Merge branch 'dev' into feat-amp-cli-module

internal/api/handlers/management/auth_files.go

@@ -1443,6 +1443,87 @@ func (h *Handler) RequestIFlowToken(c *gin.Context) {
 	c.JSON(http.StatusOK, gin.H{"status": "ok", "url": authURL, "state": state})
 }
 
+func (h *Handler) RequestIFlowCookieToken(c *gin.Context) {
+	ctx := context.Background()
+
+	var payload struct {
+		Cookie string `json:"cookie"`
+	}
+	if err := c.ShouldBindJSON(&payload); err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "cookie is required"})
+		return
+	}
+
+	cookieValue := strings.TrimSpace(payload.Cookie)
+
+	if cookieValue == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "cookie is required"})
+		return
+	}
+
+	cookieValue, errNormalize := iflowauth.NormalizeCookie(cookieValue)
+	if errNormalize != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": errNormalize.Error()})
+		return
+	}
+
+	authSvc := iflowauth.NewIFlowAuth(h.cfg)
+	tokenData, errAuth := authSvc.AuthenticateWithCookie(ctx, cookieValue)
+	if errAuth != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": errAuth.Error()})
+		return
+	}
+
+	tokenData.Cookie = cookieValue
+
+	tokenStorage := authSvc.CreateCookieTokenStorage(tokenData)
+	email := strings.TrimSpace(tokenStorage.Email)
+	if email == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "failed to extract email from token"})
+		return
+	}
+
+	fileName := iflowauth.SanitizeIFlowFileName(email)
+	if fileName == "" {
+		fileName = fmt.Sprintf("iflow-%d", time.Now().UnixMilli())
+	}
+
+	tokenStorage.Email = email
+
+	record := &coreauth.Auth{
+		ID:       fmt.Sprintf("iflow-%s.json", fileName),
+		Provider: "iflow",
+		FileName: fmt.Sprintf("iflow-%s.json", fileName),
+		Storage:  tokenStorage,
+		Metadata: map[string]any{
+			"email":        email,
+			"api_key":      tokenStorage.APIKey,
+			"expired":      tokenStorage.Expire,
+			"cookie":       tokenStorage.Cookie,
+			"type":         tokenStorage.Type,
+			"last_refresh": tokenStorage.LastRefresh,
+		},
+		Attributes: map[string]string{
+			"api_key": tokenStorage.APIKey,
+		},
+	}
+
+	savedPath, errSave := h.saveTokenRecord(ctx, record)
+	if errSave != nil {
+		c.JSON(http.StatusInternalServerError, gin.H{"status": "error", "error": "failed to save authentication tokens"})
+		return
+	}
+
+	fmt.Printf("iFlow cookie authentication successful. Token saved to %s\n", savedPath)
+	c.JSON(http.StatusOK, gin.H{
+		"status":     "ok",
+		"saved_path": savedPath,
+		"email":      email,
+		"expired":    tokenStorage.Expire,
+		"type":       tokenStorage.Type,
+	})
+}
+
 type projectSelectionRequiredError struct{}
 
 func (e *projectSelectionRequiredError) Error() string {

internal/api/server.go

@@ -534,6 +534,7 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.GET("/gemini-cli-auth-url", s.mgmt.RequestGeminiCLIToken)
 		mgmt.GET("/qwen-auth-url", s.mgmt.RequestQwenToken)
 		mgmt.GET("/iflow-auth-url", s.mgmt.RequestIFlowToken)
+		mgmt.POST("/iflow-auth-url", s.mgmt.RequestIFlowCookieToken)
 		mgmt.GET("/get-auth-status", s.mgmt.GetAuthStatus)
 	}
 }

internal/auth/iflow/cookie_helpers.go

@@ -0,0 +1,38 @@
+package iflow
+
+import (
+	"fmt"
+	"strings"
+)
+
+// NormalizeCookie normalizes raw cookie strings for iFlow authentication flows.
+func NormalizeCookie(raw string) (string, error) {
+	trimmed := strings.TrimSpace(raw)
+	if trimmed == "" {
+		return "", fmt.Errorf("cookie cannot be empty")
+	}
+
+	combined := strings.Join(strings.Fields(trimmed), " ")
+	if !strings.HasSuffix(combined, ";") {
+		combined += ";"
+	}
+	if !strings.Contains(combined, "BXAuth=") {
+		return "", fmt.Errorf("cookie missing BXAuth field")
+	}
+	return combined, nil
+}
+
+// SanitizeIFlowFileName normalizes user identifiers for safe filename usage.
+func SanitizeIFlowFileName(raw string) string {
+	if raw == "" {
+		return ""
+	}
+	cleanEmail := strings.ReplaceAll(raw, "*", "x")
+	var result strings.Builder
+	for _, r := range cleanEmail {
+		if (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') || (r >= '0' && r <= '9') || r == '_' || r == '@' || r == '.' || r == '-' {
+			result.WriteRune(r)
+		}
+	}
+	return strings.TrimSpace(result.String())
+}

internal/cmd/iflow_cookie.go

@@ -71,22 +71,9 @@ func promptForCookie(promptFn func(string) (string, error)) (string, error) {
 		return "", fmt.Errorf("failed to read cookie: %w", err)
 	}
 
