) {
+	batchRequestDurationObsever.WithLabelValues(channel, channelName, tag, baseURL, model, group, code).Observe(duration)
 }
 
 // Token metrics functions
-func IncrementInputTokens(channel, model, group, userId, userName, tokenName string, add float64) {
-	inputTokensCounter.WithLabelValues(channel, model, group, userId, userName, tokenName).Add(add)
+func IncrementInputTokens(channel, channelName, model, group, userId, userName, tokenName string, add float64) {
+	inputTokensCounter.WithLabelValues(channel, channelName, model, group, userId, userName, tokenName).Add(add)
+}
+
+func IncrementOutputTokens(channel, channelName, model, group, userId, userName, tokenName string, add float64) {
+	outputTokensCounter.WithLabelValues(channel, channelName, model, group, userId, userName, tokenName).Add(add)
 }
 
-func IncrementOutputTokens(channel, model, group, userId, userName, tokenName string, add float64) {
-	outputTokensCounter.WithLabelValues(channel, model, group, userId, userName, tokenName).Add(add)
+func IncrementCacheHitTokens(channel, channelName, model, group, userId, userName, tokenName string, add float64) {
+	cacheHitTokensCounter.WithLabelValues(channel, channelName, model, group, userId, userName, tokenName).Add(add)
 }
 
-func IncrementCacheHitTokens(channel, model, group, userId, userName, tokenName string, add float64) {
-	cacheHitTokensCounter.WithLabelValues(channel, model, group, userId, userName, tokenName).Add(add)
+func IncrementInferenceTokens(channel, channelName, model, group, userId, userName, tokenName string, add float64) {
+	inferenceTokensCounter.WithLabelValues(channel, channelName, model, group, userId, userName, tokenName).Add(add)
 }
 
-func IncrementInferenceTokens(channel, model, group, userId, userName, tokenName string, add float64) {
-	inferenceTokensCounter.WithLabelValues(channel, model, group, userId, userName, tokenName).Add(add)
+// Error log metrics function
+func IncrementErrorLog(channel, channelName, errorCode, errorType, model, group, tokenName string, add float64) {
+	errorLogCounter.WithLabelValues(channel, channelName, errorCode, errorType, model, group, tokenName).Add(add)
 }
diff --git a/middleware/distributor.go b/middleware/distributor.go
index 748dac718557..eeacaea46510 100644
--- a/middleware/distributor.go
+++ b/middleware/distributor.go
@@ -1,8 +1,10 @@
 package middleware
 
 import (
-	"errors"
+	"bytes"
+	"encoding/json"
 	"fmt"
+	"io"
 	"net/http"
 	"one-api/common"
 	"one-api/constant"
@@ -138,7 +140,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 			midjourneyRequest := dto.MidjourneyRequest{}
 			err = common.UnmarshalBodyReusable(c, &midjourneyRequest)
 			if err != nil {
-				return nil, false, err
+				return nil, false, fmt.Errorf("无效的请求, %s", err.Error())
 			}
 			midjourneyModel, mjErr, success := service.GetMjRequestModel(relayMode, &midjourneyRequest)
 			if mjErr != nil {
@@ -167,11 +169,40 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 		c.Set("platform", string(constant.TaskPlatformSuno))
 		c.Set("relay_mode", relayMode)
 	} else if !strings.HasPrefix(c.Request.URL.Path, "/v1/audio/transcriptions") {
-		err = common.UnmarshalBodyReusable(c, &modelRequest)
-	}
-	if err != nil {
-		return nil, false, errors.New("无效的请求, " + err.Error())
+		// 检查请求体是否为空
+		body, err := io.ReadAll(c.Request.Body)
+		if err != nil {
+			return nil, false, fmt.Errorf("无效的请求, 读取请求体失败: %s", err.Error())
+		}
+		// 重置请求体
+		c.Request.Body = io.NopCloser(bytes.NewBuffer(body))
+
+		// 如果请求体为空，根据路径设置默认模型
+		if len(body) == 0 {
+			if strings.HasPrefix(c.Request.URL.Path, "/v1/moderations") {
+				modelRequest.Model = "text-moderation-stable"
+			} else if strings.HasSuffix(c.Request.URL.Path, "embeddings") {
+				modelRequest.Model = c.Param("model")
+			} else if strings.HasPrefix(c.Request.URL.Path, "/v1/images/generations") {
+				modelRequest.Model = "dall-e"
+			} else if strings.HasPrefix(c.Request.URL.Path, "/v1/audio/speech") {
+				modelRequest.Model = "tts-1"
+			} else if strings.HasPrefix(c.Request.URL.Path, "/v1/audio/translations") {
+				modelRequest.Model = "whisper-1"
+			} else if strings.HasPrefix(c.Request.URL.Path, "/v1/audio/transcriptions") {
+				modelRequest.Model = "whisper-1"
+			} else if strings.HasPrefix(c.Request.URL.Path, "/v1/realtime") {
+				modelRequest.Model = c.Query("model")
+			}
+		} else {
+			// 请求体不为空，尝试解析 JSON
+			err = json.Unmarshal(body, &modelRequest)
+			if err != nil {
+				return nil, false, fmt.Errorf("无效的请求, JSON 解析失败: %s", err.Error())
+			}
+		}
 	}
+
 	if strings.HasPrefix(c.Request.URL.Path, "/v1/realtime") {
 		//wss://api.openai.com/v1/realtime?model=gpt-4o-realtime-preview-2024-10-01
 		modelRequest.Model = c.Query("model")
@@ -225,6 +256,7 @@ func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, mode
 	c.Set("status_code_mapping", channel.GetStatusCodeMapping())
 	c.Request.Header.Set("Authorization", fmt.Sprintf("Bearer %s", channel.Key))
 	c.Set("base_url", channel.GetBaseURL())
+	c.Set("endpoint", channel.GetEndpoint())
 	// TODO: api_version统一
 	switch channel.Type {
 	case common.ChannelTypeAzure:
diff --git a/middleware/request-id.go b/middleware/request-id.go
index 816396530cca..20fd086fe3de 100644
--- a/middleware/request-id.go
+++ b/middleware/request-id.go
@@ -2,8 +2,11 @@ package middleware
 
 import (
 	"crypto/md5"
+	"crypto/rand"
+	"encoding/hex"
 	"one-api/common"
 	"strconv"
+	"time"
 
 	"github.com/gin-gonic/gin"
 )
@@ -12,35 +15,28 @@ func RequestId() func(c *gin.Context) {
 	return func(c *gin.Context) {
 		id := c.GetHeader(common.RequestIdKey)
 		if id == "" {
-			id = common.GetTimeString() + common.GetRandomString(8)
+			// 使用更安全的request ID生成方法
+			id = GenerateUniqueRequestId()
 			c.Header(common.RequestIdKey, id)
 		}
 		c.Set(common.RequestIdKey, id)
+		c.Next()
+	}
+}
 
-		// 优先使用上游传递的哈希值
-		originHashValue := c.GetHeader("X-Origin-Hash-Value")
+// GenerateUniqueRequestId 生成唯一的request ID
+func GenerateUniqueRequestId() string {
+	// 获取当前时间戳（纳秒精度）
+	timestamp := time.Now().UnixNano()
 
-		if originHashValue != "" {
-			// 将上游的哈希值转换为整数
-			value, err := strconv.Atoi(originHashValue)
-			if err == nil {
-				// 确保值在0-99范围内
-				value = value % 100
-				c.Set("hash_value", value)
-				c.Next()
-				return
-			}
-		}
+	// 生成16字节的随机数
+	randomBytes := make([]byte, 16)
+	rand.Read(randomBytes)
 
-		// 如果没有上游哈希值或转换失败，则计算新的哈希值
-		hash := md5.Sum([]byte(id))
-		hashValue := 0
-		for i := 0; i < len(hash); i++ {
-			hashValue = (hashValue*31 + int(hash[i])) % 100
-		}
-		// 将哈希值存入上下文
-		c.Set("hash_value", hashValue)
+	// 将时间戳和随机数组合
+	combined := strconv.FormatInt(timestamp, 10) + hex.EncodeToString(randomBytes)
 
-		c.Next()
-	}
+	// 使用MD5生成最终的request ID（32位十六进制字符串）
+	hash := md5.Sum([]byte(combined))
+	return hex.EncodeToString(hash[:])
 }
diff --git a/model/channel.go b/model/channel.go
index e0ee2d2741d7..7b1d7940501b 100644
--- a/model/channel.go
+++ b/model/channel.go
@@ -22,6 +22,7 @@ type Channel struct {
 	TestTime           int64   `json:"test_time" gorm:"bigint"`
 	ResponseTime       int     `json:"response_time"` // in milliseconds
 	BaseURL            *string `json:"base_url" gorm:"column:base_url;default:''"`
+	Endpoint           *string `json:"endpoint" gorm:"column:endpoint;default:''"`
 	Other              string  `json:"other"`
 	Balance            float64 `json:"balance"` // in USD
 	BalanceUpdatedTime int64   `json:"balance_updated_time" gorm:"bigint"`
@@ -223,6 +224,13 @@ func (channel *Channel) GetBaseURL() string {
 	return *channel.BaseURL
 }
 
+func (channel *Channel) GetEndpoint() string {
+	if channel.Endpoint == nil {
+		return ""
+	}
+	return *channel.Endpoint
+}
+
 func (channel *Channel) GetModelMapping() string {
 	if channel.ModelMapping == nil {
 		return ""
diff --git a/relay/channel/api_request.go b/relay/channel/api_request.go
index 4e2c016ac054..248b88560c35 100644
--- a/relay/channel/api_request.go
+++ b/relay/channel/api_request.go
@@ -43,6 +43,14 @@ func SetupApiRequestHeader(info *common.RelayInfo, c *gin.Context, req *http.Hea
 	for key, value := range info.Headers {
 		req.Set(key, value)
 	}
+
+	// 添加指定的header - 从原始请求中获取
+	if retryRequestId := c.GetHeader("retry_request_id"); retryRequestId != "" {
+		req.Set("retry_request_id", retryRequestId)
+	}
+	if retry := c.GetHeader("retry"); retry != "" {
+		req.Set("retry", retry)
+	}
 }
 
 func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody io.Reader) (*http.Response, error) {
@@ -167,7 +175,14 @@ func doRequest(c *gin.Context, req *http.Request, info *common.RelayInfo) (*http
 	req.Header.Set("X-Origin-User-ID", strconv.Itoa(info.UserId))
 	req.Header.Set("X-Origin-Channel-ID", strconv.Itoa(info.ChannelId))
 	req.Header.Set("X-Retry-Count", strconv.Itoa(info.RetryCount))
-	req.Header.Set("X-Origin-Hash-Value", strconv.Itoa(c.GetInt("hash_value")))
+
+	// 添加指定的header - 从原始请求中获取
+	if retryRequestId := c.GetHeader("retry_request_id"); retryRequestId != "" {
+		req.Header.Set("retry_request_id", retryRequestId)
+	}
+	if retry := c.GetHeader("retry"); retry != "" {
+		req.Header.Set("retry", retry)
+	}
 
