feat: 完善audio计费

This commit is contained in:
CalciumIon
2024-11-07 16:12:09 +08:00
parent be652fa3c2
commit 97fdcd8e8f
9 changed files with 173 additions and 19 deletions

View File

@@ -19,12 +19,26 @@ func GenerateTextOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, m
return other
}
func GenerateWssOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.RealtimeUsage, modelRatio, groupRatio, completionRatio, modelPrice float64) map[string]interface{} {
func GenerateWssOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.RealtimeUsage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice float64) map[string]interface{} {
info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
info["ws"] = true
info["audio_input"] = usage.InputTokenDetails.AudioTokens
info["audio_output"] = usage.OutputTokenDetails.AudioTokens
info["text_input"] = usage.InputTokenDetails.TextTokens
info["text_output"] = usage.OutputTokenDetails.TextTokens
info["audio_ratio"] = audioRatio
info["audio_completion_ratio"] = audioCompletionRatio
return info
}
func GenerateAudioOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.Usage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice float64) map[string]interface{} {
info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
info["audio"] = true
info["audio_input"] = usage.PromptTokensDetails.AudioTokens
info["audio_output"] = usage.CompletionTokenDetails.AudioTokens
info["text_input"] = usage.PromptTokensDetails.TextTokens
info["text_output"] = usage.CompletionTokenDetails.TextTokens
info["audio_ratio"] = audioRatio
info["audio_completion_ratio"] = audioCompletionRatio
return info
}

View File

@@ -9,7 +9,6 @@ import (
"one-api/dto"
"one-api/model"
relaycommon "one-api/relay/common"
"strings"
"time"
)
@@ -124,18 +123,79 @@ func PostWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, mod
}
logModel := modelName
if strings.HasPrefix(logModel, "gpt-4-gizmo") {
logModel = "gpt-4-gizmo-*"
logContent += fmt.Sprintf(",模型 %s", modelName)
}
if strings.HasPrefix(logModel, "gpt-4o-gizmo") {
logModel = "gpt-4o-gizmo-*"
logContent += fmt.Sprintf(",模型 %s", modelName)
}
if extraContent != "" {
logContent += ", " + extraContent
}
other := GenerateWssOtherInfo(ctx, relayInfo, usage, modelRatio, groupRatio, completionRatio, modelPrice)
other := GenerateWssOtherInfo(ctx, relayInfo, usage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice)
model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, usage.InputTokens, usage.OutputTokens, logModel,
tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, other)
}
func PostAudioConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
usage *dto.Usage, ratio float64, preConsumedQuota int, userQuota int, modelRatio float64,
groupRatio float64,
modelPrice float64, usePrice bool, extraContent string) {
useTimeSeconds := time.Now().Unix() - relayInfo.StartTime.Unix()
textInputTokens := usage.PromptTokensDetails.TextTokens
textOutTokens := usage.CompletionTokenDetails.TextTokens
audioInputTokens := usage.PromptTokensDetails.AudioTokens
audioOutTokens := usage.CompletionTokenDetails.AudioTokens
tokenName := ctx.GetString("token_name")
completionRatio := common.GetCompletionRatio(relayInfo.UpstreamModelName)
audioRatio := common.GetAudioRatio(relayInfo.UpstreamModelName)
audioCompletionRatio := common.GetAudioCompletionRatio(relayInfo.UpstreamModelName)
quota := 0
if !usePrice {
quota = int(math.Round(float64(textInputTokens) + float64(textOutTokens)*completionRatio))
quota += int(math.Round(float64(audioInputTokens)*audioRatio + float64(audioOutTokens)*audioRatio*audioCompletionRatio))
quota = int(math.Round(float64(quota) * ratio))
if ratio != 0 && quota <= 0 {
quota = 1
}
} else {
quota = int(modelPrice * common.QuotaPerUnit * groupRatio)
}
totalTokens := usage.TotalTokens
var logContent string
if !usePrice {
logContent = fmt.Sprintf("模型倍率 %.2f,补全倍率 %.2f,音频倍率 %.2f,音频补全倍率 %.2f,分组倍率 %.2f", modelRatio, completionRatio, audioRatio, audioCompletionRatio, groupRatio)
} else {
logContent = fmt.Sprintf("模型价格 %.2f,分组倍率 %.2f", modelPrice, groupRatio)
}
// record all the consume log even if quota is 0
if totalTokens == 0 {
// in this case, must be some error happened
// we cannot just return, because we may have to return the pre-consumed quota
quota = 0
logContent += fmt.Sprintf("(可能是上游超时)")
common.LogError(ctx, fmt.Sprintf("total tokens is 0, cannot consume quota, userId %d, channelId %d, "+
"tokenId %d, model %s pre-consumed quota %d", relayInfo.UserId, relayInfo.ChannelId, relayInfo.TokenId, relayInfo.UpstreamModelName, preConsumedQuota))
} else {
quotaDelta := quota - preConsumedQuota
if quotaDelta != 0 {
err := model.PostConsumeTokenQuota(relayInfo, userQuota, quotaDelta, preConsumedQuota, true)
if err != nil {
common.LogError(ctx, "error consuming token remain quota: "+err.Error())
}
}
err := model.CacheUpdateUserQuota(relayInfo.UserId)
if err != nil {
common.LogError(ctx, "error update user quota cache: "+err.Error())
}
model.UpdateUserUsedQuotaAndRequestCount(relayInfo.UserId, quota)
model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
}
logModel := relayInfo.UpstreamModelName
if extraContent != "" {
logContent += ", " + extraContent
}
other := GenerateAudioOtherInfo(ctx, relayInfo, usage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice)
model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, usage.PromptTokens, usage.CompletionTokens, logModel,
tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, other)
}