xinyin025/new-api
1
0
Fork 0
mirror of https://github.com/linux-do/new-api.git synced 2025-11-13 01:23:41 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

merge upstream

Browse Source 
Signed-off-by: wozulong <>
This commit is contained in:
wozulong
2024-05-31 11:14:25 +08:00
parent bcf1b160d1 8a27977284
commit 69b9950aa0
32 changed files with 4523 additions and 369 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
relay/channel/claude/relay-claude.go
View File

@@ -370,7 +370,7 @@ func claudeHandler(requestMode int, c *gin.Context, resp *http.Response, promptT
}, nil
}
fullTextResponse := ResponseClaude2OpenAI(requestMode, &claudeResponse)
completionTokens, err, _ := service.CountTokenText(claudeResponse.Completion, model, false)
completionTokens, err := service.CountTokenText(claudeResponse.Completion, model)
if err != nil {
return service.OpenAIErrorWrapper(err, "count_token_text_failed", http.StatusInternalServerError), nil
}

2
relay/channel/gemini/relay-gemini.go
View File

@@ -256,7 +256,7 @@ func geminiChatHandler(c *gin.Context, resp *http.Response, promptTokens int, mo
}, nil
}
fullTextResponse := responseGeminiChat2OpenAI(&geminiResponse)
completionTokens, _, _ := service.CountTokenText(geminiResponse.GetResponseText(), model, false)
completionTokens, _ := service.CountTokenText(geminiResponse.GetResponseText(), model)
usage := dto.Usage{
PromptTokens: promptTokens,
CompletionTokens: completionTokens,

2
relay/channel/openai/relay-openai.go
View File

@@ -190,7 +190,7 @@ func OpenaiHandler(c *gin.Context, resp *http.Response, promptTokens int, model
if simpleResponse.Usage.TotalTokens == 0 {
completionTokens := 0
for _, choice := range simpleResponse.Choices {
ctkm, _, _ := service.CountTokenText(string(choice.Message.Content), model, false)
ctkm, _ := service.CountTokenText(string(choice.Message.Content), model)
completionTokens += ctkm
}
simpleResponse.Usage = dto.Usage{

2
relay/channel/palm/relay-palm.go
View File

@@ -156,7 +156,7 @@ func palmHandler(c *gin.Context, resp *http.Response, promptTokens int, model st
}, nil
}
fullTextResponse := responsePaLM2OpenAI(&palmResponse)
completionTokens, _, _ := service.CountTokenText(palmResponse.Candidates[0].Content, model, false)
completionTokens, _ := service.CountTokenText(palmResponse.Candidates[0].Content, model)
usage := dto.Usage{
PromptTokens: promptTokens,
CompletionTokens: completionTokens,

10
relay/relay-audio.go
View File

@@ -55,7 +55,13 @@ func AudioHelper(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode {
promptTokens := 0
preConsumedTokens := common.PreConsumedQuota
if strings.HasPrefix(audioRequest.Model, "tts-1") {
promptTokens, err, _ = service.CountAudioToken(audioRequest.Input, audioRequest.Model, constant.ShouldCheckPromptSensitive())
if constant.ShouldCheckPromptSensitive() {
err = service.CheckSensitiveInput(audioRequest.Input)
if err != nil {
return service.OpenAIErrorWrapper(err, "sensitive_words_detected", http.StatusBadRequest)
}
}
promptTokens, err = service.CountAudioToken(audioRequest.Input, audioRequest.Model)
if err != nil {
return service.OpenAIErrorWrapper(err, "count_audio_token_failed", http.StatusInternalServerError)
}
@@ -178,7 +184,7 @@ func AudioHelper(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode {
if strings.HasPrefix(audioRequest.Model, "tts-1") {
quota = promptTokens
} else {
quota, err, _ = service.CountAudioToken(audioResponse.Text, audioRequest.Model, false)
quota, err = service.CountAudioToken(audioResponse.Text, audioRequest.Model)
}
quota = int(float64(quota) * ratio)
if ratio != 0 && quota <= 0 {

8
relay/relay-image.go
View File

@@ -10,6 +10,7 @@ import (
"io"
"net/http"
"one-api/common"
"one-api/constant"
"one-api/dto"
"one-api/model"
relaycommon "one-api/relay/common"
@@ -47,6 +48,13 @@ func RelayImageHelper(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusC
return service.OpenAIErrorWrapper(errors.New("prompt is required"), "required_field_missing", http.StatusBadRequest)
}
if constant.ShouldCheckPromptSensitive() {
err = service.CheckSensitiveInput(imageRequest.Prompt)
if err != nil {
return service.OpenAIErrorWrapper(err, "sensitive_words_detected", http.StatusBadRequest)
}
}
if strings.Contains(imageRequest.Size, "×") {
return service.OpenAIErrorWrapper(errors.New("size an unexpected error occurred in the parameter, please use 'x' instead of the multiplication sign '×'"), "invalid_field_value", http.StatusBadRequest)
}

4
relay/relay-mj.go
View File

@@ -158,7 +158,7 @@ func RelaySwapFace(c *gin.Context) *dto.MidjourneyResponse {
modelPrice, success := common.GetModelPrice(modelName, true)
// 如果没有配置价格，则使用默认价格
if !success {
defaultPrice, ok := common.DefaultModelPrice[modelName]
defaultPrice, ok := common.GetDefaultModelRatioMap()[modelName]
if !ok {
modelPrice = 0.1
} else {
@@ -457,7 +457,7 @@ func RelayMidjourneySubmit(c *gin.Context, relayMode int) *dto.MidjourneyRespons
modelPrice, success := common.GetModelPrice(modelName, true)
// 如果没有配置价格，则使用默认价格
if !success {
defaultPrice, ok := common.DefaultModelPrice[modelName]
defaultPrice, ok := common.GetDefaultModelRatioMap()[modelName]
if !ok {
modelPrice = 0.1
} else {