 	// 打印请求头
 	requestId := c.GetString(onecommon.RequestIdKey)
diff --git a/relay/channel/aws/adaptor.go b/relay/channel/aws/adaptor.go
index 28afe4d44639..602464584bf8 100644
--- a/relay/channel/aws/adaptor.go
+++ b/relay/channel/aws/adaptor.go
@@ -2,13 +2,14 @@ package aws
 
 import (
 	"errors"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
 	"one-api/dto"
 	"one-api/relay/channel/claude"
 	relaycommon "one-api/relay/common"
 	"one-api/setting/model_setting"
+
+	"github.com/gin-gonic/gin"
 )
 
 const (
@@ -40,6 +41,15 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
 	model_setting.GetClaudeSettings().WriteHeaders(info.OriginModelName, req)
+
+	// 添加指定的header - 从原始请求中获取
+	if retryRequestId := c.GetHeader("retry_request_id"); retryRequestId != "" {
+		req.Set("retry_request_id", retryRequestId)
+	}
+	if retry := c.GetHeader("retry"); retry != "" {
+		req.Set("retry", retry)
+	}
+
 	return nil
 }
 
diff --git a/relay/channel/gemini/relay-gemini.go b/relay/channel/gemini/relay-gemini.go
index 6ba1a00b5da3..82ac7f120b03 100644
--- a/relay/channel/gemini/relay-gemini.go
+++ b/relay/channel/gemini/relay-gemini.go
@@ -212,6 +212,18 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 						},
 					})
 				}
+			} else if part.Type == dto.ContentTypeInputAudio {
+				// 处理音频内容
+				audioData := part.InputAudio.(dto.MessageInputAudio)
+				// 添加调试日志
+				common.SysLog(fmt.Sprintf("Processing audio data: format=%s, data length=%d", audioData.Format, len(audioData.Data)))
+				// 将音频数据转换为Gemini的InlineData格式
+				parts = append(parts, GeminiPart{
+					InlineData: &GeminiInlineData{
+						MimeType: audioData.Format,
+						Data:     audioData.Data,
+					},
+				})
 			} else if part.Type == dto.ContentTypeYoutube {
 				parts = append(parts, GeminiPart{
 					FileData: &GeminiFileData{
diff --git a/relay/channel/task/suno/adaptor.go b/relay/channel/task/suno/adaptor.go
index 03d60516f06a..1eaa31fd2c82 100644
--- a/relay/channel/task/suno/adaptor.go
+++ b/relay/channel/task/suno/adaptor.go
@@ -5,7 +5,6 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
 	"one-api/common"
@@ -16,6 +15,8 @@ import (
 	"one-api/service"
 	"strings"
 	"time"
+
+	"github.com/gin-gonic/gin"
 )
 
 type TaskAdaptor struct {
@@ -64,6 +65,15 @@ func (a *TaskAdaptor) BuildRequestHeader(c *gin.Context, req *http.Request, info
 	req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
 	req.Header.Set("Accept", c.Request.Header.Get("Accept"))
 	req.Header.Set("Authorization", "Bearer "+info.ApiKey)
+
+	// 添加指定的header - 从原始请求中获取
+	if retryRequestId := c.GetHeader("retry_request_id"); retryRequestId != "" {
+		req.Header.Set("retry_request_id", retryRequestId)
+	}
+	if retry := c.GetHeader("retry"); retry != "" {
+		req.Header.Set("retry", retry)
+	}
+
 	return nil
 }
 
diff --git a/relay/channel/volcengine/adaptor.go b/relay/channel/volcengine/adaptor.go
index f6b3d33fc4d9..55098993a284 100644
--- a/relay/channel/volcengine/adaptor.go
+++ b/relay/channel/volcengine/adaptor.go
@@ -76,6 +76,10 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 }
 
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
+	// 检查模型名称是否包含 "batch"，如果是则使用批量接口
+	if strings.Contains(strings.ToLower(info.OriginModelName), "batch") {
+		return DoBatchChatRequest(c, info, requestBody)
+	}
 	return channel.DoApiRequest(a, c, info, requestBody)
 }
 
