oadmin/new-api
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: gemini audio input billing

Browse Source
This commit is contained in:
creamlike1024
2025-06-07 12:26:23 +08:00
parent f17b4f0760
commit 9496dac448
8 changed files with 386 additions and 260 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
relay/channel/gemini/dto.go
View File

@@ -112,10 +112,16 @@ type GeminiChatResponse struct {
}
type GeminiUsageMetadata struct {
PromptTokenCount int `json:"promptTokenCount"`
CandidatesTokenCount int `json:"candidatesTokenCount"`
TotalTokenCount int `json:"totalTokenCount"`
ThoughtsTokenCount int `json:"thoughtsTokenCount"`
PromptTokenCount int `json:"promptTokenCount"`
CandidatesTokenCount int `json:"candidatesTokenCount"`
TotalTokenCount int `json:"totalTokenCount"`
ThoughtsTokenCount int `json:"thoughtsTokenCount"`
PromptTokensDetails []GeminiPromptTokensDetails `json:"promptTokensDetails"`
}
type GeminiPromptTokensDetails struct {
Modality string `json:"modality"`
TokenCount int `json:"tokenCount"`
}
// Imagen related structs

19
relay/channel/gemini/relay-gemini-native.go
View File

@@ -55,6 +55,16 @@ func GeminiTextGenerationHandler(c *gin.Context, resp *http.Response, info *rela
TotalTokens: geminiResponse.UsageMetadata.TotalTokenCount,
}
usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
if detail.Modality == "AUDIO" {
usage.PromptTokensDetails.AudioTokens = detail.TokenCount
} else if detail.Modality == "TEXT" {
usage.PromptTokensDetails.TextTokens = detail.TokenCount
}
}
// 直接返回 Gemini 原生格式的 JSON 响应
jsonResponse, err := json.Marshal(geminiResponse)
if err != nil {
@@ -100,6 +110,14 @@ func GeminiTextGenerationStreamHandler(c *gin.Context, resp *http.Response, info
usage.PromptTokens = geminiResponse.UsageMetadata.PromptTokenCount
usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount
usage.TotalTokens = geminiResponse.UsageMetadata.TotalTokenCount
usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
if detail.Modality == "AUDIO" {
usage.PromptTokensDetails.AudioTokens = detail.TokenCount
} else if detail.Modality == "TEXT" {
usage.PromptTokensDetails.TextTokens = detail.TokenCount
}
}
}
// 直接发送 GeminiChatResponse 响应
@@ -118,7 +136,6 @@ func GeminiTextGenerationStreamHandler(c *gin.Context, resp *http.Response, info
}
// 计算最终使用量
usage.PromptTokensDetails.TextTokens = usage.PromptTokens
usage.CompletionTokens = usage.TotalTokens - usage.PromptTokens
// 移除流式响应结尾的[Done]，因为Gemini API没有发送Done的行为

19
relay/channel/gemini/relay-gemini.go
View File

@@ -313,13 +313,13 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest, info *relaycommon
if part.GetInputAudio().Data == "" {
return nil, fmt.Errorf("only base64 audio is supported in gemini")
}
format, base64String, err := service.DecodeBase64FileData(part.GetInputAudio().Data)
base64String, err := service.DecodeBase64AudioData(part.GetInputAudio().Data)
if err != nil {
return nil, fmt.Errorf("decode base64 audio data failed: %s", err.Error())
}
parts = append(parts, GeminiPart{
InlineData: &GeminiInlineData{
MimeType: format,
MimeType: "audio/" + part.GetInputAudio().Format,
Data: base64String,
},
})
@@ -771,6 +771,13 @@ func GeminiChatStreamHandler(c *gin.Context, resp *http.Response, info *relaycom
usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount
usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
usage.TotalTokens = geminiResponse.UsageMetadata.TotalTokenCount
for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
if detail.Modality == "AUDIO" {
usage.PromptTokensDetails.AudioTokens = detail.TokenCount
} else if detail.Modality == "TEXT" {
usage.PromptTokensDetails.TextTokens = detail.TokenCount
}
}
}
err = helper.ObjectData(c, response)
if err != nil {
@@ -845,6 +852,14 @@ func GeminiChatHandler(c *gin.Context, resp *http.Response, info *relaycommon.Re
usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
usage.CompletionTokens = usage.TotalTokens - usage.PromptTokens
for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
if detail.Modality == "AUDIO" {
usage.PromptTokensDetails.AudioTokens = detail.TokenCount
} else if detail.Modality == "TEXT" {
usage.PromptTokensDetails.TextTokens = detail.TokenCount
}
}
fullTextResponse.Usage = usage
jsonResponse, err := json.Marshal(fullTextResponse)
if err != nil {