50
relay/relay-text.go
View File

@@ -98,13 +98,17 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
var ratio float64
var modelRatio float64
//err := service.SensitiveWordsCheck(textRequest)
promptTokens, err, sensitiveTrigger := getPromptTokens(textRequest, relayInfo)
// count messages token error 计算promptTokens错误
if err != nil {
if sensitiveTrigger {
if constant.ShouldCheckPromptSensitive() {
err = checkRequestSensitive(textRequest, relayInfo)
if err != nil {
return service.OpenAIErrorWrapperLocal(err, "sensitive_words_detected", http.StatusBadRequest)
}
}
promptTokens, err := getPromptTokens(textRequest, relayInfo)
// count messages token error 计算promptTokens错误
if err != nil {
return service.OpenAIErrorWrapper(err, "count_token_messages_failed", http.StatusInternalServerError)
}
@@ -128,7 +132,7 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
adaptor := GetAdaptor(relayInfo.ApiType)
if adaptor == nil {
return service.OpenAIErrorWrapper(fmt.Errorf("invalid api type: %d", relayInfo.ApiType), "invalid_api_type", http.StatusBadRequest)
return service.OpenAIErrorWrapperLocal(fmt.Errorf("invalid api type: %d", relayInfo.ApiType), "invalid_api_type", http.StatusBadRequest)
}
adaptor.Init(relayInfo, *textRequest)
var requestBody io.Reader
@@ -136,7 +140,7 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
if isModelMapped {
jsonStr, err := json.Marshal(textRequest)
if err != nil {
return service.OpenAIErrorWrapper(err, "marshal_text_request_failed", http.StatusInternalServerError)
return service.OpenAIErrorWrapperLocal(err, "marshal_text_request_failed", http.StatusInternalServerError)
}
requestBody = bytes.NewBuffer(jsonStr)
} else {
@@ -145,11 +149,11 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
} else {
convertedRequest, err := adaptor.ConvertRequest(c, relayInfo.RelayMode, textRequest)
if err != nil {
return service.OpenAIErrorWrapper(err, "convert_request_failed", http.StatusInternalServerError)
return service.OpenAIErrorWrapperLocal(err, "convert_request_failed", http.StatusInternalServerError)
}
jsonData, err := json.Marshal(convertedRequest)
if err != nil {
return service.OpenAIErrorWrapper(err, "json_marshal_failed", http.StatusInternalServerError)
return service.OpenAIErrorWrapperLocal(err, "json_marshal_failed", http.StatusInternalServerError)
}
requestBody = bytes.NewBuffer(jsonData)
}
@@ -182,15 +186,14 @@ func TextHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
return nil
}
func getPromptTokens(textRequest *dto.GeneralOpenAIRequest, info *relaycommon.RelayInfo) (int, error, bool) {
func getPromptTokens(textRequest *dto.GeneralOpenAIRequest, info *relaycommon.RelayInfo) (int, error) {
var promptTokens int
var err error
var sensitiveTrigger bool
checkSensitive := constant.ShouldCheckPromptSensitive()
switch info.RelayMode {
case relayconstant.RelayModeChatCompletions:
promptTokens, err, sensitiveTrigger = service.CountTokenChatRequest(*textRequest, textRequest.Model, checkSensitive)
promptTokens, err = service.CountTokenChatRequest(*textRequest, textRequest.Model)
case relayconstant.RelayModeCompletions:
promptTokens, err = service.CountTokenInput(textRequest.Prompt, textRequest.Model)
prompts := textRequest.Prompt
switch v := prompts.(type) {
case string:
@@ -199,17 +202,32 @@ func getPromptTokens(textRequest *dto.GeneralOpenAIRequest, info *relaycommon.Re
prompts = append(v, textRequest.Suffix)
}
promptTokens, err, sensitiveTrigger = service.CountTokenInput(prompts, textRequest.Model, checkSensitive)
promptTokens, err = service.CountTokenInput(prompts, textRequest.Model)
case relayconstant.RelayModeModerations:
promptTokens, err, sensitiveTrigger = service.CountTokenInput(textRequest.Input, textRequest.Model, checkSensitive)
promptTokens, err = service.CountTokenInput(textRequest.Input, textRequest.Model)
case relayconstant.RelayModeEmbeddings:
promptTokens, err, sensitiveTrigger = service.CountTokenInput(textRequest.Input, textRequest.Model, checkSensitive)
promptTokens, err = service.CountTokenInput(textRequest.Input, textRequest.Model)
default:
err = errors.New("unknown relay mode")
promptTokens = 0
}
info.PromptTokens = promptTokens
return promptTokens, err, sensitiveTrigger
return promptTokens, err
}
func checkRequestSensitive(textRequest *dto.GeneralOpenAIRequest, info *relaycommon.RelayInfo) error {
var err error
switch info.RelayMode {
case relayconstant.RelayModeChatCompletions:
err = service.CheckSensitiveMessages(textRequest.Messages)
case relayconstant.RelayModeCompletions:
err = service.CheckSensitiveInput(textRequest.Prompt)
case relayconstant.RelayModeModerations:
err = service.CheckSensitiveInput(textRequest.Input)
case relayconstant.RelayModeEmbeddings:
err = service.CheckSensitiveInput(textRequest.Input)
}
return err
}
// 预扣费并返回用户剩余配额