diff --git a/relay/channel/volcengine/batchchat.go b/relay/channel/volcengine/batchchat.go
new file mode 100644
index 000000000000..994e40a8cea2
--- /dev/null
+++ b/relay/channel/volcengine/batchchat.go
@@ -0,0 +1,989 @@
+package volcengine
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"math/rand"
+	"net/http"
+	"one-api/common"
+	"one-api/dto"
+	"one-api/metrics"
+	relaycommon "one-api/relay/common"
+	"os"
+	"strconv"
+	"strings"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	"github.com/gin-gonic/gin"
+	"github.com/volcengine/volcengine-go-sdk/service/arkruntime"
+	"github.com/volcengine/volcengine-go-sdk/service/arkruntime/model"
+)
+
+// 全局配置变量
+const (
+	MaxParallelRequests = 2000
+
+	// 限速器默认大小为 MaxParallelRequests 的 3 倍
+	RateLimiterSize = MaxParallelRequests * 3
+	// 异步调用超时时间配置
+	MinAsyncTimeout = 30 * time.Second
+	MaxAsyncTimeout = 60 * time.Second
+	// 限流等待时间配置 - 用于等待可用请求槽位的超时时间
+	MinRateLimitWaitTime = 100 * time.Millisecond
+	MaxRateLimitWaitTime = 1000 * time.Millisecond
+
+	// CreateBatchChatCompletion 调用的超时时间
+	BatchCompletionTimeout = 24 * time.Hour
+
+	// 子协程最大存活时间
+	SubGoroutineMaxLifetime = 24 * time.Hour
+
+	// 分布式锁过期时间
+	DistributedLockExpiration = 24 * time.Hour
+)
+
+// 异步调用的超时时间 - 可通过环境变量VOLCENGINE_ASYNC_CALL_TIMEOUT配置，默认30秒
+var AsyncCallTimeout = time.Duration(common.GetEnvOrDefault("VOLCENGINE_ASYNC_CALL_TIMEOUT", 30)) * time.Second
+
+// 客户端缓存
+var (
+	clientCache = make(map[string]*arkruntime.Client)
+	clientMutex sync.RWMutex
+)
+
+// 请求计数器
+var (
+	requestCounter int64 = 0
+)
+
+// 限速器
+var (
+	rateLimiter = make(chan struct{}, RateLimiterSize)
+)
+
+// 建议重试时间相关变量
+var (
+	batchRequestAvgDuration      float64 = 30.0 // 默认30秒
+	batchRequestAvgDurationMutex sync.RWMutex
+)
+
+// NewBatchClient 创建一个新的批量请求客户端实例
+func NewBatchClient(apiKey string) *arkruntime.Client {
+	return arkruntime.NewClientWithApiKey(
+		apiKey,
+		arkruntime.WithBatchMaxParallel(MaxParallelRequests), // 使用全局变量设置发起请求的最大并发数量
+	)
+}
+
+// GetBatchClient 根据 channel ID 获取或创建客户端实例
+func GetBatchClient(channelId string, apiKey string) *arkruntime.Client {
+	clientMutex.RLock()
+	if client, exists := clientCache[channelId]; exists {
+		clientMutex.RUnlock()
+		return client
+	}
+	clientMutex.RUnlock()
+
+	// 如果缓存中没有，创建新的客户端
+	clientMutex.Lock()
+	defer clientMutex.Unlock()
+
+	// 双重检查，防止并发创建
+	if client, exists := clientCache[channelId]; exists {
+		return client
+	}
+
+	client := NewBatchClient(apiKey)
+	clientCache[channelId] = client
+	return client
+}
+
+// acquireRequestSlot 获取请求槽位，如果达到上限则返回错误
+func acquireRequestSlot() error {
+	// 获取当前计数器值
+	currentCount := atomic.LoadInt64(&requestCounter)
+
+	// 如果已达到上限，返回错误
+	if currentCount >= int64(RateLimiterSize) {
+		return fmt.Errorf("request limit reached, please retry later")
+	}
+
+	// 尝试获取限速器槽位
+	select {
+	case rateLimiter <- struct{}{}:
+		// 成功获取槽位，增加计数器
+		atomic.AddInt64(&requestCounter, 1)
+		return nil
+	default:
+		// 限速器已满，返回错误
+		return fmt.Errorf("request limit reached, please retry later")
+	}
+}
+
+// releaseRequestSlot 释放请求槽位
+func releaseRequestSlot() {
+	// 减少计数器
+	atomic.AddInt64(&requestCounter, -1)
+	// 释放限速器槽位
+	select {
+	case <-rateLimiter:
+	default:
+		// 如果限速器为空，忽略
+	}
+}
+
+// waitForAvailableSlot 等待可用的请求槽位
+func waitForAvailableSlot(ctx context.Context) error {
+	ticker := time.NewTicker(100 * time.Millisecond) // 每100ms检查一次
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-ctx.Done():
+			return ctx.Err()
+		case <-ticker.C:
+			if err := acquireRequestSlot(); err == nil {
+				return nil
+			}
+		}
+	}
+}
+
+// 从 context 获取异步调用超时时间，如果没有设置则使用随机计算的时间
+func getAsyncCallTimeout(ctx context.Context) time.Duration {
+	if timeout, ok := ctx.Value("async_call_timeout").(time.Duration); ok && timeout > 0 {
+		return timeout
+	}
+	// 使用随机等待时间作为异步调用超时时间
+	return calculateRandomWaitTime()
+}
+
+// 从 context 获取批量推理超时时间，如果没有设置则使用默认值
+func getBatchCompletionTimeout(ctx context.Context) time.Duration {
+	if timeout, ok := ctx.Value("batch_completion_timeout").(time.Duration); ok && timeout > 0 {
+		return timeout
+	}
+	return BatchCompletionTimeout
+}
+
+// GetCurrentTimeouts 获取当前 context 中的超时配置
+func GetCurrentTimeouts(ctx context.Context) map[string]time.Duration {
+	return map[string]time.Duration{
+		"async_call_timeout":       getAsyncCallTimeout(ctx),
+		"batch_completion_timeout": getBatchCompletionTimeout(ctx),
+		"min_async_timeout":        MinAsyncTimeout,
+		"max_async_timeout":        MaxAsyncTimeout,
+		"min_rate_limit_wait":      MinRateLimitWaitTime,
+		"max_rate_limit_wait":      MaxRateLimitWaitTime,
+		"default_async_timeout":    AsyncCallTimeout,
+		"default_batch_timeout":    BatchCompletionTimeout,
+	}
+}
+
+// getRequestID 从gin context获取请求ID，使用middleware设置的ID
+func getRequestID(c *gin.Context) string {
+	// 优先使用retry_request_id
+	requestID := c.GetHeader("retry_request_id")
+	if requestID == "" {
+		// 如果没有retry_request_id，则使用正常的requestID
+		requestID = c.GetHeader(common.RequestIdKey)
+	}
+	return requestID
+}
+
+// isRetryRequest 检查是否为重试请求
+func isRetryRequest(c *gin.Context) bool {
+	retryHeader := c.GetHeader("retry")
+	return retryHeader == "true"
+}
+
+func DoBatchChatRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (*http.Response, error) {
+	// 获取请求ID
+	requestID := getRequestID(c)
+	c.Set("Retry_request_id", requestID)
+
+	// 尝试获取分布式锁，避免重复执行
+	lockKey := requestID + "_lock"
+	lockAcquired, err := TryAcquireLock(lockKey, DistributedLockExpiration)
+	if err != nil {
+		return nil, fmt.Errorf("failed to acquire lock: %w", err)
+	}
+
+	if !lockAcquired {
+		// 返回内部错误响应
+		errorResponse := gin.H{
+			"error": gin.H{
+				"message":    fmt.Sprintf("request %s is already being processed, another request is in progress", requestID),
+				"type":       "internal_error",
+				"code":       "lock_acquisition_failed",
+				"request_id": requestID,
+			},
+		}
+		errorJson, _ := json.Marshal(errorResponse)
+		response := &http.Response{
+			StatusCode: dto.StatusRequestConflict,
+			Body:       io.NopCloser(strings.NewReader(string(errorJson))),
+			Header:     make(http.Header),
+		}
+		response.Header.Set("Content-Type", "application/json")
+		return response, nil
+	}
+
+	// 确保在函数结束时释放锁
+	defer func() {
+		if releaseErr := ReleaseLock(lockKey); releaseErr != nil {
+			common.LogError(c, fmt.Sprintf("Failed to release lock for request %s: %v", requestID, releaseErr))
+		}
+	}()
+
+	// 检查是否为重试请求
+	if isRetryRequest(c) {
+		// 从Redis获取结果，使用当前的requestID（可能是retry_request_id）
+		resultData, err := GetBatchResultFromRedis(requestID)
+		if err == nil {
+			// 检查Result是否为空且状态为pending，这种情况说明第一次请求可能超时了
+			if resultData.Result == "" && resultData.Status == "pending" {
+				common.LogInfo(c.Request.Context(), fmt.Sprintf("Found pending request for %s, returning retry response", requestID))
+
+				// 返回重试提示
+				errorResponse := gin.H{
+					"error": gin.H{
+						"message":    "Request is still being processed, please retry later",
+						"type":       "request_in_progress",
+						"code":       "request_still_processing",
+						"request_id": requestID,
+					},
+				}
+				errorJson, _ := json.Marshal(errorResponse)
+
+				response := &http.Response{
+					StatusCode: dto.StatusNewAPIBatchSubmitted, // 203 - 批量请求已提交，需要重试获取结果
+					Body:       io.NopCloser(strings.NewReader(string(errorJson))),
+					Header:     make(http.Header),
+				}
+				response.Header.Set("Content-Type", "application/json")
+				response.Header.Set("Retry_request_id", requestID)
+				// 添加建议重试时间header
+				avgDuration := GetBatchRequestAverageDuration()
+				response.Header.Set("X-Suggested-Retry-After", fmt.Sprintf("%.0f", avgDuration))
+				c.Writer.Header().Set("Retry_request_id", requestID)
+				c.Writer.Header().Set("X-Suggested-Retry-After", fmt.Sprintf("%.0f", avgDuration))
+				return response, nil
+			} else if resultData.Result != "" {
+				// 只有在Result不为空时才处理缓存结果
+				// 删除Redis中的key
+				err = DeleteBatchResultFromRedis(requestID)
+				if err != nil {
+					common.LogError(c, err.Error())
+				}
+
+				// 先用火山引擎格式解析，再转换为SimpleResponse
+				openaiResponse, err := convertVolcEngineResponseToOpenAI([]byte(resultData.Result))
+				if err != nil {
+					return nil, fmt.Errorf("failed to convert cached response format: %w", err)
+				}
+
+				// 将转换后的结果序列化为JSON
+				openaiResponseJson, err := json.Marshal(openaiResponse)
+				if err != nil {
+					return nil, fmt.Errorf("failed to marshal cached OpenAI response: %w", err)
+				}
+
+				// 找到结果，返回并删除key
+				response := &http.Response{
+					StatusCode: http.StatusOK,
+					Body:       io.NopCloser(strings.NewReader(string(openaiResponseJson))),
+					Header:     make(http.Header),
+				}
+				response.Header.Set("Content-Type", "application/json")
+				return response, nil
+			} else {
+				// Result为空但状态不是pending，说明是错误状态
+				common.LogInfo(c.Request.Context(), fmt.Sprintf("Found error status for request %s, continuing with new request", requestID))
+				// 删除Redis中的key，继续执行新建流程
+				err = DeleteBatchResultFromRedis(requestID)
+				if err != nil {
+					common.LogError(c, err.Error())
+				}
+			}
+		}
+		// 如果Redis中没有找到结果，继续执行新建流程
+	}
+
+	// 尝试获取请求槽位，如果无法立即获取则等待
+	if err := acquireRequestSlot(); err != nil {
+		// 如果无法立即获取槽位，等待可用槽位
+		// 随机计算等待时间：在100ms-1000ms之间随机选择
+		waitTime := calculateRateLimitWaitTime()
+
+		ctx, cancel := context.WithTimeout(c.Request.Context(), waitTime)
+		defer cancel()
+
+		if waitErr := waitForAvailableSlot(ctx); waitErr != nil {
+			// 等待超时，返回自定义限流错误
+			errorResponse := gin.H{
+				"error": gin.H{
+					"message": "Request limit reached, please retry later",
+					"type":    "new_api_batch_rate_limit_exceeded",
+					"code":    "new_api_batch_rate_limit_exceeded",
+				},
+			}
+			errorJson, _ := json.Marshal(errorResponse)
+			response := &http.Response{
+				StatusCode: dto.StatusNewAPIBatchRateLimitExceeded,
+				Body:       io.NopCloser(strings.NewReader(string(errorJson))),
+				Header:     make(http.Header),
+			}
+			response.Header.Set("Content-Type", "application/json")
+			return response, nil
+		}
+	}
+
+	// 确保在函数结束时释放槽位
+	defer releaseRequestSlot()
+
+	// 解析请求体
+	var request dto.GeneralOpenAIRequest
+	if err := json.NewDecoder(requestBody).Decode(&request); err != nil {
+		return nil, fmt.Errorf("failed to decode request body: %w", err)
+	}
+	// 转换为豆包批量请求格式
+	batchRequest, err := convertToBatchRequest(&request, info.Endpoint)
+	if err != nil {
+		return nil, fmt.Errorf("failed to convert request: %w", err)
+	}
+
+	// 检查是否有未支持的参数
+	checkUnsupportedParameters(&request)
+
+	// 使用 channel ID 获取或创建客户端实例
+	client := GetBatchClient(fmt.Sprintf("%d", info.ChannelId), info.ApiKey)
+
+	// 创建带超时的 context，用于异步调用的整体超时
+	timeoutDuration := getAsyncCallTimeout(c.Request.Context())
+	asyncCtx, asyncCancel := context.WithTimeout(c.Request.Context(), timeoutDuration)
+	defer asyncCancel()
+
+	// 创建通道用于接收异步结果
+	resultChan := make(chan interface{}, 1)
+	errChan := make(chan error, 1)
+
+	// 异步发起批量推理请求
+	go func() {
+		// 使用独立的context，不受外层asyncCtx影响
+		independentCtx := context.Background()
+
+		// 记录batch请求开始时间
+		batchStartTime := time.Now()
+
+		result, err := executeBatchRequestWithRedis(independentCtx, client, batchRequest, requestID)
+
+		// 记录batch请求指标
+		// 获取实际状态码
+		statusCode := "-1" // 默认状态码
+		if err != nil {
+			statusCode = "request_failed"
+		} else if result != nil {
+			// 尝试从结果中提取状态码
+			if resultMap, ok := result.(map[string]interface{}); ok {
+				if code, exists := resultMap["code"]; exists {
+					if codeStr, ok := code.(string); ok {
+						statusCode = codeStr
+					} else if codeInt, ok := code.(float64); ok {
+						statusCode = fmt.Sprintf("%.0f", codeInt)
+					}
+				}
+			}
+		}
+
+		// 记录指标
+		metrics.IncrementBatchRequestCounter(
+			info.ChannelTag,
+			info.ChannelName,
+			info.ChannelTag,
+			info.BaseUrl,
+			info.UpstreamModelName,
+			info.Group,
+			statusCode,
+			1,
+		)
+		metrics.ObserveBatchRequestDuration(
+			info.ChannelTag,
+			info.ChannelName,
+			info.ChannelTag,
+			info.BaseUrl,
+			info.UpstreamModelName,
+			info.Group,
+			statusCode,
+			time.Since(batchStartTime).Seconds(),
+		)
+
+		if err != nil {
+			common.LogError(c, fmt.Sprintf("Async batch request failed for requestID %s: %v", requestID, err))
+			errChan <- err
+			return
+		}
+		resultChan <- result
+	}()
+
+	// 等待结果或超时
+	var result interface{}
+	select {
+	case result = <-resultChan:
+		// 成功获取结果
+		common.LogInfo(c, fmt.Sprintf("Received result for requestID %s", requestID))
+	case err := <-errChan:
+		// 发生错误
+		common.LogError(c, fmt.Sprintf("batch request failed: %v", err))
+		return nil, fmt.Errorf("batch request failed: %w", err)
+	case <-asyncCtx.Done():
+		// 超时
+		if asyncCtx.Err() == context.DeadlineExceeded {
+			common.LogError(c, fmt.Sprintf("Async call timeout after %v for requestID %s", timeoutDuration, requestID))
+			c.Writer.Header().Set("Retry_request_id", requestID)
+
+			// 返回自定义状态码表示请求已提交
+			errorResponse := gin.H{
+				"error": gin.H{
+					"message":    fmt.Sprintf("Async call timeout after %v for requestID %s, please retry later to get the result", timeoutDuration, requestID),
+					"type":       "request_submitted",
+					"code":       "request_submitted",
+					"request_id": requestID,
+				},
+			}
+			errorJson, _ := json.Marshal(errorResponse)
+			response := &http.Response{
+				StatusCode: dto.StatusNewAPIBatchSubmitted, // 203 - 批量请求已提交，需要重试获取结果
+				Body:       io.NopCloser(strings.NewReader(string(errorJson))),
+				Header:     make(http.Header),
+			}
+			response.Header.Set("Content-Type", "application/json")
+			response.Header.Set("Retry_request_id", requestID)
+			// 添加建议重试时间header
+			avgDuration := GetBatchRequestAverageDuration()
+			response.Header.Set("X-Suggested-Retry-After", fmt.Sprintf("%.0f", avgDuration))
+			c.Writer.Header().Set("Retry_request_id", requestID)
+			c.Writer.Header().Set("X-Suggested-Retry-After", fmt.Sprintf("%.0f", avgDuration))
+			return response, nil
+		}
+		common.LogError(c, fmt.Sprintf("Async call cancelled for requestID %s: %w", requestID, asyncCtx.Err()))
+		c.Writer.Header().Set("Retry_request_id", requestID)
+		return nil, fmt.Errorf("async call cancelled: %w", asyncCtx.Err())
+	}
+
+	// 将结果转换为JSON
+	resultJson, err := json.Marshal(result)
+	if err != nil {
+		return nil, fmt.Errorf("failed to marshal result: %w", err)
+	}
+
+	// 将火山引擎的响应转换为标准的OpenAI格式
+	openaiResponse, err := convertVolcEngineResponseToOpenAI(resultJson)
+	if err != nil {
+		return nil, fmt.Errorf("failed to convert response format: %w", err)
+	}
+
+	// 将转换后的结果序列化为JSON
+	openaiResponseJson, err := json.Marshal(openaiResponse)
+	if err != nil {
+		return nil, fmt.Errorf("failed to marshal OpenAI response: %w", err)
+	}
+
+	// 创建HTTP响应
+	response := &http.Response{
+		StatusCode: http.StatusOK,
+		Body:       io.NopCloser(strings.NewReader(string(openaiResponseJson))),
+		Header:     make(http.Header),
+	}
+	response.Header.Set("Content-Type", "application/json")
+	response.Header.Set("Retry_request_id", requestID)
+	return response, nil
+}
+
+// calculateRandomWaitTime 在 MinAsyncTimeout-MaxAsyncTimeout 之间随机计算异步调用超时时间
+func calculateRandomWaitTime() time.Duration {
+	// 计算时间范围（毫秒）
+	timeRange := int(MaxAsyncTimeout.Milliseconds() - MinAsyncTimeout.Milliseconds())
+
+	// 如果时间范围为0或负数，直接返回MinAsyncTimeout
+	if timeRange <= 0 {
+		return MinAsyncTimeout
+	}
+
+	// 生成 MinAsyncTimeout-MaxAsyncTimeout 之间的随机等待时间（毫秒）
+	waitMilliseconds := rand.Intn(timeRange) + int(MinAsyncTimeout.Milliseconds())
+
+	return time.Duration(waitMilliseconds) * time.Millisecond
+}
+
+// calculateRateLimitWaitTime 在 MinRateLimitWaitTime-MaxRateLimitWaitTime 之间随机计算限流等待时间
+func calculateRateLimitWaitTime() time.Duration {
+	// 计算时间范围（毫秒）
+	timeRange := int(MaxRateLimitWaitTime.Milliseconds() - MinRateLimitWaitTime.Milliseconds())
+
+	// 如果时间范围为0或负数，直接返回MinRateLimitWaitTime
+	if timeRange <= 0 {
+		return MinRateLimitWaitTime
+	}
+
+	// 生成 MinRateLimitWaitTime-MaxRateLimitWaitTime 之间的随机等待时间（毫秒）
+	waitMilliseconds := rand.Intn(timeRange) + int(MinRateLimitWaitTime.Milliseconds())
+
+	return time.Duration(waitMilliseconds) * time.Millisecond
+}
+
+// GetCurrentRequestCount 获取当前请求计数（用于监控）
+func GetCurrentRequestCount() int64 {
+	return atomic.LoadInt64(&requestCounter)
+}
+
+// GetRateLimiterStatus 获取限速器状态（用于监控）
+func GetRateLimiterStatus() (current, capacity int) {
+	return len(rateLimiter), cap(rateLimiter)
+}
+
+func MustMarshalJson(v interface{}) string {
+	s, _ := json.Marshal(v)
+	return string(s)
+}
+
+// checkUnsupportedParameters 检查请求中是否有未支持的参数
+func checkUnsupportedParameters(request *dto.GeneralOpenAIRequest) {
+	var unsupportedParams []string
+
+	// 检查 tool_choice 参数（豆包批量推理可能不支持）
+	if request.ToolChoice != nil {
+		unsupportedParams = append(unsupportedParams, "tool_choice")
+	}
+
+	// 检查 n 参数（批量推理不支持，批量推理本身就是多个请求）
+	if request.N > 0 {
+		unsupportedParams = append(unsupportedParams, "n")
+	}
+
+	// 检查 stream 参数（批量推理不支持流式输出）
+	if request.Stream {
+		unsupportedParams = append(unsupportedParams, "stream")
+	}
+
+	// 检查 user 参数（豆包SDK支持，但批量推理可能不支持）
+	if request.User != "" {
+		unsupportedParams = append(unsupportedParams, "user")
+	}
+
+	// 检查 seed 参数（豆包可能不支持）
+	if request.Seed != 0 {
+		unsupportedParams = append(unsupportedParams, "seed")
+	}
+
+	// 检查 response_format 参数（豆包SDK支持，但批量推理可能不支持）
+	if request.ResponseFormat != nil {
+		unsupportedParams = append(unsupportedParams, "response_format")
+	}
+
+	// 检查 stream_options 参数（豆包SDK支持，但批量推理可能不支持）
+	if request.StreamOptions != nil {
+		unsupportedParams = append(unsupportedParams, "stream_options")
+	}
+
+	// 检查 functions 参数（已废弃，使用tools替代）