-	line = strings.TrimSpace(line)
+	cookie, err := iflow.NormalizeCookie(line)
-	if line == "" {
+	if err != nil {
-		return "", fmt.Errorf("cookie cannot be empty")
+		return "", err
-	}
-
-	// Clean up any extra whitespace and join multiple spaces
-	cookie := strings.Join(strings.Fields(line), " ")
-
-	// Ensure it ends properly
-	if !strings.HasSuffix(cookie, ";") {
-		cookie = cookie + ";"
-	}
-
-	// Ensure BXAuth is present in the cookie
-	if !strings.Contains(cookie, "BXAuth=") {
-		return "", fmt.Errorf("BXAuth field not found in cookie")
 	}
 
 	return cookie, nil
@@ -94,17 +81,6 @@ func promptForCookie(promptFn func(string) (string, error)) (string, error) {
 
 // getAuthFilePath returns the auth file path for the given provider and email
 func getAuthFilePath(cfg *config.Config, provider, email string) string {
-	// Clean email to make it filename-safe
+	fileName := iflow.SanitizeIFlowFileName(email)
-	cleanEmail := strings.ReplaceAll(email, "*", "x")
+	return fmt.Sprintf("%s/%s-%s.json", cfg.AuthDir, provider, fileName)
-
-	// Remove any unsafe characters, but allow standard email chars (@, ., -)
-	var result strings.Builder
-	for _, r := range cleanEmail {
-		if (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') || (r >= '0' && r <= '9') ||
-			r == '_' || r == '@' || r == '.' || r == '-' {
-			result.WriteRune(r)
-		}
-	}
-
-	return fmt.Sprintf("%s/%s-%s.json", cfg.AuthDir, provider, result.String())
 }

internal/registry/model_definitions.go

@@ -68,82 +68,13 @@ func GetClaudeModels() []*ModelInfo {
 	}
 }
 
-// GeminiModels returns the shared base Gemini model set used by multiple providers.
-func GeminiModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    time.Now().Unix(),
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-pro",
-			Object:                     "model",
-			Created:                    time.Now().Unix(),
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-pro",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Pro",
-			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-lite",
-			Object:                     "model",
-			Created:                    time.Now().Unix(),
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-lite",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Lite",
-			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-	}
-}
-
 // GetGeminiModels returns the standard Gemini model definitions
-func GetGeminiModels() []*ModelInfo { return GeminiModels() }
+func GetGeminiModels() []*ModelInfo {
-
-// GetGeminiCLIModels returns the standard Gemini model definitions
-func GetGeminiCLIModels() []*ModelInfo {
 	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    time.Now().Unix(),
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
 		{
 			ID:                         "gemini-2.5-pro",
 			Object:                     "model",
-			Created:                    time.Now().Unix(),
+			Created:                    1750118400,
 			OwnedBy:                    "google",
 			Type:                       "gemini",
 			Name:                       "models/gemini-2.5-pro",
@@ -155,10 +86,25 @@ func GetGeminiCLIModels() []*ModelInfo {
 			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
 			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
 		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
 		{
 			ID:                         "gemini-2.5-flash-lite",
 			Object:                     "model",
-			Created:                    time.Now().Unix(),
+			Created:                    1753142400,
 			OwnedBy:                    "google",
 			Type:                       "gemini",
 			Name:                       "models/gemini-2.5-flash-lite",
@@ -173,7 +119,152 @@ func GetGeminiCLIModels() []*ModelInfo {
 		{
 			ID:                         "gemini-3-pro-preview",
 			Object:                     "model",
-			Created:                    time.Now().Unix(),
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Gemini 3 Pro Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+	}
+}
+
+func GetGeminiVertexModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Gemini 3 Pro Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-image-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-image-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Image Preview",
+			Description:                "Gemini 3 Pro Image Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		},
+	}
+}
+
+// GetGeminiCLIModels returns the standard Gemini model definitions
+func GetGeminiCLIModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
 			OwnedBy:                    "google",
 			Type:                       "gemini",
 			Name:                       "models/gemini-3-pro-preview",
@@ -190,102 +281,143 @@ func GetGeminiCLIModels() []*ModelInfo {
 
