Merge remote-tracking branch 'origin/upstream/main'

relay/channel/anthropic/constants.go

@@ -2,7 +2,7 @@ package anthropic
 
 var ModelList = []string{
 	"claude-instant-1.2", "claude-2.0", "claude-2.1",
-	"claude-3-haiku-20240229",
+	"claude-3-haiku-20240307",
 	"claude-3-sonnet-20240229",
 	"claude-3-opus-20240229",
 }

relay/channel/baidu/constants.go

@@ -7,4 +7,7 @@ var ModelList = []string{
 	"ERNIE-Speed",
 	"ERNIE-Bot-turbo",
 	"Embedding-V1",
+	"bge-large-zh",
+	"bge-large-en",
+	"tao-8k",
 }

relay/channel/lingyiwanwu/constants.go

@@ -0,0 +1,9 @@
+package lingyiwanwu
+
+// https://platform.lingyiwanwu.com/docs
+
+var ModelList = []string{
+	"yi-34b-chat-0205",
+	"yi-34b-chat-200k",
+	"yi-vl-plus",
+}

relay/channel/ollama/adaptor.go

@@ -0,0 +1,65 @@
+package ollama
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/relay/channel"
+	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/model"
+	"github.com/songquanpeng/one-api/relay/util"
+	"io"
+	"net/http"
+)
+
+type Adaptor struct {
+}
+
+func (a *Adaptor) Init(meta *util.RelayMeta) {
+
+}
+
+func (a *Adaptor) GetRequestURL(meta *util.RelayMeta) (string, error) {
+	// https://github.com/ollama/ollama/blob/main/docs/api.md
+	fullRequestURL := fmt.Sprintf("%s/api/chat", meta.BaseURL)
+	return fullRequestURL, nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *util.RelayMeta) error {
+	channel.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
+	return nil
+}
+
+func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	switch relayMode {
+	case constant.RelayModeEmbeddings:
+		return nil, errors.New("not supported")
+	default:
+		return ConvertRequest(*request), nil
+	}
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, meta *util.RelayMeta, requestBody io.Reader) (*http.Response, error) {
+	return channel.DoRequestHelper(a, c, meta, requestBody)
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *util.RelayMeta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
+	if meta.IsStream {
+		err, usage = StreamHandler(c, resp)
+	} else {
+		err, usage = Handler(c, resp)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "ollama"
+}

relay/channel/ollama/constants.go

@@ -0,0 +1,5 @@
+package ollama
+
+var ModelList = []string{
+	"qwen:0.5b-chat",
+}

