diff --git a/controller/relay-audio.go b/controller/relay-audio.go
index 4cb74ae..bb1f5c5 100644
--- a/controller/relay-audio.go
+++ b/controller/relay-audio.go
@@ -162,7 +162,7 @@ func relayAudioHelper(c *gin.Context, relayMode int) *OpenAIErrorWithStatusCode
 			if quota != 0 {
 				tokenName := c.GetString("token_name")
 				logContent := fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f", modelRatio, groupRatio)
-				model.RecordConsumeLog(ctx, userId, channelId, promptTokens, 0, audioRequest.Model, tokenName, quota, logContent, tokenId)
+				model.RecordConsumeLog(ctx, userId, channelId, promptTokens, 0, audioRequest.Model, tokenName, quota, logContent, tokenId, userQuota)
 				model.UpdateUserUsedQuotaAndRequestCount(userId, quota)
 				channelId := c.GetInt("channel_id")
 				model.UpdateChannelUsedQuota(channelId, quota)
diff --git a/controller/relay-image.go b/controller/relay-image.go
index 85c4d5d..fa9dc96 100644
--- a/controller/relay-image.go
+++ b/controller/relay-image.go
@@ -169,7 +169,7 @@ func relayImageHelper(c *gin.Context, relayMode int) *OpenAIErrorWithStatusCode
 			if quota != 0 {
 				tokenName := c.GetString("token_name")
 				logContent := fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f", modelRatio, groupRatio)
-				model.RecordConsumeLog(ctx, userId, channelId, 0, 0, imageRequest.Model, tokenName, quota, logContent, tokenId)
+				model.RecordConsumeLog(ctx, userId, channelId, 0, 0, imageRequest.Model, tokenName, quota, logContent, tokenId, userQuota)
 				model.UpdateUserUsedQuotaAndRequestCount(userId, quota)
 				channelId := c.GetInt("channel_id")
 				model.UpdateChannelUsedQuota(channelId, quota)
diff --git a/controller/relay-mj.go b/controller/relay-mj.go
index 778b455..8d88473 100644
--- a/controller/relay-mj.go
+++ b/controller/relay-mj.go
@@ -370,7 +370,7 @@ func relayMidjourneySubmit(c *gin.Context, relayMode int) *MidjourneyResponse {
 			if quota != 0 {
 				tokenName := c.GetString("token_name")
 				logContent := fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f", modelRatio, groupRatio)
-				model.RecordConsumeLog(ctx, userId, channelId, 0, 0, imageModel, tokenName, quota, logContent, tokenId)
+				model.RecordConsumeLog(ctx, userId, channelId, 0, 0, imageModel, tokenName, quota, logContent, tokenId, userQuota)
 				model.UpdateUserUsedQuotaAndRequestCount(userId, quota)
 				channelId := c.GetInt("channel_id")
 				model.UpdateChannelUsedQuota(channelId, quota)
diff --git a/controller/relay-text.go b/controller/relay-text.go
index 66fc492..93a0ab6 100644
--- a/controller/relay-text.go
+++ b/controller/relay-text.go
@@ -446,7 +446,7 @@ func relayTextHelper(c *gin.Context, relayMode int) *OpenAIErrorWithStatusCode {
 			// record all the consume log even if quota is 0
 			useTimeSeconds := time.Now().Unix() - startTime.Unix()
 			logContent := fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f，用时 %d秒", modelRatio, groupRatio, useTimeSeconds)
-			model.RecordConsumeLog(ctx, userId, channelId, promptTokens, completionTokens, textRequest.Model, tokenName, quota, logContent, tokenId)
+			model.RecordConsumeLog(ctx, userId, channelId, promptTokens, completionTokens, textRequest.Model, tokenName, quota, logContent, tokenId, userQuota)
 			model.UpdateUserUsedQuotaAndRequestCount(userId, quota)
 			model.UpdateChannelUsedQuota(channelId, quota)
 			//if quota != 0 {
diff --git a/model/log.go b/model/log.go
index 269230d..bb3d7d8 100644
--- a/model/log.go
+++ b/model/log.go
@@ -54,8 +54,8 @@ func RecordLog(userId int, logType int, content string) {
 	}
 }
 
-func RecordConsumeLog(ctx context.Context, userId int, channelId int, promptTokens int, completionTokens int, modelName string, tokenName string, quota int, content string, tokenId int) {
-	common.LogInfo(ctx, fmt.Sprintf("record consume log: userId=%d, channelId=%d, promptTokens=%d, completionTokens=%d, modelName=%s, tokenName=%s, quota=%d, content=%s", userId, channelId, promptTokens, completionTokens, modelName, tokenName, quota, content))
+func RecordConsumeLog(ctx context.Context, userId int, channelId int, promptTokens int, completionTokens int, modelName string, tokenName string, quota int, content string, tokenId int, userQuota int) {
+	common.LogInfo(ctx, fmt.Sprintf("record consume log: userId=%d, 用户调用前余额=%d, channelId=%d, promptTokens=%d, completionTokens=%d, modelName=%s, tokenName=%s, quota=%d, content=%s", userId, userQuota, channelId, promptTokens, completionTokens, modelName, tokenName, quota, content))
 	if !common.LogConsumeEnabled {
 		return
 	}
diff --git a/model/token.go b/model/token.go
index d13335e..7b06309 100644
--- a/model/token.go
+++ b/model/token.go
@@ -219,6 +219,17 @@ func PostConsumeTokenQuota(tokenId int, userQuota int, quota int, preConsumedQuo
 		return err
 	}
 
+	if !token.UnlimitedQuota {
+		if quota > 0 {
+			err = DecreaseTokenQuota(tokenId, quota)
+		} else {
+			err = IncreaseTokenQuota(tokenId, -quota)
+		}
+		if err != nil {
+			return err
+		}
+	}
+
 	if sendEmail {
 		if (quota + preConsumedQuota) != 0 {
 			quotaTooLow := userQuota >= common.QuotaRemindThreshold && userQuota-(quota+preConsumedQuota) < common.QuotaRemindThreshold
@@ -247,15 +258,5 @@ func PostConsumeTokenQuota(tokenId int, userQuota int, quota int, preConsumedQuo
 		}
 	}
 
-	if !token.UnlimitedQuota {
-		if quota > 0 {
-			err = DecreaseTokenQuota(tokenId, quota)
-		} else {
-			err = IncreaseTokenQuota(tokenId, -quota)
-		}
-		if err != nil {
-			return err
-		}
-	}
 	return nil
 }