 // GetAIStudioModels returns the Gemini model definitions for AI Studio integrations
 func GetAIStudioModels() []*ModelInfo {
-	base := GeminiModels()
+	return []*ModelInfo{
-
+		{
-	return append(base,
+			ID:                         "gemini-2.5-pro",
-		[]*ModelInfo{
+			Object:                     "model",
-			{
+			Created:                    1750118400,
-				ID:                         "gemini-3-pro-preview",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-2.5-pro",
-				OwnedBy:                    "google",
+			Version:                    "2.5",
-				Type:                       "gemini",
+			DisplayName:                "Gemini 2.5 Pro",
-				Name:                       "models/gemini-3-pro-preview",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-				Version:                    "3.0",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini 3 Pro Preview",
+			OutputTokenLimit:           65536,
-				Description:                "Gemini 3 Pro Preview",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-				OutputTokenLimit:           65536,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+			ID:                         "gemini-2.5-flash",
-			},
+			Object:                     "model",
-			{
+			Created:                    1750118400,
-				ID:                         "gemini-pro-latest",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-2.5-flash",
-				OwnedBy:                    "google",
+			Version:                    "001",
-				Type:                       "gemini",
+			DisplayName:                "Gemini 2.5 Flash",
-				Name:                       "models/gemini-pro-latest",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-				Version:                    "2.5",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini Pro Latest",
+			OutputTokenLimit:           65536,
-				Description:                "Latest release of Gemini Pro",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-				OutputTokenLimit:           65536,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+			ID:                         "gemini-2.5-flash-lite",
-			},
+			Object:                     "model",
-			{
+			Created:                    1753142400,
-				ID:                         "gemini-flash-latest",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-2.5-flash-lite",
-				OwnedBy:                    "google",
+			Version:                    "2.5",
-				Type:                       "gemini",
+			DisplayName:                "Gemini 2.5 Flash Lite",
-				Name:                       "models/gemini-flash-latest",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-				Version:                    "2.5",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini Flash Latest",
+			OutputTokenLimit:           65536,
-				Description:                "Latest release of Gemini Flash",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-				OutputTokenLimit:           65536,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+			ID:                         "gemini-3-pro-preview",
-			},
+			Object:                     "model",
-			{
+			Created:                    1737158400,
-				ID:                         "gemini-flash-lite-latest",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-3-pro-preview",
-				OwnedBy:                    "google",
+			Version:                    "3.0",
-				Type:                       "gemini",
+			DisplayName:                "Gemini 3 Pro Preview",
-				Name:                       "models/gemini-flash-lite-latest",
+			Description:                "Gemini 3 Pro Preview",
-				Version:                    "2.5",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini Flash-Lite Latest",
+			OutputTokenLimit:           65536,
-				Description:                "Latest release of Gemini Flash-Lite",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-				OutputTokenLimit:           65536,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				Thinking:                   &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+			ID:                         "gemini-pro-latest",
-			},
+			Object:                     "model",
-			{
+			Created:                    1750118400,
-				ID:                         "gemini-2.5-flash-image-preview",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-pro-latest",
-				OwnedBy:                    "google",
+			Version:                    "2.5",
-				Type:                       "gemini",
+			DisplayName:                "Gemini Pro Latest",
-				Name:                       "models/gemini-2.5-flash-image-preview",
+			Description:                "Latest release of Gemini Pro",
-				Version:                    "2.5",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini 2.5 Flash Image Preview",
+			OutputTokenLimit:           65536,
-				Description:                "State-of-the-art image generation and editing model.",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-				OutputTokenLimit:           8192,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				// image models don't support thinkingConfig; leave Thinking nil
+			ID:                         "gemini-flash-latest",
-			},
+			Object:                     "model",
-			{
+			Created:                    1750118400,
-				ID:                         "gemini-2.5-flash-image",
+			OwnedBy:                    "google",
-				Object:                     "model",
+			Type:                       "gemini",
-				Created:                    time.Now().Unix(),
+			Name:                       "models/gemini-flash-latest",
-				OwnedBy:                    "google",
+			Version:                    "2.5",
-				Type:                       "gemini",
+			DisplayName:                "Gemini Flash Latest",
-				Name:                       "models/gemini-2.5-flash-image",
+			Description:                "Latest release of Gemini Flash",
-				Version:                    "2.5",
+			InputTokenLimit:            1048576,
-				DisplayName:                "Gemini 2.5 Flash Image",
+			OutputTokenLimit:           65536,
-				Description:                "State-of-the-art image generation and editing model.",
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-				InputTokenLimit:            1048576,
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-				OutputTokenLimit:           8192,
+		},
-				SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		{
-				// image models don't support thinkingConfig; leave Thinking nil
+			ID:                         "gemini-flash-lite-latest",
-			},
+			Object:                     "model",
-		}...,
+			Created:                    1753142400,
-	)
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-flash-lite-latest",
+			Version:                    "2.5",
+			DisplayName:                "Gemini Flash-Lite Latest",
+			Description:                "Latest release of Gemini Flash-Lite",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-image-preview",
+			Object:                     "model",
+			Created:                    1756166400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-image-preview",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Image Preview",
+			Description:                "State-of-the-art image generation and editing model.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           8192,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			// image models don't support thinkingConfig; leave Thinking nil
+		},
+		{
+			ID:                         "gemini-2.5-flash-image",
+			Object:                     "model",
+			Created:                    1759363200,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-image",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Image",
+			Description:                "State-of-the-art image generation and editing model.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           8192,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			// image models don't support thinkingConfig; leave Thinking nil
+		},
+	}
 }
 
 // GetOpenAIModels returns the standard OpenAI model definitions

internal/runtime/executor/gemini_cli_executor.go

@@ -99,124 +99,89 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
 	var lastStatus int
 	var lastBody []byte
 
