Merge remote-tracking branch 'upstream/main'

relay/channel/aws/dto.go

@@ -5,6 +5,7 @@ import "one-api/relay/channel/claude"
 type AwsClaudeRequest struct {
 	// AnthropicVersion should be "bedrock-2023-05-31"
 	AnthropicVersion string                 `json:"anthropic_version"`
+	System           string                 `json:"system"`
 	Messages         []claude.ClaudeMessage `json:"messages"`
 	MaxTokens        int                    `json:"max_tokens,omitempty"`
 	Temperature      float64                `json:"temperature,omitempty"`

relay/channel/claude/dto.go

@@ -24,15 +24,16 @@ type ClaudeMessage struct {
 }
 
 type ClaudeRequest struct {
-	Model         string          `json:"model"`
-	Prompt        string          `json:"prompt,omitempty"`
-	System        string          `json:"system,omitempty"`
-	Messages      []ClaudeMessage `json:"messages,omitempty"`
-	MaxTokens     uint            `json:"max_tokens,omitempty"`
-	StopSequences []string        `json:"stop_sequences,omitempty"`
-	Temperature   float64         `json:"temperature,omitempty"`
-	TopP          float64         `json:"top_p,omitempty"`
-	TopK          int             `json:"top_k,omitempty"`
+	Model             string          `json:"model"`
+	Prompt            string          `json:"prompt,omitempty"`
+	System            string          `json:"system,omitempty"`
+	Messages          []ClaudeMessage `json:"messages,omitempty"`
+	MaxTokens         uint            `json:"max_tokens,omitempty"`
+	MaxTokensToSample uint            `json:"max_tokens_to_sample,omitempty"`
+	StopSequences     []string        `json:"stop_sequences,omitempty"`
+	Temperature       float64         `json:"temperature,omitempty"`
+	TopP              float64         `json:"top_p,omitempty"`
+	TopK              int             `json:"top_k,omitempty"`
 	//ClaudeMetadata    `json:"metadata,omitempty"`
 	Stream bool `json:"stream,omitempty"`
 }

relay/channel/claude/relay-claude.go

@@ -30,15 +30,14 @@ func RequestOpenAI2ClaudeComplete(textRequest dto.GeneralOpenAIRequest) *ClaudeR
 	claudeRequest := ClaudeRequest{
 		Model:         textRequest.Model,
 		Prompt:        "",
-		MaxTokens:     textRequest.MaxTokens,
 		StopSequences: nil,
 		Temperature:   textRequest.Temperature,
 		TopP:          textRequest.TopP,
 		TopK:          textRequest.TopK,
 		Stream:        textRequest.Stream,
 	}
-	if claudeRequest.MaxTokens == 0 {
-		claudeRequest.MaxTokens = 4096
+	if claudeRequest.MaxTokensToSample == 0 {
+		claudeRequest.MaxTokensToSample = 4096
 	}
 	prompt := ""
 	for _, message := range textRequest.Messages {

relay/channel/openai/relay-openai.go

@@ -50,7 +50,7 @@ func OpenaiStreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*d
 			if data[:6] != "data: " && data[:6] != "[DONE]" {
 				continue
 			}
-			dataChan <- data
+			common.SafeSendString(dataChan, data)
 			data = data[6:]
 			if !strings.HasPrefix(data, "[DONE]") {
 				streamItems = append(streamItems, data)
@@ -123,7 +123,7 @@ func OpenaiStreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*d
 			// wait data out
 			time.Sleep(2 * time.Second)
 		}
-		common.SafeSend(stopChan, true)
+		common.SafeSendBool(stopChan, true)
 	}()
 	service.SetEventStreamHeaders(c)
 	c.Stream(func(w io.Writer) bool {

relay/relay-text.go

@@ -264,10 +264,10 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRe
 	completionTokens := usage.CompletionTokens
 
 	tokenName := ctx.GetString("token_name")
+	completionRatio := common.GetCompletionRatio(textRequest.Model)
 
 	quota := 0
 	if modelPrice == -1 {
-		completionRatio := common.GetCompletionRatio(textRequest.Model)
 		quota = promptTokens + int(float64(completionTokens)*completionRatio)
 		quota = int(float64(quota) * ratio)
 		if ratio != 0 && quota <= 0 {
@@ -279,7 +279,7 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, textRe
 	totalTokens := promptTokens + completionTokens
 	var logContent string
 	if modelPrice == -1 {
-		logContent = fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f", modelRatio, groupRatio)
+		logContent = fmt.Sprintf("模型倍率 %.2f，分组倍率 %.2f，补全倍率 %.2f", modelRatio, groupRatio, completionRatio)
 	} else {
 		logContent = fmt.Sprintf("模型价格 %.2f，分组倍率 %.2f", modelPrice, groupRatio)
 	}