relay/channel/ollama/main.go

@@ -0,0 +1,178 @@
+package ollama
+
+import (
+	"bufio"
+	"context"
+	"encoding/json"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/relay/channel/openai"
+	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
+)
+
+func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
+	ollamaRequest := ChatRequest{
+		Model: request.Model,
+		Options: &Options{
+			Seed:             int(request.Seed),
+			Temperature:      request.Temperature,
+			TopP:             request.TopP,
+			FrequencyPenalty: request.FrequencyPenalty,
+			PresencePenalty:  request.PresencePenalty,
+		},
+		Stream: request.Stream,
+	}
+	for _, message := range request.Messages {
+		ollamaRequest.Messages = append(ollamaRequest.Messages, Message{
+			Role:    message.Role,
+			Content: message.StringContent(),
+		})
+	}
+	return &ollamaRequest
+}
+
+func responseOllama2OpenAI(response *ChatResponse) *openai.TextResponse {
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    response.Message.Role,
+			Content: response.Message.Content,
+		},
+	}
+	if response.Done {
+		choice.FinishReason = "stop"
+	}
+	fullTextResponse := openai.TextResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", helper.GetUUID()),
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+		Usage: model.Usage{
+			PromptTokens:     response.PromptEvalCount,
+			CompletionTokens: response.EvalCount,
+			TotalTokens:      response.PromptEvalCount + response.EvalCount,
+		},
+	}
+	return &fullTextResponse
+}
+
+func streamResponseOllama2OpenAI(ollamaResponse *ChatResponse) *openai.ChatCompletionsStreamResponse {
+	var choice openai.ChatCompletionsStreamResponseChoice
+	choice.Delta.Role = ollamaResponse.Message.Role
+	choice.Delta.Content = ollamaResponse.Message.Content
+	if ollamaResponse.Done {
+		choice.FinishReason = &constant.StopFinishReason
+	}
+	response := openai.ChatCompletionsStreamResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", helper.GetUUID()),
+		Object:  "chat.completion.chunk",
+		Created: helper.GetTimestamp(),
+		Model:   ollamaResponse.Model,
+		Choices: []openai.ChatCompletionsStreamResponseChoice{choice},
+	}
+	return &response
+}
+
+func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var usage model.Usage
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "}\n"); i >= 0 {
+			return i + 2, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := strings.TrimPrefix(scanner.Text(), "}")
+			dataChan <- data + "}"
+		}
+		stopChan <- true
+	}()
+	common.SetEventStreamHeaders(c)
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var ollamaResponse ChatResponse
+			err := json.Unmarshal([]byte(data), &ollamaResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			if ollamaResponse.EvalCount != 0 {
+				usage.PromptTokens = ollamaResponse.PromptEvalCount
+				usage.CompletionTokens = ollamaResponse.EvalCount
+				usage.TotalTokens = ollamaResponse.PromptEvalCount + ollamaResponse.EvalCount
+			}
+			response := streamResponseOllama2OpenAI(&ollamaResponse)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
+		}
+	})
+	err := resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	return nil, &usage
+}
+
+func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	ctx := context.TODO()
+	var ollamaResponse ChatResponse
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	logger.Debugf(ctx, "ollama response: %s", string(responseBody))
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &ollamaResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	if ollamaResponse.Error != "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: ollamaResponse.Error,
+				Type:    "ollama_error",
+				Param:   "",
+				Code:    "ollama_error",
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	fullTextResponse := responseOllama2OpenAI(&ollamaResponse)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &fullTextResponse.Usage
+}

relay/channel/ollama/model.go

@@ -0,0 +1,37 @@
+package ollama
+
+type Options struct {
+	Seed             int     `json:"seed,omitempty"`
+	Temperature      float64 `json:"temperature,omitempty"`
+	TopK             int     `json:"top_k,omitempty"`
+	TopP             float64 `json:"top_p,omitempty"`
+	FrequencyPenalty float64 `json:"frequency_penalty,omitempty"`
+	PresencePenalty  float64 `json:"presence_penalty,omitempty"`
+}
+
+type Message struct {
+	Role    string   `json:"role,omitempty"`
+	Content string   `json:"content,omitempty"`
+	Images  []string `json:"images,omitempty"`
+}
+
+type ChatRequest struct {
+	Model    string    `json:"model,omitempty"`
+	Messages []Message `json:"messages,omitempty"`
+	Stream   bool      `json:"stream"`
+	Options  *Options  `json:"options,omitempty"`
+}
+
+type ChatResponse struct {
+	Model           string  `json:"model,omitempty"`
+	CreatedAt       string  `json:"created_at,omitempty"`
+	Message         Message `json:"message,omitempty"`
+	Response        string  `json:"response,omitempty"` // for stream response
+	Done            bool    `json:"done,omitempty"`
+	TotalDuration   int     `json:"total_duration,omitempty"`
+	LoadDuration    int     `json:"load_duration,omitempty"`
+	PromptEvalCount int     `json:"prompt_eval_count,omitempty"`
+	EvalCount       int     `json:"eval_count,omitempty"`
+	EvalDuration    int     `json:"eval_duration,omitempty"`
+	Error           string  `json:"error,omitempty"`
+}

relay/channel/openai/compatible.go

@@ -5,6 +5,7 @@ import (
 	"github.com/songquanpeng/one-api/relay/channel/ai360"
 	"github.com/songquanpeng/one-api/relay/channel/baichuan"
 	"github.com/songquanpeng/one-api/relay/channel/groq"
+	"github.com/songquanpeng/one-api/relay/channel/lingyiwanwu"
 	"github.com/songquanpeng/one-api/relay/channel/minimax"
 	"github.com/songquanpeng/one-api/relay/channel/mistral"
 	"github.com/songquanpeng/one-api/relay/channel/moonshot"
@@ -18,6 +19,7 @@ var CompatibleChannels = []int{
 	common.ChannelTypeMinimax,
 	common.ChannelTypeMistral,
 	common.ChannelTypeGroq,
+	common.ChannelTypeLingYiWanWu,
 }
 