-	// Get max retry count from config, default to 3 if not set
-	maxRetries := e.cfg.RequestRetry
-	if maxRetries < 0 {
-		maxRetries = 3
-	}
-
 	for idx, attemptModel := range models {
-		retryCount := 0
+		payload := append([]byte(nil), basePayload...)
-		for {
+		if action == "countTokens" {
-			payload := append([]byte(nil), basePayload...)
+			payload = deleteJSONField(payload, "project")
-			if action == "countTokens" {
+			payload = deleteJSONField(payload, "model")
-				payload = deleteJSONField(payload, "project")
+		} else {
-				payload = deleteJSONField(payload, "model")
+			payload = setJSONField(payload, "project", projectID)
-			} else {
+			payload = setJSONField(payload, "model", attemptModel)
-				payload = setJSONField(payload, "project", projectID)
-				payload = setJSONField(payload, "model", attemptModel)
-			}
-
-			tok, errTok := tokenSource.Token()
-			if errTok != nil {
-				err = errTok
-				return resp, err
-			}
-			updateGeminiCLITokenMetadata(auth, baseTokenData, tok)
-
-			url := fmt.Sprintf("%s/%s:%s", codeAssistEndpoint, codeAssistVersion, action)
-			if opts.Alt != "" && action != "countTokens" {
-				url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
-			}
-
-			reqHTTP, errReq := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(payload))
-			if errReq != nil {
-				err = errReq
-				return resp, err
-			}
-			reqHTTP.Header.Set("Content-Type", "application/json")
-			reqHTTP.Header.Set("Authorization", "Bearer "+tok.AccessToken)
-			applyGeminiCLIHeaders(reqHTTP)
-			reqHTTP.Header.Set("Accept", "application/json")
-			recordAPIRequest(ctx, e.cfg, upstreamRequestLog{
-				URL:       url,
-				Method:    http.MethodPost,
-				Headers:   reqHTTP.Header.Clone(),
-				Body:      payload,
-				Provider:  e.Identifier(),
-				AuthID:    authID,
-				AuthLabel: authLabel,
-				AuthType:  authType,
-				AuthValue: authValue,
-			})
-
-			httpResp, errDo := httpClient.Do(reqHTTP)
-			if errDo != nil {
-				recordAPIResponseError(ctx, e.cfg, errDo)
-				err = errDo
-				return resp, err
-			}
-
-			data, errRead := io.ReadAll(httpResp.Body)
-			if errClose := httpResp.Body.Close(); errClose != nil {
-				log.Errorf("gemini cli executor: close response body error: %v", errClose)
-			}
-			recordAPIResponseMetadata(ctx, e.cfg, httpResp.StatusCode, httpResp.Header.Clone())
-			if errRead != nil {
-				recordAPIResponseError(ctx, e.cfg, errRead)
-				err = errRead
-				return resp, err
-			}
-			appendAPIResponseChunk(ctx, e.cfg, data)
-			if httpResp.StatusCode >= 200 && httpResp.StatusCode < 300 {
-				reporter.publish(ctx, parseGeminiCLIUsage(data))
-				var param any
-				out := sdktranslator.TranslateNonStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), payload, data, &param)
-				resp = cliproxyexecutor.Response{Payload: []byte(out)}
-				return resp, nil
-			}
-
-			lastStatus = httpResp.StatusCode
-			lastBody = append([]byte(nil), data...)
-			log.Debugf("request error, error status: %d, error body: %s", httpResp.StatusCode, summarizeErrorBody(httpResp.Header.Get("Content-Type"), data))
-
-			// Handle 429 rate limit errors with retry
-			if httpResp.StatusCode == 429 {
-				if retryCount < maxRetries {
-					// Parse retry delay from Google's response
-					retryDelay := parseRetryDelay(data)
-					log.Infof("gemini cli executor: rate limited (429), retrying model %s in %v (retry %d/%d)", attemptModel, retryDelay, retryCount+1, maxRetries)
-					retryCount++
-
-					// Wait for the specified delay
-					select {
-					case <-time.After(retryDelay):
-						// Continue to next retry iteration
-						continue
-					case <-ctx.Done():
-						// Context cancelled, return immediately
-						err = ctx.Err()
-						return resp, err
-					}
-				} else {
-					// Exhausted retries for this model, try next model if available
-					if idx+1 < len(models) {
-						log.Infof("gemini cli executor: rate limited, exhausted %d retries for model %s, trying fallback model: %s", maxRetries, attemptModel, models[idx+1])
-						break // Break inner loop to try next model
-					} else {
-						log.Infof("gemini cli executor: rate limited, exhausted %d retries for model %s, no additional fallback model", maxRetries, attemptModel)
-						// No more models to try, will return error below
-					}
-				}
-			} else {
-				// Non-429 error, don't retry this model
-				err = statusErr{code: httpResp.StatusCode, msg: string(data)}
-				return resp, err
-			}
-
-			// Break inner loop if we hit this point (no retry needed or exhausted retries)
-			break
 		}
+
+		tok, errTok := tokenSource.Token()
+		if errTok != nil {
+			err = errTok
+			return resp, err
+		}
+		updateGeminiCLITokenMetadata(auth, baseTokenData, tok)
+
+		url := fmt.Sprintf("%s/%s:%s", codeAssistEndpoint, codeAssistVersion, action)
+		if opts.Alt != "" && action != "countTokens" {
+			url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+		}
+
+		reqHTTP, errReq := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(payload))
+		if errReq != nil {
+			err = errReq
+			return resp, err
+		}
+		reqHTTP.Header.Set("Content-Type", "application/json")
+		reqHTTP.Header.Set("Authorization", "Bearer "+tok.AccessToken)
+		applyGeminiCLIHeaders(reqHTTP)
+		reqHTTP.Header.Set("Accept", "application/json")
+		recordAPIRequest(ctx, e.cfg, upstreamRequestLog{
+			URL:       url,
+			Method:    http.MethodPost,
+			Headers:   reqHTTP.Header.Clone(),
+			Body:      payload,
+			Provider:  e.Identifier(),
+			AuthID:    authID,
+			AuthLabel: authLabel,
+			AuthType:  authType,
+			AuthValue: authValue,
+		})
+
+		httpResp, errDo := httpClient.Do(reqHTTP)
+		if errDo != nil {
+			recordAPIResponseError(ctx, e.cfg, errDo)
+			err = errDo
+			return resp, err
+		}
+
+		data, errRead := io.ReadAll(httpResp.Body)
+		if errClose := httpResp.Body.Close(); errClose != nil {
+			log.Errorf("gemini cli executor: close response body error: %v", errClose)
+		}
+		recordAPIResponseMetadata(ctx, e.cfg, httpResp.StatusCode, httpResp.Header.Clone())
+		if errRead != nil {
+			recordAPIResponseError(ctx, e.cfg, errRead)
+			err = errRead
+			return resp, err
+		}
+		appendAPIResponseChunk(ctx, e.cfg, data)
+		if httpResp.StatusCode >= 200 && httpResp.StatusCode < 300 {
+			reporter.publish(ctx, parseGeminiCLIUsage(data))
+			var param any
+			out := sdktranslator.TranslateNonStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), payload, data, &param)
+			resp = cliproxyexecutor.Response{Payload: []byte(out)}
+			return resp, nil
+		}
+
+		lastStatus = httpResp.StatusCode
+		lastBody = append([]byte(nil), data...)
+		log.Debugf("request error, error status: %d, error body: %s", httpResp.StatusCode, summarizeErrorBody(httpResp.Header.Get("Content-Type"), data))
+		if httpResp.StatusCode == 429 {
+			if idx+1 < len(models) {
+				log.Debugf("gemini cli executor: rate limited, retrying with next model: %s", models[idx+1])
+			} else {
+				log.Debug("gemini cli executor: rate limited, no additional fallback model")
+			}
+			continue
+		}
+
+		err = newGeminiStatusErr(httpResp.StatusCode, data)
+		return resp, err
 	}
 
