refactor(cache, translator): remove session ID from signature caching and clean up logic

fix(executor): only strip maxOutputTokens for non-claude models

config.example.yaml

@@ -141,6 +141,15 @@ codex-instructions-enabled: false
 #       - "claude-3-*"               # wildcard matching prefix (e.g. claude-3-7-sonnet-20250219)
 #       - "*-thinking"               # wildcard matching suffix (e.g. claude-opus-4-5-thinking)
 #       - "*haiku*"                  # wildcard matching substring (e.g. claude-3-5-haiku-20241022)
+#     cloak:                         # optional: request cloaking for non-Claude-Code clients
+#       mode: "auto"                 # "auto" (default): cloak only when client is not Claude Code
+#                                    # "always": always apply cloaking
+#                                    # "never": never apply cloaking
+#       strict-mode: false           # false (default): prepend Claude Code prompt to user system messages
+#                                    # true: strip all user system messages, keep only Claude Code prompt
+#       sensitive-words:             # optional: words to obfuscate with zero-width characters
+#         - "API"
+#         - "proxy"
 
 # OpenAI compatibility providers
 # openai-compatibility:

internal/api/handlers/management/auth_files.go

@@ -3,6 +3,8 @@ package management
 import (
 	"bytes"
 	"context"
+	"crypto/sha256"
+	"encoding/hex"
 	"encoding/json"
 	"errors"
 	"fmt"
@@ -1383,9 +1385,16 @@ func (h *Handler) RequestCodexToken(c *gin.Context) {
 		claims, _ := codex.ParseJWTToken(tokenResp.IDToken)
 		email := ""
 		accountID := ""
+		planType := ""
 		if claims != nil {
 			email = claims.GetUserEmail()
 			accountID = claims.GetAccountID()
+			planType = strings.TrimSpace(claims.CodexAuthInfo.ChatgptPlanType)
+		}
+		hashAccountID := ""
+		if accountID != "" {
+			digest := sha256.Sum256([]byte(accountID))
+			hashAccountID = hex.EncodeToString(digest[:])[:8]
 		}
 		// Build bundle compatible with existing storage
 		bundle := &codex.CodexAuthBundle{
@@ -1402,10 +1411,11 @@ func (h *Handler) RequestCodexToken(c *gin.Context) {
 
 		// Create token storage and persist
 		tokenStorage := openaiAuth.CreateTokenStorage(bundle)
+		fileName := codex.CredentialFileName(tokenStorage.Email, planType, hashAccountID, true)
 		record := &coreauth.Auth{
-			ID:       fmt.Sprintf("codex-%s.json", tokenStorage.Email),
+			ID:       fileName,
 			Provider: "codex",
-			FileName: fmt.Sprintf("codex-%s.json", tokenStorage.Email),
+			FileName: fileName,
 			Storage:  tokenStorage,
 			Metadata: map[string]any{
 				"email":      tokenStorage.Email,
@@ -2110,7 +2120,20 @@ func performGeminiCLISetup(ctx context.Context, httpClient *http.Client, storage
 			finalProjectID := projectID
 			if responseProjectID != "" {
 				if explicitProject && !strings.EqualFold(responseProjectID, projectID) {
-					log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
+					// Check if this is a free user (gen-lang-client projects or free/legacy tier)
+					isFreeUser := strings.HasPrefix(projectID, "gen-lang-client-") ||
+						strings.EqualFold(tierID, "FREE") ||
+						strings.EqualFold(tierID, "LEGACY")
+
+					if isFreeUser {
+						// For free users, use backend project ID for preview model access
+						log.Infof("Gemini onboarding: frontend project %s maps to backend project %s", projectID, responseProjectID)
+						log.Infof("Using backend project ID: %s (recommended for preview model access)", responseProjectID)
+						finalProjectID = responseProjectID
+					} else {
+						// Pro users: keep requested project ID (original behavior)
+						log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
+					}
 				} else {
 					finalProjectID = responseProjectID
 				}