+	if request.Functions != nil {
+		unsupportedParams = append(unsupportedParams, "functions")
+	}
+
+	// 检查其他可能不支持的参数
+	if request.Prompt != nil {
+		unsupportedParams = append(unsupportedParams, "prompt")
+	}
+
+	if request.Prefix != nil {
+		unsupportedParams = append(unsupportedParams, "prefix")
+	}
+
+	if request.Suffix != nil {
+		unsupportedParams = append(unsupportedParams, "suffix")
+	}
+
+	if request.Input != nil {
+		unsupportedParams = append(unsupportedParams, "input")
+	}
+
+	if request.Instruction != "" {
+		unsupportedParams = append(unsupportedParams, "instruction")
+	}
+
+	if request.Size != "" {
+		unsupportedParams = append(unsupportedParams, "size")
+	}
+
+	if request.EncodingFormat != nil {
+		unsupportedParams = append(unsupportedParams, "encoding_format")
+	}
+
+	if request.Dimensions > 0 {
+		unsupportedParams = append(unsupportedParams, "dimensions")
+	}
+
+	if request.Modalities != nil {
+		unsupportedParams = append(unsupportedParams, "modalities")
+	}
+
+	if request.Audio != nil {
+		unsupportedParams = append(unsupportedParams, "audio")
+	}
+
+	if request.ExtraBody != nil {
+		unsupportedParams = append(unsupportedParams, "extra_body")
+	}
+
+	if request.Thinking != nil {
+		unsupportedParams = append(unsupportedParams, "thinking")
+	}
+
+	if request.ThinkingConfig != nil {
+		unsupportedParams = append(unsupportedParams, "thinking_config")
+	}
+
+	// 如果有未支持的参数，打印错误日志
+	if len(unsupportedParams) > 0 {
+		fmt.Printf("Error: Unsupported parameters detected in batch request: %v\n", unsupportedParams)
+		fmt.Printf("These parameters are not supported by VolcEngine batch inference API and will be ignored.\n")
+	}
+}
+
+// convertToBatchRequest 将 OpenAI 格式的请求转换为豆包批量请求格式
+func convertToBatchRequest(request *dto.GeneralOpenAIRequest, endpoint string) (*model.CreateChatCompletionRequest, error) {
+	// 获取消息内容
+	if len(request.Messages) == 0 {
+		return nil, fmt.Errorf("no messages found in request")
+	}
+
+	// 转换消息为豆包格式，支持多模态消息
+	messages := make([]*model.ChatCompletionMessage, 0, len(request.Messages))
+	for _, msg := range request.Messages {
+		// 解析消息内容
+		contentParts := msg.ParseContent()
+
+		var messageContent *model.ChatCompletionMessageContent
+
+		if len(contentParts) == 1 && contentParts[0].Type == "text" {
+			// 单文本消息
+			text := contentParts[0].Text
+			messageContent = &model.ChatCompletionMessageContent{
+				StringValue: &text,
+			}
+		} else {
+			// 多模态消息或复杂消息
+			parts := make([]*model.ChatCompletionMessageContentPart, 0, len(contentParts))
+			for _, part := range contentParts {
+				switch part.Type {
+				case "text":
+					parts = append(parts, &model.ChatCompletionMessageContentPart{
+						Type: "text",
+						Text: part.Text,
+					})
+				case "image_url":
+					if imageUrl, ok := part.ImageUrl.(dto.MessageImageUrl); ok {
+						detail := model.ImageURLDetail(imageUrl.Detail)
+						parts = append(parts, &model.ChatCompletionMessageContentPart{
+							Type: "image_url",
+							ImageURL: &model.ChatMessageImageURL{
+								URL:    imageUrl.Url,
+								Detail: detail,
+							},
+						})
+					}
+				}
+			}
+			if len(parts) > 0 {
+				messageContent = &model.ChatCompletionMessageContent{
+					ListValue: parts,
+				}
+			}
+		}
+
+		if messageContent != nil {
+			messages = append(messages, &model.ChatCompletionMessage{
+				Role:    msg.Role,
+				Content: messageContent,
+			})
+		}
+	}
+
+	// 转换为豆包批量请求格式
+	batchRequest := model.CreateChatCompletionRequest{
+		Model:    endpoint,
+		Messages: messages,
+	}
+
+	// 设置 max_tokens，只有当请求中包含时才设置
+	if request.MaxTokens > 0 {
+		maxTokens := int(request.MaxTokens)
+		batchRequest.MaxTokens = &maxTokens
+	}
+
+	// 设置 stop 参数，只有当请求中包含时才设置
+	if request.Stop != nil {
+		// 类型断言处理 stop 参数
+		switch stop := request.Stop.(type) {
+		case string:
+			batchRequest.Stop = []string{stop}
+		case []string:
+			batchRequest.Stop = stop
+		case []interface{}:
+			stopStrings := make([]string, 0, len(stop))
+			for _, s := range stop {
+				if str, ok := s.(string); ok {
+					stopStrings = append(stopStrings, str)
+				}
+			}
+			batchRequest.Stop = stopStrings
+		}
+	}
+
+	// 设置 frequency_penalty，只有当请求中包含且不为0时才设置
+	if request.FrequencyPenalty != 0 {
+		freqPenalty := float32(request.FrequencyPenalty)
+		batchRequest.FrequencyPenalty = &freqPenalty
+	}
+
+	// 设置 presence_penalty，只有当请求中包含且不为0时才设置
+	if request.PresencePenalty != 0 {
+		presPenalty := float32(request.PresencePenalty)
+		batchRequest.PresencePenalty = &presPenalty
+	}
+
+	// 设置 temperature，只有当请求中包含时才设置
+	if request.Temperature != nil {
+		temp := float32(*request.Temperature)
+		batchRequest.Temperature = &temp
+	}
+
+	// 设置 top_p，只有当请求中包含且不为0时才设置
+	if request.TopP != 0 {
+		topP := float32(request.TopP)
+		batchRequest.TopP = &topP
+	}
+
+	// 设置 logprobs，只有当请求中包含且为true时才设置
+	if request.LogProbs {
+		batchRequest.LogProbs = &request.LogProbs
+	}
+
+	// 设置 top_logprobs，只有当请求中包含且大于0时才设置
+	if request.TopLogProbs > 0 {
+		batchRequest.TopLogProbs = &request.TopLogProbs
+	}
+
+	// 设置 logit_bias，只有当请求中包含时才设置
+	if len(request.LogitBias) > 0 {
+		batchRequest.LogitBias = request.LogitBias
+	}
+
+	// 设置 tools，只有当请求中包含时才设置
+	if len(request.Tools) > 0 {
+		tools := make([]*model.Tool, 0, len(request.Tools))
+		for _, tool := range request.Tools {
+			// 只支持function类型的工具
+			if tool.Type == "function" {
+				chatTool := &model.Tool{
+					Type: model.ToolTypeFunction,
+					Function: &model.FunctionDefinition{
+						Name:        tool.Function.Name,
+						Description: tool.Function.Description,
+						Parameters:  tool.Function.Parameters,
+					},
+				}
+				tools = append(tools, chatTool)
+			}
+		}
+		if len(tools) > 0 {
+			batchRequest.Tools = tools
+		}
+	}
+
+	return &batchRequest, nil
+}
+
+// executeBatchRequestWithRedis 执行批量请求并保存结果到Redis
+func executeBatchRequestWithRedis(ctx context.Context, client *arkruntime.Client, batchRequest *model.CreateChatCompletionRequest, requestID string) (interface{}, error) {
+	// 在发起请求前先创建Redis key
+	if err := CreateBatchRequestKey(requestID); err != nil {
+		common.LogError(ctx, fmt.Sprintf("Failed to create initial Redis key for request %s: %v", requestID, err))
+		// 即使创建Redis key失败，也继续执行请求，只是不保存结果
+	}
+
+	// 为 CreateBatchChatCompletion 创建独立的超时 context，不复用传入的ctx
+	apiCtx, apiCancel := context.WithTimeout(context.Background(), getBatchCompletionTimeout(ctx))
+	defer apiCancel()
+
+	common.LogInfo(ctx, fmt.Sprintf("Batch chat completion request: %+v ", batchRequest))
+	result, err := client.CreateBatchChatCompletion(apiCtx, batchRequest)
+	if err != nil {
+		common.LogError(ctx, err.Error())
+		// 检查是否是超时错误
+		if apiCtx.Err() == context.DeadlineExceeded {
+			timeoutMsg := fmt.Sprintf("batch completion timeout after %v", getBatchCompletionTimeout(ctx))
+			if saveErr := SaveBatchErrorToRedis(requestID, timeoutMsg); saveErr != nil {
+				fmt.Printf("Failed to save timeout error to Redis for request %s: %v\n", requestID, saveErr)
+			}
+		} else {
+			if saveErr := SaveBatchErrorToRedis(requestID, err.Error()); saveErr != nil {
+				fmt.Printf("Failed to save error to Redis for request %s: %v\n", requestID, saveErr)
+			}
+		}
+		return nil, err
+	}
+	// 保存成功结果到Redis，子协程独立运行
+	common.LogInfo(ctx, fmt.Sprintf("Batch chat completion result: %+v", result))
+	if saveErr := SaveBatchResultToRedis(requestID, result, "completed"); saveErr != nil {
+		fmt.Printf("Failed to save result to Redis for request %s: %v\n", requestID, saveErr)
+	}
+
+	return result, nil
+}
+
+// convertVolcEngineResponseToOpenAI 将火山引擎的响应转换为标准的OpenAI格式
+func convertVolcEngineResponseToOpenAI(resultJson []byte) (*dto.SimpleResponse, error) {
+	// 添加调试日志
+	fmt.Printf("Original volcengine response: %s\n", string(resultJson))
+
+	// 直接解析为火山引擎的原始格式
+	var volcResponse map[string]interface{}
+	if err := json.Unmarshal(resultJson, &volcResponse); err != nil {
+		return nil, fmt.Errorf("failed to unmarshal volcengine response: %w", err)
+	}
+
+	// 提取choices和usage
+	choices, ok := volcResponse["choices"].([]interface{})
+	if !ok {
+		return nil, fmt.Errorf("invalid choices field in volcengine response")
+	}
+
+	usageRaw, ok := volcResponse["usage"].(map[string]interface{})
+	if !ok {
+		return nil, fmt.Errorf("invalid usage field in volcengine response")
+	}
+
+	// 转换usage
+	usage := dto.Usage{}
+	if promptTokens, ok := usageRaw["prompt_tokens"].(float64); ok {
+		usage.PromptTokens = int(promptTokens)
+	}
+	if completionTokens, ok := usageRaw["completion_tokens"].(float64); ok {
+		usage.CompletionTokens = int(completionTokens)
+	}
+	if totalTokens, ok := usageRaw["total_tokens"].(float64); ok {
+		usage.TotalTokens = int(totalTokens)
+	}
+
+	// 转换为标准的OpenAI格式
+	openaiResponse := &dto.SimpleResponse{
+		Usage:   usage,
+		Choices: []dto.OpenAITextResponseChoice{},
+	}
+
+	// 转换choices
+	for _, choice := range choices {
+		choiceMap, ok := choice.(map[string]interface{})
+		if !ok {
+			continue
+		}
+
+		message, ok := choiceMap["message"].(map[string]interface{})
+		if !ok {
+			continue
+		}
+
+		// 处理content字段
+		content := ""
+		if contentRaw, exists := message["content"]; exists {
+			switch v := contentRaw.(type) {
+			case string:
+				content = v
+			case []interface{}:
+				// 如果是数组，提取文本内容
+				for _, part := range v {
+					if partMap, ok := part.(map[string]interface{}); ok {
+						if partType, ok := partMap["type"].(string); ok && partType == "text" {
+							if text, ok := partMap["text"].(string); ok {
+								content += text
+							}
+						}
+					}
+				}
+			}
+		}
+
+		// 处理reasoning_content字段
+		if reasoningContent, exists := message["reasoning_content"]; exists {
+			if reasoningStr, ok := reasoningContent.(string); ok && reasoningStr != "" {
+				content = reasoningStr + "\n" + content
+			}
+		}
+
+		// 构建转换后的choice
+		index := 0
+		if indexRaw, ok := choiceMap["index"].(float64); ok {
+			index = int(indexRaw)
+		}
+
+		finishReason := ""
+		if finishReasonRaw, ok := choiceMap["finish_reason"].(string); ok {
+			finishReason = finishReasonRaw
+		}
+
+		role := ""
+		if roleRaw, ok := message["role"].(string); ok {
+			role = roleRaw
+		}
+
+		convertedChoice := dto.OpenAITextResponseChoice{
+			Index: index,
+			Message: dto.Message{
+				Role: role,
+			},
+			FinishReason: finishReason,
+		}
+		convertedChoice.Message.SetStringContent(content)
+
+		openaiResponse.Choices = append(openaiResponse.Choices, convertedChoice)
+	}
+
+	// 添加调试日志
+	openaiResponseJson, _ := json.Marshal(openaiResponse)
+	fmt.Printf("Converted OpenAI response: %s\n", string(openaiResponseJson))
+
+	return openaiResponse, nil
+}
+
+// GetBatchRequestAverageDuration 获取batch请求的平均耗时（秒）
+// 这个函数返回一个估算的平均耗时，用于建议重试时间
+func GetBatchRequestAverageDuration() float64 {
+	batchRequestAvgDurationMutex.RLock()
+	defer batchRequestAvgDurationMutex.RUnlock()
+	return batchRequestAvgDuration
+}
+
+// SetBatchRequestAverageDuration 设置batch请求的平均耗时（秒）
+func SetBatchRequestAverageDuration(duration float64) {
+	batchRequestAvgDurationMutex.Lock()
+	defer batchRequestAvgDurationMutex.Unlock()
+	if duration > 0 {
+		batchRequestAvgDuration = duration
+	}
+}
+
+// InitBatchRequestAverageDuration 初始化batch请求的平均耗时
+// 从环境变量读取配置，如果没有配置则使用默认值
+func InitBatchRequestAverageDuration() {
+	if avgDurationStr := os.Getenv("BATCH_REQUEST_AVG_DURATION"); avgDurationStr != "" {
+		if avgDuration, err := strconv.ParseFloat(avgDurationStr, 64); err == nil && avgDuration > 0 {
+			SetBatchRequestAverageDuration(avgDuration)
+		}
+	}
+}
diff --git a/relay/channel/volcengine/keepalive.go b/relay/channel/volcengine/keepalive.go
new file mode 100644
index 000000000000..79972600d998
--- /dev/null
+++ b/relay/channel/volcengine/keepalive.go
@@ -0,0 +1,546 @@
+package volcengine
+
+import (
+	"context"
+	"fmt"
+	"math/rand"
+	"one-api/common"
+	"one-api/middleware"
+	"strings"
+	"sync"
+	"time"
+)
+
+// KeepAliveManager 保活管理器
+type KeepAliveManager struct {
+	keys       map[string]*KeepAliveKey
+	mutex      sync.RWMutex
+	ctx        context.Context
+	cancel     context.CancelFunc
+	isRunning  bool
+	interval   time.Duration
+	expiration time.Duration
+}
+
+// KeepAliveKey 保活key的信息
+type KeepAliveKey struct {
+	Key        string        `json:"key"`
+	CreatedAt  time.Time     `json:"created_at"`  // 创建时间
+	LastTouch  time.Time     `json:"last_touch"`  // 最后触摸时间
+	Expiration time.Duration `json:"expiration"`  // 过期时间
+	Status     string        `json:"status"`      // active, inactive, error
+	ErrorCount int           `json:"error_count"` // 错误计数
+}
+
+// 全局保活管理器实例
+var (
+	keepAliveManager *KeepAliveManager
+	keepAliveOnce    sync.Once
+)
+
+// 默认配置
+const (
+	DefaultKeepAliveInterval   = 10 * time.Second // 默认保活间隔：10分钟
+	DefaultKeepAliveExpiration = 10 * time.Minute // 默认过期时间：4小时
+	MaxKeepAliveDuration       = 10 * time.Minute // 最大保活时间：4小时
+	MaxErrorCount              = 5                // 最大错误次数
+	KeepAliveTriggerTime       = 5 * time.Minute  // 保活触发时间：在key过期前5分钟开始保活
+	MinKeepAliveInterval       = 30 * time.Second // 随机保活时间范围
+	MaxKeepAliveInterval       = 2 * time.Minute  // 随机保活时间范围
+)
+
+// GetKeepAliveManager 获取全局保活管理器实例（单例模式）
+func GetKeepAliveManager() *KeepAliveManager {
+	keepAliveOnce.Do(func() {
+		keepAliveManager = NewKeepAliveManager(DefaultKeepAliveInterval, DefaultKeepAliveExpiration)
+	})
+	return keepAliveManager
+}
+
+// NewKeepAliveManager 创建新的保活管理器
+func NewKeepAliveManager(interval, expiration time.Duration) *KeepAliveManager {
+	ctx, cancel := context.WithCancel(context.Background())
+
+	manager := &KeepAliveManager{
+		keys:       make(map[string]*KeepAliveKey),
+		ctx:        ctx,
+		cancel:     cancel,
+		isRunning:  false,
+		interval:   interval,
+		expiration: expiration,
+	}
+
+	return manager
+}
+
+// Start 启动保活管理器
+func (kam *KeepAliveManager) Start() error {
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	if kam.isRunning {
+		return fmt.Errorf("keep-alive manager is already running")
+	}
+
+	kam.isRunning = true
+
+	// 启动保活协程
+	go kam.keepAliveLoop()
+
+	common.LogInfo(kam.ctx, fmt.Sprintf("Keep-alive manager started with interval: %v, expiration: %v", kam.interval, kam.expiration))
+	return nil
+}
+
+// Stop 停止保活管理器
+func (kam *KeepAliveManager) Stop() error {
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	if !kam.isRunning {
+		return fmt.Errorf("keep-alive manager is not running")
+	}
+
+	kam.isRunning = false
+	kam.cancel()
+
+	common.LogInfo(kam.ctx, "Keep-alive manager stopped")
+	return nil
+}
+
+// AddKey 添加key到保活列表
+func (kam *KeepAliveManager) AddKey(key string, expiration time.Duration) error {
+	if key == "" {
+		return fmt.Errorf("key cannot be empty")
+	}
+
+	if expiration <= 0 {
+		expiration = kam.expiration
+	}
+
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	// 检查key是否已存在
+	if _, exists := kam.keys[key]; exists {
+		return fmt.Errorf("key %s already exists in keep-alive list", key)
+	}
+
+	// 创建新的保活key
+	keepAliveKey := &KeepAliveKey{
+		Key:        key,
+		CreatedAt:  time.Now(),
+		LastTouch:  time.Now(),
+		Expiration: expiration,
+		Status:     "active",
+		ErrorCount: 0,
+	}
+
+	kam.keys[key] = keepAliveKey
+
+	common.LogInfo(kam.ctx, fmt.Sprintf("Added key %s to keep-alive list with expiration: %v", key, expiration))
+	return nil
+}
+
+// RemoveKey 从保活列表中移除key
+func (kam *KeepAliveManager) RemoveKey(key string) error {
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	if _, exists := kam.keys[key]; !exists {
+		return fmt.Errorf("key %s not found in keep-alive list", key)
+	}
+
+	delete(kam.keys, key)
+
+	common.LogInfo(kam.ctx, fmt.Sprintf("Removed key %s from keep-alive list", key))
+	return nil
+}
+
+// GetKey 获取key的信息
+func (kam *KeepAliveManager) GetKey(key string) (*KeepAliveKey, error) {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	keepAliveKey, exists := kam.keys[key]
+	if !exists {
+		return nil, fmt.Errorf("key %s not found in keep-alive list", key)
+	}
+
+	return keepAliveKey, nil
+}
+
+// GetAllKeys 获取所有保活key的信息
+func (kam *KeepAliveManager) GetAllKeys() map[string]*KeepAliveKey {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	// 创建副本以避免并发访问问题
+	result := make(map[string]*KeepAliveKey)
+	for key, value := range kam.keys {
+		result[key] = &KeepAliveKey{
+			Key:        value.Key,
+			CreatedAt:  value.CreatedAt,
+			LastTouch:  value.LastTouch,
+			Expiration: value.Expiration,
+			Status:     value.Status,
+			ErrorCount: value.ErrorCount,
+		}
+	}
+
+	return result
+}
+
+// GetKeyCount 获取保活key的数量
+func (kam *KeepAliveManager) GetKeyCount() int {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	return len(kam.keys)
+}
+
+// IsRunning 检查保活管理器是否正在运行
+func (kam *KeepAliveManager) IsRunning() bool {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	return kam.isRunning
+}
+
+// keepAliveLoop 保活循环
+func (kam *KeepAliveManager) keepAliveLoop() {
+	for {
+		select {
+		case <-kam.ctx.Done():
+			common.LogInfo(kam.ctx, "Keep-alive loop stopped")
+			return
+		default:
+			// 生成随机保活间隔时间
+			randomInterval := kam.generateRandomInterval()
+			common.LogInfo(kam.ctx, fmt.Sprintf("Next keep-alive in %v", randomInterval))
+
+			// 等待随机时间
+			select {
+			case <-kam.ctx.Done():
+				common.LogInfo(kam.ctx, "Keep-alive loop stopped")
+				return
+			case <-time.After(randomInterval):
+				kam.performKeepAlive()
+			}
+		}
+	}
+}
+
+// generateRandomInterval 生成随机保活间隔时间
+func (kam *KeepAliveManager) generateRandomInterval() time.Duration {
+	// 计算随机秒数
+	minSeconds := int(MinKeepAliveInterval.Seconds())
+	maxSeconds := int(MaxKeepAliveInterval.Seconds())
+	randomSeconds := minSeconds + rand.Intn(maxSeconds-minSeconds+1)
+
+	return time.Duration(randomSeconds) * time.Second
+}
+