 	if len(lastBody) > 0 {
@@ -225,7 +190,7 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
 	if lastStatus == 0 {
 		lastStatus = 429
 	}
-	err = statusErr{code: lastStatus, msg: string(lastBody)}
+	err = newGeminiStatusErr(lastStatus, lastBody)
 	return resp, err
 }
 
@@ -270,135 +235,77 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
 	var lastStatus int
 	var lastBody []byte
 
-	// Get max retry count from config, default to 3 if not set
-	maxRetries := e.cfg.RequestRetry
-	if maxRetries < 0 {
-		maxRetries = 3
-	}
-
 	for idx, attemptModel := range models {
-		var httpResp *http.Response
+		payload := append([]byte(nil), basePayload...)
-		var payload []byte
+		payload = setJSONField(payload, "project", projectID)
-		var errDo error
+		payload = setJSONField(payload, "model", attemptModel)
-		shouldContinueToNextModel := false
 
-		retryCount := 0
+		tok, errTok := tokenSource.Token()
-		// Inner retry loop for 429 errors on the same model
+		if errTok != nil {
-		for {
+			err = errTok
-			payload = append([]byte(nil), basePayload...)
+			return nil, err
-			payload = setJSONField(payload, "project", projectID)
+		}
-			payload = setJSONField(payload, "model", attemptModel)
+		updateGeminiCLITokenMetadata(auth, baseTokenData, tok)
 