47
relay/relay-text.go
View File

@@ -352,6 +352,7 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
promptTokens := usage.PromptTokens
cacheTokens := usage.PromptTokensDetails.CachedTokens
imageTokens := usage.PromptTokensDetails.ImageTokens
audioTokens := usage.PromptTokensDetails.AudioTokens
completionTokens := usage.CompletionTokens
modelName := relayInfo.OriginModelName
@@ -367,6 +368,7 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
dPromptTokens := decimal.NewFromInt(int64(promptTokens))
dCacheTokens := decimal.NewFromInt(int64(cacheTokens))
dImageTokens := decimal.NewFromInt(int64(imageTokens))
dAudioTokens := decimal.NewFromInt(int64(audioTokens))
dCompletionTokens := decimal.NewFromInt(int64(completionTokens))
dCompletionRatio := decimal.NewFromFloat(completionRatio)
dCacheRatio := decimal.NewFromFloat(cacheRatio)
@@ -412,23 +414,43 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
dFileSearchQuota = decimal.NewFromFloat(fileSearchPrice).
Mul(decimal.NewFromInt(int64(fileSearchTool.CallCount))).
Div(decimal.NewFromInt(1000)).Mul(dGroupRatio).Mul(dQuotaPerUnit)
extraContent += fmt.Sprintf("File Search 调用 %d 次，调用花费 $%s",
extraContent += fmt.Sprintf("File Search 调用 %d 次，调用花费 %s",
fileSearchTool.CallCount, dFileSearchQuota.String())
}
}
var quotaCalculateDecimal decimal.Decimal
if !priceData.UsePrice {
nonCachedTokens := dPromptTokens.Sub(dCacheTokens)
cachedTokensWithRatio := dCacheTokens.Mul(dCacheRatio)
promptQuota := nonCachedTokens.Add(cachedTokensWithRatio)
if imageTokens > 0 {
nonImageTokens := dPromptTokens.Sub(dImageTokens)
imageTokensWithRatio := dImageTokens.Mul(dImageRatio)
promptQuota = nonImageTokens.Add(imageTokensWithRatio)
var audioInputQuota decimal.Decimal
var audioInputPrice float64
if !priceData.UsePrice {
baseTokens := dPromptTokens
// 减去 cached tokens
var cachedTokensWithRatio decimal.Decimal
if !dCacheTokens.IsZero() {
baseTokens = baseTokens.Sub(dCacheTokens)
cachedTokensWithRatio = dCacheTokens.Mul(dCacheRatio)
}
// 减去 image tokens
var imageTokensWithRatio decimal.Decimal
if !dImageTokens.IsZero() {
baseTokens = baseTokens.Sub(dImageTokens)
imageTokensWithRatio = dImageTokens.Mul(dImageRatio)
}
// 减去 Gemini audio tokens
if !dAudioTokens.IsZero() {
audioInputPrice = operation_setting.GetGeminiInputAudioPricePerMillionTokens(modelName)
if audioInputPrice > 0 {
// 重新计算 base tokens
baseTokens = baseTokens.Sub(dAudioTokens)
audioInputQuota = decimal.NewFromFloat(audioInputPrice).Div(decimal.NewFromInt(1000000)).Mul(dAudioTokens).Mul(dGroupRatio).Mul(dQuotaPerUnit)
extraContent += fmt.Sprintf("Audio Input 花费 %s", audioInputQuota.String())
}
}
promptQuota := baseTokens.Add(cachedTokensWithRatio).Add(imageTokensWithRatio)
completionQuota := dCompletionTokens.Mul(dCompletionRatio)
quotaCalculateDecimal = promptQuota.Add(completionQuota).Mul(ratio)
@@ -442,6 +464,8 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
// 添加 responses tools call 调用的配额
quotaCalculateDecimal = quotaCalculateDecimal.Add(dWebSearchQuota)
quotaCalculateDecimal = quotaCalculateDecimal.Add(dFileSearchQuota)
// 添加 audio input 独立计费
quotaCalculateDecimal = quotaCalculateDecimal.Add(audioInputQuota)
quota := int(quotaCalculateDecimal.Round(0).IntPart())
totalTokens := promptTokens + completionTokens
@@ -512,6 +536,11 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
other["file_search_price"] = fileSearchPrice
}
}
if !audioInputQuota.IsZero() {
other["audio_input_seperate_price"] = true
other["audio_input_token_count"] = audioTokens
other["audio_input_price"] = audioInputPrice
}
model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, logModel,
tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, relayInfo.Group, other)
}