+// performKeepAlive 执行保活操作
+func (kam *KeepAliveManager) performKeepAlive() {
+	// 使用现有的RequestId生成逻辑创建ctx
+	requestID := middleware.GenerateUniqueRequestId()
+	ctx := context.WithValue(context.Background(), common.RequestIdKey, requestID)
+
+	kam.mutex.RLock()
+	keys := make([]string, 0, len(kam.keys))
+	for key := range kam.keys {
+		keys = append(keys, key)
+	}
+	kam.mutex.RUnlock()
+
+	// 记录本轮保活开始
+	common.LogInfo(ctx, fmt.Sprintf("Keep-alive round started, total keys: %d", len(keys)))
+
+	// 统计变量
+	var (
+		successCount int
+		removedCount int
+		errorCount   int
+		skippedCount int
+	)
+
+	// 逐个处理每个key
+	for i, key := range keys {
+		keyRequestID := fmt.Sprintf("%s-key-%d", requestID, i+1)
+		keyCtx := context.WithValue(ctx, "key_request_id", keyRequestID)
+
+		result := kam.touchKey(keyCtx, key)
+		switch result {
+		case "success":
+			successCount++
+		case "removed":
+			removedCount++
+		case "error":
+			errorCount++
+		case "skipped":
+			skippedCount++
+		}
+	}
+
+	// 记录本轮保活结束
+	common.LogInfo(ctx, fmt.Sprintf("Keep-alive round completed, success: %d, removed: %d, errors: %d, skipped: %d",
+		successCount, removedCount, errorCount, skippedCount))
+}
+
+// touchKey 触摸单个key以保持活跃（内部方法，带详细日志）
+func (kam *KeepAliveManager) touchKey(ctx context.Context, key string) string {
+	kam.mutex.Lock()
+	keepAliveKey, exists := kam.keys[key]
+	if !exists {
+		kam.mutex.Unlock()
+		common.LogInfo(ctx, fmt.Sprintf("Key %s not found in keep-alive list", key))
+		return "removed"
+	}
+	kam.mutex.Unlock()
+
+	// 记录开始处理key
+	common.LogInfo(ctx, fmt.Sprintf("Processing key %s, age: %v, error_count: %d",
+		key, time.Since(keepAliveKey.CreatedAt), keepAliveKey.ErrorCount))
+
+	// 检查是否超过最大保活时间
+	if time.Since(keepAliveKey.CreatedAt) > MaxKeepAliveDuration {
+		common.LogInfo(ctx, fmt.Sprintf("Key %s has exceeded max keep-alive duration (%v), removing from keep-alive list",
+			key, MaxKeepAliveDuration))
+		kam.RemoveKey(key)
+		return "removed"
+	}
+
+	// 检查是否需要保活：只在key快到期的最后5分钟进行保活
+	remainingTime := keepAliveKey.Expiration - time.Since(keepAliveKey.CreatedAt)
+	if remainingTime > KeepAliveTriggerTime {
+		common.LogInfo(ctx, fmt.Sprintf("Key %s has %v remaining, skipping keep-alive (trigger time: %v)",
+			key, remainingTime, KeepAliveTriggerTime))
+		return "skipped"
+	}
+
+	// 尝试触摸key
+	err := kam.touchRedisKey(key, keepAliveKey.Expiration)
+
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	if err != nil {
+		// 检查是否是key不存在的错误
+		if strings.Contains(err.Error(), "does not exist in Redis") {
+			common.LogInfo(ctx, fmt.Sprintf("Key %s no longer exists in Redis, removing from keep-alive list",
+				key))
+			delete(kam.keys, key)
+			return "removed"
+		}
+
+		// 其他错误，增加错误计数
+		keepAliveKey.ErrorCount++
+		keepAliveKey.Status = "error"
+
+		common.LogError(ctx, fmt.Sprintf("Failed to touch key %s: %v (error count: %d)",
+			key, err, keepAliveKey.ErrorCount))
+
+		// 如果错误次数超过阈值，移除key
+		if keepAliveKey.ErrorCount >= MaxErrorCount {
+			common.LogError(ctx, fmt.Sprintf("Key %s exceeded max error count, removing from keep-alive list",
+				key))
+			delete(kam.keys, key)
+			return "removed"
+		}
+
+		return "error"
+	} else {
+		keepAliveKey.LastTouch = time.Now()
+		keepAliveKey.ErrorCount = 0
+		keepAliveKey.Status = "active"
+
+		common.LogInfo(ctx, fmt.Sprintf("Successfully touched key %s, new expiration: %v",
+			key, keepAliveKey.Expiration))
+
+		return "success"
+	}
+}
+
+// touchRedisKey 触摸Redis中的key
+func (kam *KeepAliveManager) touchRedisKey(key string, expiration time.Duration) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 检查key是否存在
+	exists, err := redisClient.Exists(ctx, key).Result()
+	if err != nil {
+		return fmt.Errorf("failed to check key existence: %w", err)
+	}
+
+	if exists == 0 {
+		return fmt.Errorf("key %s does not exist in Redis, should be removed from keep-alive list", key)
+	}
+
+	// 更新key的过期时间
+	err = redisClient.Expire(ctx, key, expiration).Err()
+	if err != nil {
+		return fmt.Errorf("failed to update key expiration: %w", err)
+	}
+
+	return nil
+}
+
+// CleanupExpiredKeys 清理过期的key
+func (kam *KeepAliveManager) CleanupExpiredKeys() int {
+	kam.mutex.Lock()
+	defer kam.mutex.Unlock()
+
+	now := time.Now()
+	removedCount := 0
+
+	for key, keepAliveKey := range kam.keys {
+		if now.Sub(keepAliveKey.LastTouch) > keepAliveKey.Expiration {
+			delete(kam.keys, key)
+			removedCount++
+			common.LogInfo(kam.ctx, fmt.Sprintf("Cleaned up expired key: %s", key))
+		}
+	}
+
+	if removedCount > 0 {
+		common.LogInfo(kam.ctx, fmt.Sprintf("Cleaned up %d expired keys", removedCount))
+	}
+
+	return removedCount
+}
+
+// GetStats 获取保活管理器的统计信息
+func (kam *KeepAliveManager) GetStats() map[string]interface{} {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	stats := make(map[string]interface{})
+	stats["is_running"] = kam.isRunning
+	stats["total_keys"] = len(kam.keys)
+	stats["interval"] = kam.interval.String()
+	stats["expiration"] = kam.expiration.String()
+	stats["max_keep_alive_duration"] = MaxKeepAliveDuration.String()
+
+	// 统计不同状态的key数量
+	statusCount := make(map[string]int)
+	// 统计保活时间分布
+	ageDistribution := make(map[string]int)
+	now := time.Now()
+
+	for _, key := range kam.keys {
+		statusCount[key.Status]++
+
+		// 计算key的年龄并分类
+		age := now.Sub(key.CreatedAt)
+		switch {
+		case age < 1*time.Hour:
+			ageDistribution["<1h"]++
+		case age < 2*time.Hour:
+			ageDistribution["1-2h"]++
+		case age < 3*time.Hour:
+			ageDistribution["2-3h"]++
+		case age < 4*time.Hour:
+			ageDistribution["3-4h"]++
+		default:
+			ageDistribution[">4h"]++
+		}
+	}
+	stats["status_count"] = statusCount
+	stats["age_distribution"] = ageDistribution
+
+	return stats
+}
+
+// 便捷函数，用于快速添加batch_result类型的key
+func AddBatchResultKey(requestID string) error {
+	manager := GetKeepAliveManager()
+	key := "batch_result:" + requestID
+	return manager.AddKey(key, DefaultKeepAliveExpiration)
+}
+
+// 便捷函数，用于快速移除batch_result类型的key
+func RemoveBatchResultKey(requestID string) error {
+	manager := GetKeepAliveManager()
+	key := "batch_result:" + requestID
+	return manager.RemoveKey(key)
+}
+
+// 便捷函数，用于获取batch_result类型key的剩余保活时间
+func GetBatchResultKeyRemainingTime(requestID string) (time.Duration, error) {
+	manager := GetKeepAliveManager()
+	key := "batch_result:" + requestID
+	return manager.GetKeyRemainingKeepAliveTime(key)
+}
+
+// 便捷函数，用于获取batch_result类型key的年龄
+func GetBatchResultKeyAge(requestID string) (time.Duration, error) {
+	manager := GetKeepAliveManager()
+	key := "batch_result:" + requestID
+	return manager.GetKeyAge(key)
+}
+
+// InitKeepAliveManager 初始化并启动保活管理器
+func InitKeepAliveManager() error {
+	manager := GetKeepAliveManager()
+
+	// 如果已经运行，直接返回
+	if manager.IsRunning() {
+		return nil
+	}
+
+	// 启动保活管理器
+	if err := manager.Start(); err != nil {
+		return fmt.Errorf("failed to start keep-alive manager: %w", err)
+	}
+
+	common.LogInfo(context.Background(), "Keep-alive manager initialized and started successfully")
+	return nil
+}
+
+// ShutdownKeepAliveManager 关闭保活管理器
+func ShutdownKeepAliveManager() error {
+	manager := GetKeepAliveManager()
+
+	if !manager.IsRunning() {
+		return nil
+	}
+
+	if err := manager.Stop(); err != nil {
+		return fmt.Errorf("failed to stop keep-alive manager: %w", err)
+	}
+
+	common.LogInfo(context.Background(), "Keep-alive manager shutdown successfully")
+	return nil
+}
+
+// GetKeyRemainingKeepAliveTime 获取key的剩余保活时间
+func (kam *KeepAliveManager) GetKeyRemainingKeepAliveTime(key string) (time.Duration, error) {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	keepAliveKey, exists := kam.keys[key]
+	if !exists {
+		return 0, fmt.Errorf("key %s not found in keep-alive list", key)
+	}
+
+	elapsed := time.Since(keepAliveKey.CreatedAt)
+	remaining := MaxKeepAliveDuration - elapsed
+
+	if remaining <= 0 {
+		return 0, nil
+	}
+
+	return remaining, nil
+}
+
+// GetKeyAge 获取key的年龄（从创建到现在的时间）
+func (kam *KeepAliveManager) GetKeyAge(key string) (time.Duration, error) {
+	kam.mutex.RLock()
+	defer kam.mutex.RUnlock()
+
+	keepAliveKey, exists := kam.keys[key]
+	if !exists {
+		return 0, fmt.Errorf("key %s not found in keep-alive list", key)
+	}
+
+	return time.Since(keepAliveKey.CreatedAt), nil
+}
diff --git a/relay/channel/volcengine/redis_client.go b/relay/channel/volcengine/redis_client.go
new file mode 100644
index 000000000000..2b6b0043c426
--- /dev/null
+++ b/relay/channel/volcengine/redis_client.go
@@ -0,0 +1,300 @@
+package volcengine
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"one-api/common"
+	"sync"
+	"time"
+
+	"github.com/go-redis/redis/v8"
+)
+
+// Redis 客户端
+var (
+	redisClient *redis.Client
+	redisOnce   sync.Once
+)
+
+// BatchResultData 批量推理结果数据结构
+type BatchResultData struct {
+	Result    string `json:"result"`
+	Timestamp int64  `json:"timestamp"`
+	Status    string `json:"status"`
+	RequestID string `json:"request_id"`
+	Error     string `json:"error,omitempty"`
+}
+
+// getRedisClient 获取Redis客户端实例
+func getRedisClient() *redis.Client {
+	redisOnce.Do(func() {
+		// 使用项目统一的Redis客户端
+		if common.RedisEnabled && common.RDB != nil {
+			redisClient = common.RDB
+		} else {
+			// 如果项目Redis未启用，创建一个默认的本地Redis客户端
+			redisClient = redis.NewClient(&redis.Options{
+				Addr:     "localhost:6379",
+				Password: "",
+				DB:       0,
+			})
+		}
+	})
+	return redisClient
+}
+
+// CreateBatchRequestKey 在发起请求前预先创建Redis key
+func CreateBatchRequestKey(requestID string) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 设置过期时间为10分钟
+	expiration := 10 * time.Minute
+
+	// 创建初始状态的数据结构
+	resultData := BatchResultData{
+		Result:    "",
+		Timestamp: time.Now().Unix(),
+		Status:    "pending",
+		RequestID: requestID,
+	}
+
+	resultDataJson, err := json.Marshal(resultData)
+	if err != nil {
+		return fmt.Errorf("failed to marshal initial data: %w", err)
+	}
+
+	// 写入Redis
+	key := "batch_result:" + requestID
+	err = redisClient.Set(ctx, key, string(resultDataJson), expiration).Err()
+	if err != nil {
+		return fmt.Errorf("failed to create initial key in Redis: %w", err)
+	}
+
+	// 将key添加到保活管理器
+	if err := AddBatchResultKey(requestID); err != nil {
+		// 即使添加到保活管理器失败，也不影响Redis key的创建
+		fmt.Printf("Warning: Failed to add key %s to keep-alive manager: %v\n", key, err)
+	}
+
+	fmt.Printf("Successfully created initial Redis key for request %s with status: pending\n", requestID)
+	return nil
+}
+
+// SaveBatchResultToRedis 保存批量推理结果到Redis
+func SaveBatchResultToRedis(requestID string, result interface{}, status string) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 设置过期时间为24小时
+	expiration := 24 * time.Hour
+
+	// 将结果转换为JSON
+	var resultJson []byte
+	var err error
+
+	if result == nil {
+		// 如果结果为nil，使用空字符串
+		resultJson = []byte("")
+	} else {
+		resultJson, err = json.Marshal(result)
+		if err != nil {
+			return fmt.Errorf("failed to marshal result: %w", err)
+		}
+	}
+
+	// 创建结果数据结构
+	resultData := BatchResultData{
+		Result:    string(resultJson),
+		Timestamp: time.Now().Unix(),
+		Status:    status,
+		RequestID: requestID,
+	}
+
+	resultDataJson, err := json.Marshal(resultData)
+	if err != nil {
+		return fmt.Errorf("failed to marshal result data: %w", err)
+	}
+
+	// 写入Redis
+	key := "batch_result:" + requestID
+	err = redisClient.Set(ctx, key, string(resultDataJson), expiration).Err()
+	if err != nil {
+		return fmt.Errorf("failed to write result to Redis: %w", err)
+	}
+
+	fmt.Printf("Successfully wrote result to Redis for request %s with status: %s\n", requestID, status)
+	return nil
+}
+
+// SaveBatchErrorToRedis 保存批量推理错误到Redis
+func SaveBatchErrorToRedis(requestID string, errorMsg string) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 设置过期时间为24小时
+	expiration := 24 * time.Hour
+
+	// 创建错误数据结构
+	resultData := BatchResultData{
+		Result:    "",
+		Timestamp: time.Now().Unix(),
+		Status:    "error",
+		RequestID: requestID,
+		Error:     errorMsg,
+	}
+
+	resultDataJson, err := json.Marshal(resultData)
+	if err != nil {
+		return fmt.Errorf("failed to marshal error data: %w", err)
+	}
+
+	// 写入Redis
+	key := "batch_result:" + requestID
+	err = redisClient.Set(ctx, key, string(resultDataJson), expiration).Err()
+	if err != nil {
+		return fmt.Errorf("failed to write error to Redis: %w", err)
+	}
+
+	fmt.Printf("Successfully wrote error to Redis for request %s\n", requestID)
+	return nil
+}
+
+// GetBatchResultFromRedis 从Redis获取批量推理结果
+func GetBatchResultFromRedis(requestID string) (*BatchResultData, error) {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	key := "batch_result:" + requestID
+	result, err := redisClient.Get(ctx, key).Result()
+	if err != nil {
+		if err == redis.Nil {
+			return nil, fmt.Errorf("result not found for request ID: %s", requestID)
+		}
+		return nil, fmt.Errorf("failed to get result from Redis: %w", err)
+	}
+
+	var resultData BatchResultData
+	if err := json.Unmarshal([]byte(result), &resultData); err != nil {
+		return nil, fmt.Errorf("failed to unmarshal result data: %w", err)
+	}
+
+	return &resultData, nil
+}
+
+// DeleteBatchResultFromRedis 从Redis删除批量推理结果
+func DeleteBatchResultFromRedis(requestID string) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	key := "batch_result:" + requestID
+	err := redisClient.Del(ctx, key).Err()
+	if err != nil {
+		return fmt.Errorf("failed to delete result from Redis: %w", err)
+	}
+
+	fmt.Printf("Successfully deleted result from Redis for request %s\n", requestID)
+	return nil
+}
+
+// ListBatchResultsFromRedis 列出所有批量推理结果
+func ListBatchResultsFromRedis() ([]string, error) {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	pattern := "batch_result:*"
+	keys, err := redisClient.Keys(ctx, pattern).Result()
+	if err != nil {
+		return nil, fmt.Errorf("failed to list keys from Redis: %w", err)
+	}
+
+	return keys, nil
+}
+
+// GetBatchResultCount 获取批量推理结果数量
+func GetBatchResultCount() (int64, error) {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	pattern := "batch_result:*"
+	count, err := redisClient.Keys(ctx, pattern).Result()
+	if err != nil {
+		return 0, fmt.Errorf("failed to count keys from Redis: %w", err)
+	}
+
+	return int64(len(count)), nil
+}
+
+// CleanExpiredBatchResults 清理过期的批量推理结果
+func CleanExpiredBatchResults() error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 获取所有批量推理结果的key
+	keys, err := ListBatchResultsFromRedis()
+	if err != nil {
+		return fmt.Errorf("failed to list keys: %w", err)
+	}
+
+	// 检查每个key的TTL，如果小于等于0则删除
+	for _, key := range keys {
+		ttl, err := redisClient.TTL(ctx, key).Result()
+		if err != nil {
+			fmt.Printf("Failed to get TTL for key %s: %v\n", key, err)
+			continue
+		}
+
+		if ttl <= 0 {
+			err := redisClient.Del(ctx, key).Err()
+			if err != nil {
+				fmt.Printf("Failed to delete expired key %s: %v\n", key, err)
+			} else {
+				fmt.Printf("Successfully deleted expired key %s\n", key)
+			}
+		}
+	}
+
+	return nil
+}
+
+// PingRedis 测试Redis连接
+func PingRedis() error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	_, err := redisClient.Ping(ctx).Result()
+	if err != nil {
+		return fmt.Errorf("failed to ping Redis: %w", err)
+	}
+
+	return nil
+}
+
+// TryAcquireLock 尝试获取分布式锁
+func TryAcquireLock(lockKey string, expiration time.Duration) (bool, error) {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 使用SET命令的NX和EX选项实现分布式锁
+	result, err := redisClient.SetNX(ctx, "lock:"+lockKey, "locked", expiration).Result()
+	if err != nil {
+		return false, fmt.Errorf("failed to acquire lock: %w", err)
+	}
+
+	return result, nil
+}
+
+// ReleaseLock 释放分布式锁
+func ReleaseLock(lockKey string) error {
+	redisClient := getRedisClient()
+	ctx := context.Background()
+
+	// 删除锁
+	err := redisClient.Del(ctx, "lock:"+lockKey).Err()
+	if err != nil {
+		return fmt.Errorf("failed to release lock: %w", err)
+	}
+
+	return nil
+}
diff --git a/relay/common/relay_info.go b/relay/common/relay_info.go
index 5405fc2d4eef..8834c6eb7345 100644
--- a/relay/common/relay_info.go
+++ b/relay/common/relay_info.go
@@ -21,6 +21,7 @@ type RelayInfo struct {
 	ChannelType       int
 	ChannelId         int
 	ChannelTag        string
+	ChannelName       string
 	TokenId           int
 	TokenKey          string
 	UserId            int
@@ -44,6 +45,7 @@ type RelayInfo struct {
 	ApiKey               string
 	Organization         string
 	BaseUrl              string
+	Endpoint             string
 	SupportStreamOptions bool
 	ShouldIncludeUsage   bool
 	IsModelMapped        bool
@@ -91,6 +93,7 @@ func GenRelayInfo(c *gin.Context) *RelayInfo {
 	channelId := c.GetInt("channel_id")
 	channelSetting := c.GetStringMap("channel_setting")
 	channelTag := c.GetString("channel_tag")
+	channelName := c.GetString("channel_name")
 	tokenId := c.GetInt("token_id")
 	tokenKey := c.GetString("token_key")
 	userId := c.GetInt("id")
@@ -108,10 +111,12 @@ func GenRelayInfo(c *gin.Context) *RelayInfo {
 		isFirstResponse:   true,
 		RelayMode:         relayconstant.Path2RelayMode(c.Request.URL.Path),
 		BaseUrl:           c.GetString("base_url"),
+		Endpoint:          c.GetString("endpoint"),
 		RequestURLPath:    c.Request.URL.String(),
 		ChannelType:       channelType,
 		ChannelId:         channelId,
 		ChannelTag:        channelTag,
+		ChannelName:       channelName,
 		TokenId:           tokenId,
 		TokenKey:          tokenKey,
 		UserId:            userId,
diff --git a/relay/constant/api_type.go b/relay/constant/api_type.go
index 273988cf9665..77812503defb 100644
--- a/relay/constant/api_type.go
+++ b/relay/constant/api_type.go
@@ -32,6 +32,7 @@ const (
 	APITypeBaiduV2
 	APITypeOpenRouter
 	APITypeXai
+	APITypeDoubaoOffline
 	APITypeDummy // this one is only for count, do not add any channel after this
 )
 
