feat: enhance token usage details for upstream OpenRouter

dto/openai_request.go

@@ -28,7 +28,6 @@ type GeneralOpenAIRequest struct {
 	MaxTokens           uint              `json:"max_tokens,omitempty"`
 	MaxCompletionTokens uint              `json:"max_completion_tokens,omitempty"`
 	ReasoningEffort     string            `json:"reasoning_effort,omitempty"`
-	//Reasoning           json.RawMessage   `json:"reasoning,omitempty"`
 	Temperature         *float64          `json:"temperature,omitempty"`
 	TopP                float64           `json:"top_p,omitempty"`
 	TopK                int               `json:"top_k,omitempty"`
@@ -55,6 +54,9 @@ type GeneralOpenAIRequest struct {
 	EnableThinking      any               `json:"enable_thinking,omitempty"` // ali
 	ExtraBody           any               `json:"extra_body,omitempty"`
 	WebSearchOptions    *WebSearchOptions `json:"web_search_options,omitempty"`
+	// OpenRouter Params
+	Usage     json.RawMessage `json:"usage,omitempty"`
+	Reasoning json.RawMessage `json:"reasoning,omitempty"`
 }
 
 type ToolCallRequest struct {

relay/channel/openai/adaptor.go

@@ -152,6 +152,9 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 	if info.ChannelType != common.ChannelTypeOpenAI && info.ChannelType != common.ChannelTypeAzure {
 		request.StreamOptions = nil
 	}
+	if info.ChannelType == common.ChannelTypeOpenRouter {
+		request.Usage = json.RawMessage("{\"include\": true}")
+	}
 	if strings.HasPrefix(request.Model, "o") {
 		if request.MaxCompletionTokens == 0 && request.MaxTokens != 0 {
 			request.MaxCompletionTokens = request.MaxTokens

service/convert.go

@@ -246,12 +246,15 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 		}
 		if info.Done {
 			claudeResponses = append(claudeResponses, generateStopBlock(info.ClaudeConvertInfo.Index))
-			if info.ClaudeConvertInfo.Usage != nil {
+			oaiUsage := info.ClaudeConvertInfo.Usage
+			if oaiUsage != nil {
 				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
 					Type: "message_delta",
 					Usage: &dto.ClaudeUsage{
-						InputTokens:  info.ClaudeConvertInfo.Usage.PromptTokens,
-						OutputTokens: info.ClaudeConvertInfo.Usage.CompletionTokens,
+						InputTokens:              oaiUsage.PromptTokens,
+						OutputTokens:             oaiUsage.CompletionTokens,
+						CacheCreationInputTokens: oaiUsage.PromptTokensDetails.CachedCreationTokens,
+						CacheReadInputTokens:     oaiUsage.PromptTokensDetails.CachedTokens,
 					},
 					Delta: &dto.ClaudeMediaMessage{
 						StopReason: common.GetPointer[string](stopReasonOpenAI2Claude(info.FinishReason)),