internal/auth/codex/filename.go

@@ -0,0 +1,57 @@
+package codex
+
+import (
+	"fmt"
+	"strings"
+	"unicode"
+
+	"golang.org/x/text/cases"
+	"golang.org/x/text/language"
+)
+
+// CredentialFileName returns the filename used to persist Codex OAuth credentials.
+// When planType is available (e.g. "plus", "team"), it is appended after the email
+// as a suffix to disambiguate subscriptions.
+func CredentialFileName(email, planType, hashAccountID string, includeProviderPrefix bool) string {
+	email = strings.TrimSpace(email)
+	plan := normalizePlanTypeForFilename(planType)
+
+	prefix := ""
+	if includeProviderPrefix {
+		prefix = "codex"
+	}
+
+	if plan == "" {
+		return fmt.Sprintf("%s-%s.json", prefix, email)
+	} else if plan == "team" {
+		return fmt.Sprintf("%s-%s-%s-%s.json", prefix, hashAccountID, email, plan)
+	}
+	return fmt.Sprintf("%s-%s-%s.json", prefix, email, plan)
+}
+
+func normalizePlanTypeForFilename(planType string) string {
+	planType = strings.TrimSpace(planType)
+	if planType == "" {
+		return ""
+	}
+
+	parts := strings.FieldsFunc(planType, func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	})
+	if len(parts) == 0 {
+		return ""
+	}
+
+	for i, part := range parts {
+		parts[i] = titleToken(part)
+	}
+	return strings.Join(parts, "-")
+}
+
+func titleToken(token string) string {
+	token = strings.TrimSpace(token)
+	if token == "" {
+		return ""
+	}
+	return cases.Title(language.English).String(token)
+}

internal/cache/signature_cache.go

@@ -3,6 +3,8 @@ package cache
 import (
 	"crypto/sha256"
 	"encoding/hex"
+	"fmt"
+	"strings"
 	"sync"
 	"time"
 )
@@ -94,17 +96,17 @@ func purgeExpiredSessions() {
 
 // CacheSignature stores a thinking signature for a given session and text.
 // Used for Claude models that require signed thinking blocks in multi-turn conversations.
-func CacheSignature(sessionID, text, signature string) {
-	if sessionID == "" || text == "" || signature == "" {
+func CacheSignature(modelName, text, signature string) {
+	if text == "" || signature == "" {
 		return
 	}
 	if len(signature) < MinValidSignatureLen {
 		return
 	}
 
-	sc := getOrCreateSession(sessionID)
+	text = fmt.Sprintf("%s#%s", GetModelGroup(modelName), text)
 	textHash := hashText(text)
-
+	sc := getOrCreateSession(textHash)
 	sc.mu.Lock()
 	defer sc.mu.Unlock()
 
@@ -116,13 +118,21 @@ func CacheSignature(sessionID, text, signature string) {
 
 // GetCachedSignature retrieves a cached signature for a given session and text.
 // Returns empty string if not found or expired.
-func GetCachedSignature(sessionID, text string) string {
-	if sessionID == "" || text == "" {
+func GetCachedSignature(modelName, text string) string {
+	family := GetModelGroup(modelName)
+
+	if text == "" {
+		if family == "gemini" {
+			return "skip_thought_signature_validator"
+		}
 		return ""
 	}
-
-	val, ok := signatureCache.Load(sessionID)
+	text = fmt.Sprintf("%s#%s", GetModelGroup(modelName), text)
+	val, ok := signatureCache.Load(hashText(text))
 	if !ok {
+		if family == "gemini" {
+			return "skip_thought_signature_validator"
+		}
 		return ""
 	}
 	sc := val.(*sessionCache)
@@ -135,11 +145,17 @@ func GetCachedSignature(sessionID, text string) string {
 	entry, exists := sc.entries[textHash]
 	if !exists {
 		sc.mu.Unlock()
+		if family == "gemini" {
+			return "skip_thought_signature_validator"
+		}
 		return ""
 	}
 	if now.Sub(entry.Timestamp) > SignatureCacheTTL {
 		delete(sc.entries, textHash)
 		sc.mu.Unlock()
+		if family == "gemini" {
+			return "skip_thought_signature_validator"
+		}
 		return ""
 	}
 