 func GetCompatibleChannelMeta(channelType int) (string, []string) {
@@ -36,6 +38,8 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) {
 		return "mistralai", mistral.ModelList
 	case common.ChannelTypeGroq:
 		return "groq", groq.ModelList
+	case common.ChannelTypeLingYiWanWu:
+		return "lingyiwanwu", lingyiwanwu.ModelList
 	default:
 		return "openai", ModelList
 	}

relay/constant/api_type.go

@@ -15,6 +15,7 @@ const (
 	APITypeAIProxyLibrary
 	APITypeTencent
 	APITypeGemini
+	APITypeOllama
 
 	APITypeDummy // this one is only for count, do not add any channel after this
 )
@@ -40,6 +41,8 @@ func ChannelType2APIType(channelType int) int {
 		apiType = APITypeTencent
 	case common.ChannelTypeGemini:
 		apiType = APITypeGemini
+	case common.ChannelTypeOllama:
+		apiType = APITypeOllama
 	}
 	return apiType
 }

relay/controller/audio.go

@@ -23,6 +23,7 @@ import (
 )
 
 func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode {
+	ctx := c.Request.Context()
 	audioModel := "whisper-1"
 
 	tokenId := c.GetInt("token_id")
@@ -51,16 +52,16 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 	// groupRatio := common.GetGroupRatio(group)
 	groupRatio := c.GetFloat64("channel_ratio")
 	ratio := modelRatio * groupRatio
-	var quota int
-	var preConsumedQuota int
+	var quota int64
+	var preConsumedQuota int64
 	switch relayMode {
 	case constant.RelayModeAudioSpeech:
-		preConsumedQuota = int(float64(len(ttsRequest.Input)) * ratio)
+		preConsumedQuota = int64(float64(len(ttsRequest.Input)) * ratio)
 		quota = preConsumedQuota
 	default:
-		preConsumedQuota = int(float64(config.PreConsumedQuota) * ratio)
+		preConsumedQuota = int64(float64(config.PreConsumedQuota) * ratio)
 	}
-	userQuota, err := model.CacheGetUserQuota(userId)
+	userQuota, err := model.CacheGetUserQuota(ctx, userId)
 	if err != nil {
 		return openai.ErrorWrapper(err, "get_user_quota_failed", http.StatusInternalServerError)
 	}
@@ -185,7 +186,7 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		if err != nil {
 			return openai.ErrorWrapper(err, "get_text_from_body_err", http.StatusInternalServerError)
 		}
-		quota = openai.CountTokenText(text, audioModel)
+		quota = int64(openai.CountTokenText(text, audioModel))
 		resp.Body = io.NopCloser(bytes.NewBuffer(responseBody))
 	}
 	if resp.StatusCode != http.StatusOK {

relay/controller/helper.go

@@ -107,18 +107,18 @@ func getPromptTokens(textRequest *relaymodel.GeneralOpenAIRequest, relayMode int
 	return 0
 }
 
-func getPreConsumedQuota(textRequest *relaymodel.GeneralOpenAIRequest, promptTokens int, ratio float64) int {
+func getPreConsumedQuota(textRequest *relaymodel.GeneralOpenAIRequest, promptTokens int, ratio float64) int64 {
 	preConsumedTokens := config.PreConsumedQuota
 	if textRequest.MaxTokens != 0 {
-		preConsumedTokens = promptTokens + textRequest.MaxTokens
+		preConsumedTokens = int64(promptTokens) + int64(textRequest.MaxTokens)
 	}
-	return int(float64(preConsumedTokens) * ratio)
+	return int64(float64(preConsumedTokens) * ratio)
 }
 
