Ecosystem/ai-gateway
5
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

disable MaxTokens

Browse Source
This commit is contained in:
quzard 2023-06-14 17:44:47 +08:00
parent d086f34cff
commit 90cd939df1
3 changed files with 22 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
controller/channel-test.go
View File

@ -67,8 +67,8 @@ func testChannel(channel *model.Channel, request *ChatRequest) error {
func buildTestRequest(c *gin.Context) *ChatRequest { func buildTestRequest(c *gin.Context) *ChatRequest {
model_ := c.Query("model") model_ := c.Query("model")
testRequest := &ChatRequest{ testRequest := &ChatRequest{
Model: model_, Model: model_,
MaxTokens: 1, //MaxTokens: 1,
} }
testMessage := Message{ testMessage := Message{
Role: "user", Role: "user",

2
controller/relay-palm.go
View File

@ -51,7 +51,7 @@ func relayPaLM(openAIRequest GeneralOpenAIRequest, c *gin.Context) *OpenAIErrorW
Temperature: openAIRequest.Temperature, Temperature: openAIRequest.Temperature,
CandidateCount: openAIRequest.N, CandidateCount: openAIRequest.N,
TopP: openAIRequest.TopP, TopP: openAIRequest.TopP,
TopK: openAIRequest.MaxTokens, //TopK: openAIRequest.MaxTokens,
} }
// TODO: forward request to PaLM & convert response // TODO: forward request to PaLM & convert response
fmt.Print(request) fmt.Print(request)

38
controller/relay.go
View File

@ -30,28 +30,28 @@ const (
// https://platform.openai.com/docs/api-reference/chat // https://platform.openai.com/docs/api-reference/chat
type GeneralOpenAIRequest struct { type GeneralOpenAIRequest struct {
Model string `json:"model"` Model string `json:"model"`
Messages []Message `json:"messages"` Messages []Message `json:"messages"`
Prompt string `json:"prompt"` Prompt string `json:"prompt"`
Stream bool `json:"stream"` Stream bool `json:"stream"`
MaxTokens int `json:"max_tokens"` //MaxTokens int `json:"max_tokens"`
Temperature float64 `json:"temperature"` Temperature float64 `json:"temperature"`
TopP float64 `json:"top_p"` TopP float64 `json:"top_p"`
N int `json:"n"` N int `json:"n"`
Input any `json:"input"` Input any `json:"input"`
} }
type ChatRequest struct { type ChatRequest struct {
Model string `json:"model"` Model string `json:"model"`
Messages []Message `json:"messages"` Messages []Message `json:"messages"`
MaxTokens int `json:"max_tokens"` //MaxTokens int `json:"max_tokens"`
} }
type TextRequest struct { type TextRequest struct {
Model string `json:"model"` Model string `json:"model"`
Messages []Message `json:"messages"` Messages []Message `json:"messages"`
Prompt string `json:"prompt"` Prompt string `json:"prompt"`
MaxTokens int `json:"max_tokens"` //MaxTokens int `json:"max_tokens"`
//Stream bool `json:"stream"` //Stream bool `json:"stream"`
} }
@ -193,9 +193,9 @@ func relayHelper(c *gin.Context, relayMode int) *OpenAIErrorWithStatusCode {
promptTokens = countTokenInput(textRequest.Input, textRequest.Model) promptTokens = countTokenInput(textRequest.Input, textRequest.Model)
} }
preConsumedTokens := common.PreConsumedQuota preConsumedTokens := common.PreConsumedQuota
if textRequest.MaxTokens != 0 { //if textRequest.MaxTokens != 0 {
preConsumedTokens = promptTokens + textRequest.MaxTokens // preConsumedTokens = promptTokens + textRequest.MaxTokens
} //}
modelRatio := common.GetModelRatio(textRequest.Model) modelRatio := common.GetModelRatio(textRequest.Model)
groupRatio := common.GetGroupRatio(group) groupRatio := common.GetGroupRatio(group)
ratio := modelRatio * groupRatio ratio := modelRatio * groupRatio