@@ -154,7 +170,13 @@ func GetCachedSignature(sessionID, text string) string {
 // ClearSignatureCache clears signature cache for a specific session or all sessions.
 func ClearSignatureCache(sessionID string) {
 	if sessionID != "" {
-		signatureCache.Delete(sessionID)
+		signatureCache.Range(func(key, _ any) bool {
+			kStr, ok := key.(string)
+			if ok && strings.HasSuffix(kStr, "#"+sessionID) {
+				signatureCache.Delete(key)
+			}
+			return true
+		})
 	} else {
 		signatureCache.Range(func(key, _ any) bool {
 			signatureCache.Delete(key)
@@ -164,6 +186,17 @@ func ClearSignatureCache(sessionID string) {
 }
 
 // HasValidSignature checks if a signature is valid (non-empty and long enough)
-func HasValidSignature(signature string) bool {
-	return signature != "" && len(signature) >= MinValidSignatureLen
+func HasValidSignature(modelName, signature string) bool {
+	return (signature != "" && len(signature) >= MinValidSignatureLen) || (signature == "skip_thought_signature_validator" && GetModelGroup(modelName) == "gemini")
+}
+
+func GetModelGroup(modelName string) string {
+	if strings.Contains(modelName, "gpt") {
+		return "gpt"
+	} else if strings.Contains(modelName, "claude") {
+		return "claude"
+	} else if strings.Contains(modelName, "gemini") {
+		return "gemini"
+	}
+	return modelName
 }

internal/cache/signature_cache_test.go

@@ -8,15 +8,14 @@ import (
 func TestCacheSignature_BasicStorageAndRetrieval(t *testing.T) {
 	ClearSignatureCache("")
 
-	sessionID := "test-session-1"
 	text := "This is some thinking text content"
 	signature := "abc123validSignature1234567890123456789012345678901234567890"
 
 	// Store signature
-	CacheSignature(sessionID, text, signature)
+	CacheSignature("test-model", text, signature)
 
 	// Retrieve signature
-	retrieved := GetCachedSignature(sessionID, text)
+	retrieved := GetCachedSignature("test-model", text)
 	if retrieved != signature {
 		t.Errorf("Expected signature '%s', got '%s'", signature, retrieved)
 	}
@@ -29,13 +28,13 @@ func TestCacheSignature_DifferentSessions(t *testing.T) {
 	sig1 := "signature1_1234567890123456789012345678901234567890123456"
 	sig2 := "signature2_1234567890123456789012345678901234567890123456"
 
-	CacheSignature("session-a", text, sig1)
-	CacheSignature("session-b", text, sig2)
+	CacheSignature("test-model", text, sig1)
+	CacheSignature("test-model", text, sig2)
 
-	if GetCachedSignature("session-a", text) != sig1 {
+	if GetCachedSignature("test-model", text) != sig1 {
 		t.Error("Session-a signature mismatch")
 	}
-	if GetCachedSignature("session-b", text) != sig2 {
+	if GetCachedSignature("test-model", text) != sig2 {
 		t.Error("Session-b signature mismatch")
 	}
 }
@@ -44,13 +43,13 @@ func TestCacheSignature_NotFound(t *testing.T) {
 	ClearSignatureCache("")
 
 	// Non-existent session
-	if got := GetCachedSignature("nonexistent", "some text"); got != "" {
+	if got := GetCachedSignature("test-model", "some text"); got != "" {
 		t.Errorf("Expected empty string for nonexistent session, got '%s'", got)
 	}
 