-func preConsumeQuota(ctx context.Context, textRequest *relaymodel.GeneralOpenAIRequest, promptTokens int, ratio float64, meta *util.RelayMeta) (int, *relaymodel.ErrorWithStatusCode) {
+func preConsumeQuota(ctx context.Context, textRequest *relaymodel.GeneralOpenAIRequest, promptTokens int, ratio float64, meta *util.RelayMeta) (int64, *relaymodel.ErrorWithStatusCode) {
 	preConsumedQuota := getPreConsumedQuota(textRequest, promptTokens, ratio)
 
-	userQuota, err := model.CacheGetUserQuota(meta.UserId)
+	userQuota, err := model.CacheGetUserQuota(ctx, meta.UserId)
 	if err != nil {
 		return preConsumedQuota, openai.ErrorWrapper(err, "get_user_quota_failed", http.StatusInternalServerError)
 	}
@@ -144,16 +144,16 @@ func preConsumeQuota(ctx context.Context, textRequest *relaymodel.GeneralOpenAIR
 	return preConsumedQuota, nil
 }
 
-func postConsumeQuota(ctx context.Context, usage *relaymodel.Usage, meta *util.RelayMeta, textRequest *relaymodel.GeneralOpenAIRequest, ratio float64, preConsumedQuota int, modelRatio float64, groupRatio float64) {
+func postConsumeQuota(ctx context.Context, usage *relaymodel.Usage, meta *util.RelayMeta, textRequest *relaymodel.GeneralOpenAIRequest, ratio float64, preConsumedQuota int64, modelRatio float64, groupRatio float64) {
 	if usage == nil {
 		logger.Error(ctx, "usage is nil, which is unexpected")
 		return
 	}
-	quota := 0
+	var quota int64
 	completionRatio := common.GetCompletionRatio(textRequest.Model)
 	promptTokens := usage.PromptTokens
 	completionTokens := usage.CompletionTokens
-	quota = int(math.Ceil((float64(promptTokens) + float64(completionTokens)*completionRatio) * ratio))
+	quota = int64(math.Ceil((float64(promptTokens) + float64(completionTokens)*completionRatio) * ratio))
 	if ratio != 0 && quota <= 0 {
 		quota = 1
 	}
@@ -168,7 +168,7 @@ func postConsumeQuota(ctx context.Context, usage *relaymodel.Usage, meta *util.R
 	if err != nil {
 		logger.Error(ctx, "error consuming token remain quota: "+err.Error())
 	}
-	err = model.CacheUpdateUserQuota(meta.UserId)
+	err = model.CacheUpdateUserQuota(ctx, meta.UserId)
 	if err != nil {
 		logger.Error(ctx, "error update user quota cache: "+err.Error())
 	}

relay/controller/image.go

@@ -81,9 +81,9 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 	// groupRatio := common.GetGroupRatio(meta.Group)
 	groupRatio := c.GetFloat64("channel_ratio") // pre-selected cheapest channel ratio
 	ratio := modelRatio * groupRatio
-	userQuota, err := model.CacheGetUserQuota(meta.UserId)
+	userQuota, err := model.CacheGetUserQuota(ctx, meta.UserId)
 
-	quota := int(ratio*imageCostRatio*1000) * imageRequest.N
+	quota := int64(ratio*imageCostRatio*1000) * int64(imageRequest.N)
 
 	if userQuota-quota < 0 {
 		return openai.ErrorWrapper(errors.New("user quota is not enough"), "insufficient_user_quota", http.StatusForbidden)
@@ -127,7 +127,7 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		if err != nil {
 			logger.SysError("error consuming token remain quota: " + err.Error())
 		}
-		err = model.CacheUpdateUserQuota(meta.UserId)
+		err = model.CacheUpdateUserQuota(ctx, meta.UserId)
 		if err != nil {
 			logger.SysError("error update user quota cache: " + err.Error())
 		}

relay/controller/text.go

@@ -77,6 +77,7 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 		if err != nil {
 			return openai.ErrorWrapper(err, "json_marshal_failed", http.StatusInternalServerError)
 		}
+		logger.Debugf(ctx, "converted request: \n%s", string(jsonData))
 		requestBody = bytes.NewBuffer(jsonData)
 	}
 