@@ -92,6 +93,8 @@ func ChannelType2APIType(channelType int) (int, bool) {
 		apiType = APITypeOpenRouter
 	case common.ChannelTypeXai:
 		apiType = APITypeXai
+	case common.ChannelTypeDoubaoOffline:
+		apiType = APITypeVolcEngine
 	}
 	if apiType == -1 {
 		return APITypeOpenAI, false
diff --git a/relay/helper/common.go b/relay/helper/common.go
index 2a72d30a9ce3..225da677d047 100644
--- a/relay/helper/common.go
+++ b/relay/helper/common.go
@@ -4,11 +4,12 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
-	"github.com/gorilla/websocket"
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
+
+	"github.com/gin-gonic/gin"
+	"github.com/gorilla/websocket"
 )
 
 func SetEventStreamHeaders(c *gin.Context) {
diff --git a/relay/relay-audio.go b/relay/relay-audio.go
index b46956e9220f..456ae80876e6 100644
--- a/relay/relay-audio.go
+++ b/relay/relay-audio.go
@@ -3,7 +3,6 @@ package relay
 import (
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
@@ -16,6 +15,8 @@ import (
 	"strconv"
 	"strings"
 	"time"
+
+	"github.com/gin-gonic/gin"
 )
 
 func getAndValidAudioRequest(c *gin.Context, info *relaycommon.RelayInfo) (*dto.AudioRequest, error) {
@@ -71,13 +72,14 @@ func AudioInfo(c *gin.Context) (*relaycommon.RelayInfo, *dto.AudioRequest, *dto.
 func AudioHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, audioRequest *dto.AudioRequest) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 	startTime := time.Now()
 	var funcErr *dto.OpenAIErrorWithStatusCode
-	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, 1)
+	var statusCode int = -1
+	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, 1)
 	defer func() {
 		if funcErr != nil {
-			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
+			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
 		} else {
-			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, 1)
-			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
+			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, strconv.Itoa(statusCode), 1)
+			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, audioRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
 		}
 	}()
 	var (
@@ -160,6 +162,9 @@ func AudioHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, audioRequest
 		return openaiErr
 	}
 
