xinyin025/new-api
1
0
Fork 0
mirror of https://github.com/linux-do/new-api.git synced 2025-11-09 15:43:41 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

feat: support cohere rerank

Browse Source
This commit is contained in:
CalciumIon
2024-07-06 17:09:22 +08:00
parent afe02c6aa5
commit 8af4e28f75
25 changed files with 347 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
relay/relay-text.go
View File

@@ -182,7 +182,7 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
service.ResetStatusCode(openaiErr, statusCodeMappingStr)
return openaiErr
}
postConsumeQuota(c, relayInfo, *textRequest, usage, ratio, preConsumedQuota, userQuota, modelRatio, groupRatio, modelPrice, success)
postConsumeQuota(c, relayInfo, textRequest.Model, usage, ratio, preConsumedQuota, userQuota, modelRatio, groupRatio, modelPrice, success)
return nil
}
@@ -272,7 +272,7 @@ func returnPreConsumedQuota(c *gin.Context, tokenId int, userQuota int, preConsu
}
}
func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRequest dto.GeneralOpenAIRequest,
func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, modelName string,
usage *dto.Usage, ratio float64, preConsumedQuota int, userQuota int, modelRatio float64, groupRatio float64,
modelPrice float64, usePrice bool) {
@@ -281,7 +281,7 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRe
completionTokens := usage.CompletionTokens
tokenName := ctx.GetString("token_name")
completionRatio := common.GetCompletionRatio(textRequest.Model)
completionRatio := common.GetCompletionRatio(modelName)
quota := 0
if !usePrice {
@@ -307,7 +307,8 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRe
// 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, textRequest.Model, preConsumedQuota))
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, modelName, preConsumedQuota))
} else {
//if sensitiveResp != nil {
// logContent += fmt.Sprintf("，敏感词：%s", strings.Join(sensitiveResp.SensitiveWords, ", "))
@@ -327,13 +328,14 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRe
model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
}
logModel := textRequest.Model
logModel := modelName
if strings.HasPrefix(logModel, "gpt-4-gizmo") {
logModel = "gpt-4-gizmo-*"
logContent += fmt.Sprintf("，模型 %s", textRequest.Model)
logContent += fmt.Sprintf("，模型 %s", modelName)
}
other := service.GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, logModel, tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, other)
model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, logModel,
tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, other)
//if quota != 0 {
//