relay/helper/main.go

@@ -5,6 +5,7 @@ import (
 	"github.com/songquanpeng/one-api/relay/channel/aiproxy"
 	"github.com/songquanpeng/one-api/relay/channel/anthropic"
 	"github.com/songquanpeng/one-api/relay/channel/gemini"
+	"github.com/songquanpeng/one-api/relay/channel/ollama"
 	"github.com/songquanpeng/one-api/relay/channel/openai"
 	"github.com/songquanpeng/one-api/relay/channel/palm"
 	"github.com/songquanpeng/one-api/relay/constant"
@@ -26,12 +27,14 @@ func GetAdaptor(apiType int) channel.Adaptor {
 		return &openai.Adaptor{}
 	case constant.APITypePaLM:
 		return &palm.Adaptor{}
-		// case constant.APITypeTencent:
-		// 	return &tencent.Adaptor{}
-		// case constant.APITypeXunfei:
-		// 	return &xunfei.Adaptor{}
-		// case constant.APITypeZhipu:
-		// 	return &zhipu.Adaptor{}
+	// case constant.APITypeTencent:
+	// 	return &tencent.Adaptor{}
+	// case constant.APITypeXunfei:
+	// 	return &xunfei.Adaptor{}
+	// case constant.APITypeZhipu:
+	// 	return &zhipu.Adaptor{}
+	case constant.APITypeOllama:
+		return &ollama.Adaptor{}
 	}
 	return nil
 }

relay/util/billing.go

@@ -6,7 +6,7 @@ import (
 	"github.com/songquanpeng/one-api/model"
 )
 
-func ReturnPreConsumedQuota(ctx context.Context, preConsumedQuota int, tokenId int) {
+func ReturnPreConsumedQuota(ctx context.Context, preConsumedQuota int64, tokenId int) {
 	if preConsumedQuota != 0 {
 		go func(ctx context.Context) {
 			// return pre-consumed quota

relay/util/common.go

@@ -35,10 +35,17 @@ func ShouldDisableChannel(err *relaymodel.Error, statusCode int) bool {
 		return true
 	case "permission_error":
 		return true
+	case "forbidden":
+		return true
 	}
 	if err.Code == "invalid_api_key" || err.Code == "account_deactivated" {
 		return true
 	}
+	if strings.HasPrefix(err.Message, "Your credit balance is too low") { // anthropic
+		return true
+	} else if strings.HasPrefix(err.Message, "This organization has been disabled.") {
+		return true
+	}
 	return false
 }
 
@@ -148,20 +155,20 @@ func GetFullRequestURL(baseURL string, requestURL string, channelType int) strin
 	return fullRequestURL
 }
 
-func PostConsumeQuota(ctx context.Context, tokenId int, quotaDelta int, totalQuota int, userId int, channelId int, modelRatio float64, groupRatio float64, modelName string, tokenName string) {
+func PostConsumeQuota(ctx context.Context, tokenId int, quotaDelta int64, totalQuota int64, userId int, channelId int, modelRatio float64, groupRatio float64, modelName string, tokenName string) {
 	// quotaDelta is remaining quota to be consumed
 	err := model.PostConsumeTokenQuota(tokenId, quotaDelta)
 	if err != nil {
 		logger.SysError("error consuming token remain quota: " + err.Error())
 	}
-	err = model.CacheUpdateUserQuota(userId)
+	err = model.CacheUpdateUserQuota(ctx, userId)
 	if err != nil {
 		logger.SysError("error update user quota cache: " + err.Error())
 	}
 	// totalQuota is total quota consumed
 	if totalQuota >= 0 {
 		logContent := fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f", modelRatio, groupRatio)
-		model.RecordConsumeLog(ctx, userId, channelId, totalQuota, 0, modelName, tokenName, totalQuota, logContent)
+		model.RecordConsumeLog(ctx, userId, channelId, int(totalQuota), 0, modelName, tokenName, totalQuota, logContent)
 		model.UpdateUserUsedQuotaAndRequestCount(userId, totalQuota)
 		model.UpdateChannelUsedQuota(channelId, totalQuota)
 	}