+	// 设置状态码用于指标记录
+	statusCode = resp.(*http.Response).StatusCode
+
 	postConsumeQuota(c, relayInfo, usage.(*dto.Usage), preConsumedQuota, userQuota, priceData, "")
 
 	return nil
diff --git a/relay/relay-image.go b/relay/relay-image.go
index 4d21e617ef76..a638c59dfce2 100644
--- a/relay/relay-image.go
+++ b/relay/relay-image.go
@@ -5,7 +5,6 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
 	"one-api/common"
@@ -19,6 +18,8 @@ import (
 	"strconv"
 	"strings"
 	"time"
+
+	"github.com/gin-gonic/gin"
 )
 
 func getAndValidImageRequest(c *gin.Context, info *relaycommon.RelayInfo) (*dto.ImageRequest, error) {
@@ -87,13 +88,14 @@ func ImageInfo(c *gin.Context) (*relaycommon.RelayInfo, *dto.ImageRequest, *dto.
 func ImageHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, imageRequest *dto.ImageRequest) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 	startTime := time.Now()
 	var funcErr *dto.OpenAIErrorWithStatusCode
-	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, 1)
+	var statusCode int = -1
+	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, 1)
 	defer func() {
 		if funcErr != nil {
-			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
+			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
 		} else {
-			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, 1)
-			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
+			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, strconv.Itoa(statusCode), 1)
+			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, imageRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
 		}
 	}()
 	err := helper.ModelMappedHelper(c, relayInfo)
@@ -116,6 +118,9 @@ func ImageHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, imageRequest
 	}
 
 	userQuota, err := model.GetUserQuota(relayInfo.UserId, false)
+	if err != nil {
+		common.LogError(c, fmt.Sprintf("get_user_quota_failed: %s", err.Error()))
+	}
 
 	sizeRatio := 1.0
 	// Size
@@ -195,6 +200,9 @@ func ImageHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, imageRequest
 		return openaiErr
 	}
 
+	// 设置状态码用于指标记录
+	statusCode = resp.(*http.Response).StatusCode
+
 	usage := &dto.Usage{
 		PromptTokens: imageRequest.N,
 		TotalTokens:  imageRequest.N,
diff --git a/relay/relay-text.go b/relay/relay-text.go
index 10f98738eb18..f8a2637678f6 100644
--- a/relay/relay-text.go
+++ b/relay/relay-text.go
@@ -96,13 +96,14 @@ func TextInfo(c *gin.Context) (*relaycommon.RelayInfo, *dto.GeneralOpenAIRequest
 func TextHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest *dto.GeneralOpenAIRequest) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 	startTime := common.GetBeijingTime()
 	var funcErr *dto.OpenAIErrorWithStatusCode
-	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, 1)
+	var statusCode int = -1
+	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, 1)
 	defer func() {
 		if funcErr != nil {
-			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, strconv.Itoa(openaiErr.StatusCode), 1)
+			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, strconv.Itoa(openaiErr.StatusCode), 1)
 		} else {
-			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, 1)
-			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
+			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, strconv.Itoa(statusCode), 1)
+			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, textRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
 		}
 	}()
 
@@ -141,7 +142,7 @@ func TextHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest *d
 	// Record input tokens metric
 	tokenName := c.GetString("token_name")
 	userName := c.GetString("username")
-	metrics.IncrementInputTokens(strconv.Itoa(relayInfo.ChannelId), textRequest.Model, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(promptTokens))
+	metrics.IncrementInputTokens(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, textRequest.Model, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(promptTokens))
 
 	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptTokens, int(textRequest.MaxTokens))
 	if err != nil {
@@ -221,9 +222,26 @@ func TextHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest *d
 		relayInfo.Headers["X-Test-Traffic"] = "true"
 	}
 