-			tok, errTok := tokenSource.Token()
+		url := fmt.Sprintf("%s/%s:%s", codeAssistEndpoint, codeAssistVersion, "streamGenerateContent")
-			if errTok != nil {
+		if opts.Alt == "" {
-				err = errTok
+			url = url + "?alt=sse"
+		} else {
+			url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+		}
+
+		reqHTTP, errReq := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(payload))
+		if errReq != nil {
+			err = errReq
+			return nil, err
+		}
+		reqHTTP.Header.Set("Content-Type", "application/json")
+		reqHTTP.Header.Set("Authorization", "Bearer "+tok.AccessToken)
+		applyGeminiCLIHeaders(reqHTTP)
+		reqHTTP.Header.Set("Accept", "text/event-stream")
+		recordAPIRequest(ctx, e.cfg, upstreamRequestLog{
+			URL:       url,
+			Method:    http.MethodPost,
+			Headers:   reqHTTP.Header.Clone(),
+			Body:      payload,
+			Provider:  e.Identifier(),
+			AuthID:    authID,
+			AuthLabel: authLabel,
+			AuthType:  authType,
+			AuthValue: authValue,
+		})
+
+		httpResp, errDo := httpClient.Do(reqHTTP)
+		if errDo != nil {
+			recordAPIResponseError(ctx, e.cfg, errDo)
+			err = errDo
+			return nil, err
+		}
+		recordAPIResponseMetadata(ctx, e.cfg, httpResp.StatusCode, httpResp.Header.Clone())
+		if httpResp.StatusCode < 200 || httpResp.StatusCode >= 300 {
+			data, errRead := io.ReadAll(httpResp.Body)
+			if errClose := httpResp.Body.Close(); errClose != nil {
+				log.Errorf("gemini cli executor: close response body error: %v", errClose)
+			}
+			if errRead != nil {
+				recordAPIResponseError(ctx, e.cfg, errRead)
+				err = errRead
 				return nil, err
 			}
-			updateGeminiCLITokenMetadata(auth, baseTokenData, tok)
+			appendAPIResponseChunk(ctx, e.cfg, data)
-
+			lastStatus = httpResp.StatusCode
-			url := fmt.Sprintf("%s/%s:%s", codeAssistEndpoint, codeAssistVersion, "streamGenerateContent")
+			lastBody = append([]byte(nil), data...)
-			if opts.Alt == "" {
+			log.Debugf("request error, error status: %d, error body: %s", httpResp.StatusCode, summarizeErrorBody(httpResp.Header.Get("Content-Type"), data))
-				url = url + "?alt=sse"
+			if httpResp.StatusCode == 429 {
-			} else {
+				if idx+1 < len(models) {
-				url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+					log.Debugf("gemini cli executor: rate limited, retrying with next model: %s", models[idx+1])
-			}
-
-			reqHTTP, errReq := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(payload))
-			if errReq != nil {
-				err = errReq
-				return nil, err
-			}
-			reqHTTP.Header.Set("Content-Type", "application/json")
-			reqHTTP.Header.Set("Authorization", "Bearer "+tok.AccessToken)
-			applyGeminiCLIHeaders(reqHTTP)
-			reqHTTP.Header.Set("Accept", "text/event-stream")
-			recordAPIRequest(ctx, e.cfg, upstreamRequestLog{
-				URL:       url,
-				Method:    http.MethodPost,
-				Headers:   reqHTTP.Header.Clone(),
-				Body:      payload,
-				Provider:  e.Identifier(),
-				AuthID:    authID,
-				AuthLabel: authLabel,
-				AuthType:  authType,
-				AuthValue: authValue,
-			})
-
-			httpResp, errDo = httpClient.Do(reqHTTP)
-			if errDo != nil {
-				recordAPIResponseError(ctx, e.cfg, errDo)
-				err = errDo
-				return nil, err
-			}
-			recordAPIResponseMetadata(ctx, e.cfg, httpResp.StatusCode, httpResp.Header.Clone())
-			if httpResp.StatusCode < 200 || httpResp.StatusCode >= 300 {
-				data, errRead := io.ReadAll(httpResp.Body)
-				if errClose := httpResp.Body.Close(); errClose != nil {
-					log.Errorf("gemini cli executor: close response body error: %v", errClose)
-				}
-				if errRead != nil {
-					recordAPIResponseError(ctx, e.cfg, errRead)
-					err = errRead
-					return nil, err
-				}
-				appendAPIResponseChunk(ctx, e.cfg, data)
-				lastStatus = httpResp.StatusCode
-				lastBody = append([]byte(nil), data...)
-				log.Debugf("request error, error status: %d, error body: %s", httpResp.StatusCode, summarizeErrorBody(httpResp.Header.Get("Content-Type"), data))
-
-				// Handle 429 rate limit errors with retry
-				if httpResp.StatusCode == 429 {
-					if retryCount < maxRetries {
-						// Parse retry delay from Google's response
-						retryDelay := parseRetryDelay(data)
-						log.Infof("gemini cli executor: rate limited (429), retrying stream model %s in %v (retry %d/%d)", attemptModel, retryDelay, retryCount+1, maxRetries)
-						retryCount++
-
-						// Wait for the specified delay
-						select {
-						case <-time.After(retryDelay):
-							// Continue to next retry iteration
-							continue
-						case <-ctx.Done():
-							// Context cancelled, return immediately
-							err = ctx.Err()
-							return nil, err
-						}
-					} else {
-						// Exhausted retries for this model, try next model if available
-						if idx+1 < len(models) {
-							log.Infof("gemini cli executor: rate limited, exhausted %d retries for stream model %s, trying fallback model: %s", maxRetries, attemptModel, models[idx+1])
-							shouldContinueToNextModel = true
-							break // Break inner loop to try next model
-						} else {
-							log.Infof("gemini cli executor: rate limited, exhausted %d retries for stream model %s, no additional fallback model", maxRetries, attemptModel)
-							// No more models to try, will return error below
-						}
-					}
 				} else {
-					// Non-429 error, don't retry this model
+					log.Debug("gemini cli executor: rate limited, no additional fallback model")
-					err = statusErr{code: httpResp.StatusCode, msg: string(data)}
-					return nil, err
 				}
-
+				continue
-				// Break inner loop if we hit this point (no retry needed or exhausted retries)
-				break
 			}
-
+			err = newGeminiStatusErr(httpResp.StatusCode, data)
-			// Success - httpResp.StatusCode is 2xx, break out of retry loop
+			return nil, err
-			// and proceed to streaming logic below
-			break
-		}
-
-		// If we need to try the next fallback model, skip streaming logic
-		if shouldContinueToNextModel {
-			continue
-		}
-
-		// If we have a failed response (non-2xx), don't attempt streaming
-		// Continue outer loop to try next model or return error
-		if httpResp == nil || httpResp.StatusCode < 200 || httpResp.StatusCode >= 300 {
-			continue
 		}
 
 		out := make(chan cliproxyexecutor.StreamChunk)
@@ -470,7 +377,7 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
 	if lastStatus == 0 {
 		lastStatus = 429
 	}
-	err = statusErr{code: lastStatus, msg: string(lastBody)}
+	err = newGeminiStatusErr(lastStatus, lastBody)
 	return nil, err
 }
 
