refactor: update GeminiThinkingConfig initialization

relay/channel/gemini/relay-gemini.go

@@ -28,9 +28,6 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest, info *relaycommon
 			TopP:            textRequest.TopP,
 			MaxOutputTokens: textRequest.MaxTokens,
 			Seed:            int64(textRequest.Seed),
-			ThinkingConfig: &GeminiThinkingConfig{
-				IncludeThoughts: true,
-			},
 		},
 	}
 
@@ -47,9 +44,14 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest, info *relaycommon
 			if budgetTokens == 0 || budgetTokens > 24576 {
 				budgetTokens = 24576
 			}
-			geminiRequest.GenerationConfig.ThinkingConfig.SetThinkingBudget(int(budgetTokens))
+			geminiRequest.GenerationConfig.ThinkingConfig = &GeminiThinkingConfig{
+				ThinkingBudget:  common.GetPointer(int(budgetTokens)),
+				IncludeThoughts: true,
+			}
 		} else if strings.HasSuffix(info.OriginModelName, "-nothinking") {
-			geminiRequest.GenerationConfig.ThinkingConfig.SetThinkingBudget(0)
+			geminiRequest.GenerationConfig.ThinkingConfig = &GeminiThinkingConfig{
+				ThinkingBudget: common.GetPointer(0),
+			}
 		}
 	}
 

relay/channel/siliconflow/adaptor.go

@@ -74,13 +74,9 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	switch info.RelayMode {
 	case constant.RelayModeRerank:
 		err, usage = siliconflowRerankHandler(c, resp)
-	case constant.RelayModeChatCompletions:
-		if info.IsStream {
-			err, usage = openai.OaiStreamHandler(c, resp, info)
-		} else {
-			err, usage = openai.OpenaiHandler(c, resp, info)
-		}
 	case constant.RelayModeCompletions:
+		fallthrough
+	case constant.RelayModeChatCompletions:
 		if info.IsStream {
 			err, usage = openai.OaiStreamHandler(c, resp, info)
 		} else {