+	// // 如果请求中包含 retry_request_id 头，则添加到 relayInfo 中
+	// if retryRequestId := c.GetHeader("retry_request_id"); retryRequestId != "" {
+	// 	if relayInfo.Headers == nil {
+	// 		relayInfo.Headers = make(map[string]string)
+	// 	}
+	// 	relayInfo.Headers["retry_request_id"] = retryRequestId
+	// }
+
+	// // 如果请求中包含 retry 头，则添加到 relayInfo 中
+	// if retry := c.GetHeader("retry"); retry != "" {
+	// 	if relayInfo.Headers == nil {
+	// 		relayInfo.Headers = make(map[string]string)
+	// 	}
+	// 	relayInfo.Headers["retry"] = retry
+	// }
+
 	statusCodeMappingStr := c.GetString("status_code_mapping")
 	var httpResp *http.Response
 	resp, err := adaptor.DoRequest(c, relayInfo, requestBody)
+
 	if err != nil {
 		funcErr = service.OpenAIErrorWrapper(err, "do_request_failed", http.StatusInternalServerError)
 		return funcErr
@@ -231,13 +249,22 @@ func TextHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest *d
 
 	if resp != nil {
 		httpResp = resp.(*http.Response)
-		// 添加来源标识和重试标记
-		httpResp.Header.Set("X-Origin-User-ID", strconv.Itoa(relayInfo.UserId))
-		httpResp.Header.Set("X-Origin-Channel-ID", strconv.Itoa(relayInfo.ChannelId))
-		httpResp.Header.Set("X-Retry-Count", strconv.Itoa(relayInfo.RetryCount))
-
+		// // 直接设置到 gin 的响应头
+		// c.Writer.Header().Set("X-Origin-User-ID", strconv.Itoa(relayInfo.UserId))
+		// c.Writer.Header().Set("X-Origin-Channel-ID", strconv.Itoa(relayInfo.ChannelId))
+		// c.Writer.Header().Set("X-Retry-Count", strconv.Itoa(relayInfo.RetryCount))
+		// if c.GetHeader("Retry_request_id") != "" {
+		// 	c.Writer.Header().Set("Retry_request_id", c.GetHeader("Retry_request_id"))
+		// }
 		relayInfo.IsStream = relayInfo.IsStream || strings.HasPrefix(httpResp.Header.Get("Content-Type"), "text/event-stream")
 		if httpResp.StatusCode != http.StatusOK {
+			for k, v := range httpResp.Header {
+				if k == "Content-Length" {
+					continue
+				}
+				c.Writer.Header().Set(k, v[0])
+				// common.LogInfo(c, fmt.Sprintf("set header %s = %s", k, v[0]))
+			}
 			openaiErr = service.RelayErrorHandler(httpResp)
 			funcErr = openaiErr
 			// reset status code 重置状态码
@@ -264,6 +291,9 @@ func TextHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest *d
 		return openaiErr
 	}
 	common.LogInfo(c, fmt.Sprintf("response status code: %d, Usage: %+v", httpResp.StatusCode, usage))
+	// 设置状态码用于指标记录
+	statusCode = resp.(*http.Response).StatusCode
+
 	// Store request and response data together if persistence is enabled and status code is 200
 	if model.RequestPersistenceEnabled && httpResp.StatusCode == http.StatusOK && !(c.GetHeader("X-Test-Traffic") == "true") {
 		// 读取请求数据
@@ -486,14 +516,13 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 	}
 
 	// Record token metrics
-	metrics.IncrementOutputTokens(strconv.Itoa(relayInfo.ChannelId), modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(completionTokens))
+	metrics.IncrementOutputTokens(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(completionTokens))
 
 	if cacheTokens > 0 {
-		metrics.IncrementCacheHitTokens(strconv.Itoa(relayInfo.ChannelId), modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(cacheTokens))
+		metrics.IncrementCacheHitTokens(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(cacheTokens))
 	}
 
-	metrics.IncrementInferenceTokens(strconv.Itoa(relayInfo.ChannelId), modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(thinkingTokens))
-	fmt.Println(strconv.Itoa(relayInfo.ChannelId), modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(completionTokens))
+	metrics.IncrementInferenceTokens(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, modelName, relayInfo.Group, strconv.Itoa(relayInfo.UserId), userName, tokenName, float64(thinkingTokens))
 	other := service.GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, cacheTokens, cacheRatio, modelPrice)
 	model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, thinkingTokens, logModel,
 		tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, relayInfo.Group, other)
diff --git a/relay/relay_adaptor.go b/relay/relay_adaptor.go
index bba872d3e4b8..6245d749cb12 100644
--- a/relay/relay_adaptor.go
+++ b/relay/relay_adaptor.go
@@ -89,7 +89,8 @@ func GetAdaptor(apiType int) channel.Adaptor {
 		return &openrouter.Adaptor{}
 	case constant.APITypeXai:
 		return &xai.Adaptor{}
-
+	case constant.APITypeDoubaoOffline:
+		return &volcengine.Adaptor{}
 	}
 	return nil
 }
diff --git a/relay/relay_embedding.go b/relay/relay_embedding.go
index 7607f7d7849e..8dd76439162e 100644
--- a/relay/relay_embedding.go
+++ b/relay/relay_embedding.go
@@ -4,7 +4,6 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
@@ -15,6 +14,8 @@ import (
 	"one-api/service"
 	"strconv"
 	"time"
+
+	"github.com/gin-gonic/gin"
 )
 
 func getEmbeddingPromptToken(embeddingRequest dto.EmbeddingRequest) int {
@@ -50,13 +51,14 @@ func EmbeddingInfo(c *gin.Context) (*relaycommon.RelayInfo, *dto.EmbeddingReques
 func EmbeddingHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, embeddingRequest *dto.EmbeddingRequest) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 	startTime := time.Now()
 	var funcErr *dto.OpenAIErrorWithStatusCode
-	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, 1)
+	var statusCode int = -1
+	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, 1)
 	defer func() {
 		if funcErr != nil {
-			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
+			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
 		} else {
-			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, 1)
-			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
+			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, strconv.Itoa(statusCode), 1)
+			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, embeddingRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
 		}
 	}()
 
@@ -139,6 +141,10 @@ func EmbeddingHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, embedding
 		service.ResetStatusCode(openaiErr, statusCodeMappingStr)
 		return openaiErr
 	}
+
+	// 设置状态码用于指标记录
+	statusCode = resp.(*http.Response).StatusCode
+
 	postConsumeQuota(c, relayInfo, usage.(*dto.Usage), preConsumedQuota, userQuota, priceData, "")
 	return nil
 }
diff --git a/relay/relay_rerank.go b/relay/relay_rerank.go
index 17a5934b09bc..e3c3f0832953 100644
--- a/relay/relay_rerank.go
+++ b/relay/relay_rerank.go
@@ -4,7 +4,6 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
@@ -14,6 +13,8 @@ import (
 	"one-api/service"
 	"strconv"
 	"time"
+
+	"github.com/gin-gonic/gin"
 )
 
 func getRerankPromptToken(rerankRequest dto.RerankRequest) int {
@@ -42,13 +43,14 @@ func RerankInfo(c *gin.Context) (*relaycommon.RelayInfo, *dto.RerankRequest, *dt
 func RerankHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, rerankRequest *dto.RerankRequest) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 	startTime := time.Now()
 	var funcErr *dto.OpenAIErrorWithStatusCode
-	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, 1)
+	var statusCode int = -1
+	metrics.IncrementRelayRequestTotalCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, 1)
 	defer func() {
 		if funcErr != nil {
-			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
+			metrics.IncrementRelayRequestFailedCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, strconv.Itoa(funcErr.StatusCode), 1)
 		} else {
-			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, 1)
-			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
+			metrics.IncrementRelayRequestSuccessCounter(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, strconv.Itoa(statusCode), 1)
+			metrics.ObserveRelayRequestDuration(strconv.Itoa(relayInfo.ChannelId), relayInfo.ChannelName, relayInfo.ChannelTag, relayInfo.BaseUrl, rerankRequest.Model, relayInfo.Group, time.Since(startTime).Seconds())
 		}
 	}()
 	if rerankRequest.Query == "" {
@@ -132,6 +134,10 @@ func RerankHelper(c *gin.Context, relayInfo *relaycommon.RelayInfo, rerankReques
 		service.ResetStatusCode(openaiErr, statusCodeMappingStr)
 		return openaiErr
 	}
+
+	// 设置状态码用于指标记录
+	statusCode = resp.(*http.Response).StatusCode
+
 	postConsumeQuota(c, relayInfo, usage.(*dto.Usage), preConsumedQuota, userQuota, priceData, "")
 	return nil
 }
diff --git a/relay/relay_task.go b/relay/relay_task.go
index 01237c37bbe8..ee66647eafd4 100644
--- a/relay/relay_task.go
+++ b/relay/relay_task.go
@@ -5,7 +5,6 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
 	"one-api/common"
@@ -17,6 +16,8 @@ import (
 	"one-api/service"
 	"one-api/setting"
 	"one-api/setting/operation_setting"
+
+	"github.com/gin-gonic/gin"
 )
 
 /*
diff --git a/router/api-router.go b/router/api-router.go
index 1701313bf8a8..48333309248f 100644
--- a/router/api-router.go
+++ b/router/api-router.go
@@ -13,6 +13,7 @@ func SetApiRouter(router *gin.Engine) {
 	apiRouter.Use(gzip.Gzip(gzip.DefaultCompression))
 	apiRouter.Use(middleware.GlobalAPIRateLimit())
 	{
+		apiRouter.GET("/ping", controller.Ping)
 		apiRouter.GET("/status", controller.GetStatus)
 		apiRouter.GET("/models", middleware.UserAuth(), controller.DashboardListModels)
 		apiRouter.GET("/status/test", middleware.AdminAuth(), controller.TestStatus)
diff --git a/router/relay-router.go b/router/relay-router.go
index f547767ebd6b..54799467e1e8 100644
--- a/router/relay-router.go
+++ b/router/relay-router.go
@@ -1,15 +1,20 @@
 package router
 
 import (
-	"github.com/gin-gonic/gin"
 	"one-api/controller"
 	"one-api/middleware"
 	"one-api/relay"
+
+	"github.com/gin-gonic/gin"
 )
 
 func SetRelayRouter(router *gin.Engine) {
 	router.Use(middleware.CORS())
 	router.Use(middleware.DecompressRequestMiddleware())
+
+	// 添加ping路由用于测试
+	router.GET("/ping", controller.Ping)
+
 	// https://platform.openai.com/docs/api-reference/introduction
 	modelsRouter := router.Group("/v1/models")
 	modelsRouter.Use(middleware.TokenAuth())
diff --git a/web/src/constants/channel.constants.js b/web/src/constants/channel.constants.js
index 5738d656bb1f..60641cca44d6 100644
--- a/web/src/constants/channel.constants.js
+++ b/web/src/constants/channel.constants.js
@@ -108,5 +108,10 @@ export const CHANNEL_OPTIONS = [
     value: 44,
     color: 'purple',
     label: '嵌入模型：MokaAI M3E'
+  },
+  {
+    value: 100,
+    color: 'blue',
+    label: '豆包离线'
   }
 ];
diff --git a/web/src/pages/Channel/EditChannel.js b/web/src/pages/Channel/EditChannel.js
index bfc611fe7fd5..824dc4898b46 100644
--- a/web/src/pages/Channel/EditChannel.js
+++ b/web/src/pages/Channel/EditChannel.js
@@ -77,6 +77,7 @@ const EditChannel = (props) => {
     openai_organization: '',
     max_input_tokens: 0,
     base_url: '',
+    endpoint: '',
     other: '',
     model_mapping: '',
     status_code_mapping: '',
@@ -537,7 +538,7 @@ const EditChannel = (props) => {
             value={inputs.name}
             autoComplete="new-password"
           />
-          {inputs.type !== 3 && inputs.type !== 8 && inputs.type !== 22 && inputs.type !== 36 && inputs.type !== 45 && (
+          {inputs.type !== 3 && inputs.type !== 8 && inputs.type !== 22 && inputs.type !== 36 && inputs.type !== 45 && inputs.type !== 100 && (
             <>
               <div style={{ marginTop: 10 }}>
                 <Typography.Text strong>{t('代理站地址')}：</Typography.Text>
@@ -556,6 +557,25 @@ const EditChannel = (props) => {
               </Tooltip>
             </>
           )}
+          {inputs.type === 100 && (
+            <>
+              <div style={{ marginTop: 10 }}>
+                <Typography.Text strong>{t('接入点地址')}：</Typography.Text>
+              </div>
+              <Tooltip content={t('请输入豆包离线部署的接入点地址')}>
+                <Input
+                  label={t('接入点地址')}
+                  name="endpoint"
+                  placeholder={t('请输入豆包离线部署的接入点地址，例如：ep-bi-20250620135343-67crm')}
+                  onChange={(value) => {
+                    handleInputChange('endpoint', value);
+                  }}
+                  value={inputs.endpoint}
+                  autoComplete="new-password"
+                />
+              </Tooltip>
+            </>
+          )}
           <div style={{ marginTop: 10 }}>
             <Typography.Text strong>{t('密钥')}：</Typography.Text>
           </div>