@@ -578,7 +485,7 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
 	if lastStatus == 0 {
 		lastStatus = 429
 	}
-	return cliproxyexecutor.Response{}, statusErr{code: lastStatus, msg: string(lastBody)}
+	return cliproxyexecutor.Response{}, newGeminiStatusErr(lastStatus, lastBody)
 }
 
 func (e *GeminiCLIExecutor) Refresh(ctx context.Context, auth *cliproxyauth.Auth) (*cliproxyauth.Auth, error) {
@@ -863,19 +770,25 @@ func fixGeminiCLIImageAspectRatio(modelName string, rawJSON []byte) []byte {
 	return rawJSON
 }
 
+func newGeminiStatusErr(statusCode int, body []byte) statusErr {
+	err := statusErr{code: statusCode, msg: string(body)}
+	if statusCode == http.StatusTooManyRequests {
+		if retryAfter, parseErr := parseRetryDelay(body); parseErr == nil && retryAfter != nil {
+			err.retryAfter = retryAfter
+		}
+	}
+	return err
+}
+
 // parseRetryDelay extracts the retry delay from a Google API 429 error response.
 // The error response contains a RetryInfo.retryDelay field in the format "0.847655010s".
-// Returns the duration to wait, or a default duration if parsing fails.
+// Returns the parsed duration or an error if it cannot be determined.
-func parseRetryDelay(errorBody []byte) time.Duration {
+func parseRetryDelay(errorBody []byte) (*time.Duration, error) {
-	const defaultDelay = 1 * time.Second
-	const maxDelay = 60 * time.Second
-
 	// Try to parse the retryDelay from the error response
 	// Format: error.details[].retryDelay where @type == "type.googleapis.com/google.rpc.RetryInfo"
 	details := gjson.GetBytes(errorBody, "error.details")
 	if !details.Exists() || !details.IsArray() {
-		log.Debugf("parseRetryDelay: no error.details found, using default delay %v", defaultDelay)
+		return nil, fmt.Errorf("no error.details found")
-		return defaultDelay
 	}
 
 	for _, detail := range details.Array() {
@@ -886,24 +799,12 @@ func parseRetryDelay(errorBody []byte) time.Duration {
 				// Parse duration string like "0.847655010s"
 				duration, err := time.ParseDuration(retryDelay)
 				if err != nil {
-					log.Debugf("parseRetryDelay: failed to parse duration %q: %v, using default", retryDelay, err)
+					return nil, fmt.Errorf("failed to parse duration")
-					return defaultDelay
 				}
-				// Cap at maxDelay to prevent excessive waits
+				return &duration, nil
-				if duration > maxDelay {
-					log.Debugf("parseRetryDelay: capping delay from %v to %v", duration, maxDelay)
-					return maxDelay
-				}
-				if duration < 0 {
-					log.Debugf("parseRetryDelay: negative delay %v, using default", duration)
-					return defaultDelay
-				}
-				log.Debugf("parseRetryDelay: using delay %v from API response", duration)
-				return duration
 			}
 		}
 	}
 
-	log.Debugf("parseRetryDelay: no RetryInfo found, using default delay %v", defaultDelay)
+	return nil, fmt.Errorf("no RetryInfo found")
-	return defaultDelay
 }

internal/runtime/executor/openai_compat_executor.go

@@ -8,6 +8,7 @@ import (
 	"io"
 	"net/http"
 	"strings"
+	"time"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
@@ -340,8 +341,9 @@ func (e *OpenAICompatExecutor) overrideModel(payload []byte, model string) []byt
 }
 
 type statusErr struct {
-	code int
+	code       int
-	msg  string
+	msg        string
+	retryAfter *time.Duration
 }
 
 func (e statusErr) Error() string {
@@ -350,4 +352,5 @@ func (e statusErr) Error() string {
 	}
 	return fmt.Sprintf("status %d", e.code)
 }
-func (e statusErr) StatusCode() int { return e.code }
+func (e statusErr) StatusCode() int            { return e.code }
+func (e statusErr) RetryAfter() *time.Duration { return e.retryAfter }

internal/translator/codex/openai/responses/codex_openai-responses_request.go

@@ -22,6 +22,7 @@ func ConvertOpenAIResponsesRequestToCodex(modelName string, inputRawJSON []byte,
 	rawJSON, _ = sjson.DeleteBytes(rawJSON, "max_completion_tokens")
 	rawJSON, _ = sjson.DeleteBytes(rawJSON, "temperature")
 	rawJSON, _ = sjson.DeleteBytes(rawJSON, "top_p")
+	rawJSON, _ = sjson.DeleteBytes(rawJSON, "service_tier")
 
 	originalInstructions := ""
 	originalInstructionsText := ""

