feat(user): simplify user response structure in JSON output

fix: revert model ratio

controller/user.go

@@ -110,18 +110,17 @@ func setupLogin(user *model.User, c *gin.Context) {
 		})
 		return
 	}
-	cleanUser := model.User{
-		Id:          user.Id,
-		Username:    user.Username,
-		DisplayName: user.DisplayName,
-		Role:        user.Role,
-		Status:      user.Status,
-		Group:       user.Group,
-	}
 	c.JSON(http.StatusOK, gin.H{
 		"message": "",
 		"success": true,
-		"data":    cleanUser,
+		"data": map[string]any{
+			"id":           user.Id,
+			"username":     user.Username,
+			"display_name": user.DisplayName,
+			"role":         user.Role,
+			"status":       user.Status,
+			"group":        user.Group,
+		},
 	})
 }
 

relay/channel/claude/relay-claude.go

@@ -483,9 +483,11 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *dto.ClaudeResponse
 				}
 			}
 		} else if claudeResponse.Type == "message_delta" {
-			finishReason := stopReasonClaude2OpenAI(*claudeResponse.Delta.StopReason)
-			if finishReason != "null" {
-				choice.FinishReason = &finishReason
+			if claudeResponse.Delta != nil && claudeResponse.Delta.StopReason != nil {
+				finishReason := stopReasonClaude2OpenAI(*claudeResponse.Delta.StopReason)
+				if finishReason != "null" {
+					choice.FinishReason = &finishReason
+				}
 			}
 			//claudeUsage = &claudeResponse.Usage
 		} else if claudeResponse.Type == "message_stop" {

relay/channel/openai/relay-openai.go

@@ -596,7 +596,7 @@ func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, res
 		if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
 			usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
 		}
-	case constant.ChannelTypeZhipu_v4:
+	case constant.ChannelTypeZhipu_v4, constant.ChannelTypeMoonshot:
 		if usage.PromptTokensDetails.CachedTokens == 0 {
 			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
 				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens

relay/compatible_handler.go

@@ -300,14 +300,20 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage
 	if !relayInfo.PriceData.UsePrice {
 		baseTokens := dPromptTokens
 		// 减去 cached tokens
+		// Anthropic API 的 input_tokens 已经不包含缓存 tokens，不需要减去
+		// OpenAI/OpenRouter 等 API 的 prompt_tokens 包含缓存 tokens，需要减去
 		var cachedTokensWithRatio decimal.Decimal
 		if !dCacheTokens.IsZero() {
-			baseTokens = baseTokens.Sub(dCacheTokens)
+			if relayInfo.ChannelType != constant.ChannelTypeAnthropic {
+				baseTokens = baseTokens.Sub(dCacheTokens)
+			}
 			cachedTokensWithRatio = dCacheTokens.Mul(dCacheRatio)
 		}
 		var dCachedCreationTokensWithRatio decimal.Decimal
 		if !dCachedCreationTokens.IsZero() {
-			baseTokens = baseTokens.Sub(dCachedCreationTokens)
+			if relayInfo.ChannelType != constant.ChannelTypeAnthropic {
+				baseTokens = baseTokens.Sub(dCachedCreationTokens)
+			}
 			dCachedCreationTokensWithRatio = dCachedCreationTokens.Mul(dCachedCreationRatio)
 		}
 

service/convert.go

@@ -389,25 +389,29 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
 				}
 
 				idx := blockIndex
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_start",
-					ContentBlock: &dto.ClaudeMediaMessage{
-						Id:    toolCall.ID,
-						Type:  "tool_use",
-						Name:  toolCall.Function.Name,
-						Input: map[string]interface{}{},
-					},
-				})
-
-				claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
-					Index: &idx,
-					Type:  "content_block_delta",
-					Delta: &dto.ClaudeMediaMessage{
-						Type:        "input_json_delta",
-						PartialJson: &toolCall.Function.Arguments,
-					},
-				})
+				if toolCall.Function.Name != "" {
+					claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+						Index: &idx,
+						Type:  "content_block_start",
+						ContentBlock: &dto.ClaudeMediaMessage{
+							Id:    toolCall.ID,
+							Type:  "tool_use",
+							Name:  toolCall.Function.Name,
+							Input: map[string]interface{}{},
+						},
+					})
+				}
+				
+				if len(toolCall.Function.Arguments) > 0 {
+					claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
+						Index: &idx,
+						Type:  "content_block_delta",
+						Delta: &dto.ClaudeMediaMessage{
+							Type:        "input_json_delta",
+							PartialJson: &toolCall.Function.Arguments,
+						},
+					})
+				}
 
 				info.ClaudeConvertInfo.Index = blockIndex
 			}

setting/ratio_setting/model_ratio.go

@@ -7,7 +7,6 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/setting/operation_setting"
-	"github.com/QuantumNous/new-api/setting/reasoning"
 )
 
 // from songquanpeng/one-api
@@ -829,10 +828,6 @@ func FormatMatchingModelName(name string) string {
 		name = handleThinkingBudgetModel(name, "gemini-2.5-pro", "gemini-2.5-pro-thinking-*")
 	}
 
-	if base, _, ok := reasoning.TrimEffortSuffix(name); ok {
-		name = base
-	}
-
 	if strings.HasPrefix(name, "gpt-4-gizmo") {
 		name = "gpt-4-gizmo-*"
 	}