sdk/cliproxy/auth/manager.go

@@ -62,6 +62,8 @@ type Result struct {
 	Model string
 	// Success marks whether the execution succeeded.
 	Success bool
+	// RetryAfter carries a provider supplied retry hint (e.g. 429 retryDelay).
+	RetryAfter *time.Duration
 	// Error describes the failure when Success is false.
 	Error *Error
 }
@@ -325,6 +327,9 @@ func (m *Manager) executeWithProvider(ctx context.Context, provider string, req
 			if errors.As(errExec, &se) && se != nil {
 				result.Error.HTTPStatus = se.StatusCode()
 			}
+			if ra := retryAfterFromError(errExec); ra != nil {
+				result.RetryAfter = ra
+			}
 			m.MarkResult(execCtx, result)
 			lastErr = errExec
 			continue
@@ -370,6 +375,9 @@ func (m *Manager) executeCountWithProvider(ctx context.Context, provider string,
 			if errors.As(errExec, &se) && se != nil {
 				result.Error.HTTPStatus = se.StatusCode()
 			}
+			if ra := retryAfterFromError(errExec); ra != nil {
+				result.RetryAfter = ra
+			}
 			m.MarkResult(execCtx, result)
 			lastErr = errExec
 			continue
@@ -415,6 +423,7 @@ func (m *Manager) executeStreamWithProvider(ctx context.Context, provider string
 				rerr.HTTPStatus = se.StatusCode()
 			}
 			result := Result{AuthID: auth.ID, Provider: provider, Model: req.Model, Success: false, Error: rerr}
+			result.RetryAfter = retryAfterFromError(errStream)
 			m.MarkResult(execCtx, result)
 			lastErr = errStream
 			continue
@@ -556,17 +565,23 @@ func (m *Manager) MarkResult(ctx context.Context, result Result) {
 					suspendReason = "payment_required"
 					shouldSuspendModel = true
 				case 429:
-					cooldown, nextLevel := nextQuotaCooldown(state.Quota.BackoffLevel)
 					var next time.Time
-					if cooldown > 0 {
+					backoffLevel := state.Quota.BackoffLevel
-						next = now.Add(cooldown)
+					if result.RetryAfter != nil {
+						next = now.Add(*result.RetryAfter)
+					} else {
+						cooldown, nextLevel := nextQuotaCooldown(backoffLevel)
+						if cooldown > 0 {
+							next = now.Add(cooldown)
+						}
+						backoffLevel = nextLevel
 					}
 					state.NextRetryAfter = next
 					state.Quota = QuotaState{
 						Exceeded:      true,
 						Reason:        "quota",
 						NextRecoverAt: next,
-						BackoffLevel:  nextLevel,
+						BackoffLevel:  backoffLevel,
 					}
 					suspendReason = "quota"
 					shouldSuspendModel = true
@@ -582,7 +597,7 @@ func (m *Manager) MarkResult(ctx context.Context, result Result) {
 				auth.UpdatedAt = now
 				updateAggregatedAvailability(auth, now)
 			} else {
-				applyAuthFailureState(auth, result.Error, now)
+				applyAuthFailureState(auth, result.Error, result.RetryAfter, now)
 			}
 		}
 
@@ -742,6 +757,25 @@ func cloneError(err *Error) *Error {
 	}
 }
 
+func retryAfterFromError(err error) *time.Duration {
+	if err == nil {
+		return nil
+	}
+	type retryAfterProvider interface {
+		RetryAfter() *time.Duration
+	}
+	rap, ok := err.(retryAfterProvider)
+	if !ok || rap == nil {
+		return nil
+	}
+	retryAfter := rap.RetryAfter()
+	if retryAfter == nil {
+		return nil
+	}
+	val := *retryAfter
+	return &val
+}
+
 func statusCodeFromResult(err *Error) int {
 	if err == nil {
 		return 0
@@ -749,7 +783,7 @@ func statusCodeFromResult(err *Error) int {
 	return err.StatusCode()
 }
 
-func applyAuthFailureState(auth *Auth, resultErr *Error, now time.Time) {
+func applyAuthFailureState(auth *Auth, resultErr *Error, retryAfter *time.Duration, now time.Time) {
 	if auth == nil {
 		return
 	}
@@ -774,13 +808,17 @@ func applyAuthFailureState(auth *Auth, resultErr *Error, now time.Time) {
 		auth.StatusMessage = "quota exhausted"
 		auth.Quota.Exceeded = true
 		auth.Quota.Reason = "quota"
-		cooldown, nextLevel := nextQuotaCooldown(auth.Quota.BackoffLevel)
 		var next time.Time
-		if cooldown > 0 {
+		if retryAfter != nil {
-			next = now.Add(cooldown)
+			next = now.Add(*retryAfter)
+		} else {
+			cooldown, nextLevel := nextQuotaCooldown(auth.Quota.BackoffLevel)
+			if cooldown > 0 {
+				next = now.Add(cooldown)
+			}
+			auth.Quota.BackoffLevel = nextLevel
 		}
 		auth.Quota.NextRecoverAt = next
-		auth.Quota.BackoffLevel = nextLevel
 		auth.NextRetryAfter = next
 	case 408, 500, 502, 503, 504:
 		auth.StatusMessage = "transient upstream error"

sdk/cliproxy/service.go

@@ -629,7 +629,7 @@ func (s *Service) registerModelsForAuth(a *coreauth.Auth) {
 		models = registry.GetGeminiModels()
 	case "vertex":
 		// Vertex AI Gemini supports the same model identifiers as Gemini.
-		models = registry.GetGeminiModels()
+		models = registry.GetGeminiVertexModels()
 	case "gemini-cli":
 		models = registry.GetGeminiCLIModels()
 	case "aistudio":