 	// Existing session but different text
-	CacheSignature("session-x", "text-a", "sigA12345678901234567890123456789012345678901234567890")
-	if got := GetCachedSignature("session-x", "text-b"); got != "" {
+	CacheSignature("test-model", "text-a", "sigA12345678901234567890123456789012345678901234567890")
+	if got := GetCachedSignature("test-model", "text-b"); got != "" {
 		t.Errorf("Expected empty string for different text, got '%s'", got)
 	}
 }
@@ -59,12 +58,12 @@ func TestCacheSignature_EmptyInputs(t *testing.T) {
 	ClearSignatureCache("")
 
 	// All empty/invalid inputs should be no-ops
-	CacheSignature("", "text", "sig12345678901234567890123456789012345678901234567890")
-	CacheSignature("session", "", "sig12345678901234567890123456789012345678901234567890")
-	CacheSignature("session", "text", "")
-	CacheSignature("session", "text", "short") // Too short
+	CacheSignature("test-model", "text", "sig12345678901234567890123456789012345678901234567890")
+	CacheSignature("test-model", "", "sig12345678901234567890123456789012345678901234567890")
+	CacheSignature("test-model", "text", "")
+	CacheSignature("test-model", "text", "short") // Too short
 
-	if got := GetCachedSignature("session", "text"); got != "" {
+	if got := GetCachedSignature("test-model", "text"); got != "" {
 		t.Errorf("Expected empty after invalid cache attempts, got '%s'", got)
 	}
 }
@@ -72,13 +71,12 @@ func TestCacheSignature_EmptyInputs(t *testing.T) {
 func TestCacheSignature_ShortSignatureRejected(t *testing.T) {
 	ClearSignatureCache("")
 
-	sessionID := "test-short-sig"
 	text := "Some text"
 	shortSig := "abc123" // Less than 50 chars
 
-	CacheSignature(sessionID, text, shortSig)
+	CacheSignature("test-model", text, shortSig)
 
-	if got := GetCachedSignature(sessionID, text); got != "" {
+	if got := GetCachedSignature("test-model", text); got != "" {
 		t.Errorf("Short signature should be rejected, got '%s'", got)
 	}
 }
@@ -87,15 +85,15 @@ func TestClearSignatureCache_SpecificSession(t *testing.T) {
 	ClearSignatureCache("")
 
 	sig := "validSig1234567890123456789012345678901234567890123456"
-	CacheSignature("session-1", "text", sig)
-	CacheSignature("session-2", "text", sig)
+	CacheSignature("test-model", "text", sig)
+	CacheSignature("test-model", "text", sig)
 
 	ClearSignatureCache("session-1")
 
-	if got := GetCachedSignature("session-1", "text"); got != "" {
+	if got := GetCachedSignature("test-model", "text"); got != "" {
 		t.Error("session-1 should be cleared")
 	}
-	if got := GetCachedSignature("session-2", "text"); got != sig {
+	if got := GetCachedSignature("test-model", "text"); got != sig {
 		t.Error("session-2 should still exist")
 	}
 }
@@ -104,15 +102,15 @@ func TestClearSignatureCache_AllSessions(t *testing.T) {
 	ClearSignatureCache("")
 
 	sig := "validSig1234567890123456789012345678901234567890123456"
-	CacheSignature("session-1", "text", sig)
-	CacheSignature("session-2", "text", sig)
+	CacheSignature("test-model", "text", sig)
+	CacheSignature("test-model", "text", sig)
 
 	ClearSignatureCache("")
 
-	if got := GetCachedSignature("session-1", "text"); got != "" {
+	if got := GetCachedSignature("test-model", "text"); got != "" {
 		t.Error("session-1 should be cleared")
 	}
-	if got := GetCachedSignature("session-2", "text"); got != "" {
+	if got := GetCachedSignature("test-model", "text"); got != "" {
 		t.Error("session-2 should be cleared")
 	}
 }
@@ -132,7 +130,7 @@ func TestHasValidSignature(t *testing.T) {
 
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			result := HasValidSignature(tt.signature)
+			result := HasValidSignature("claude-sonnet-4-5-thinking", tt.signature)
 			if result != tt.expected {
 				t.Errorf("HasValidSignature(%q) = %v, expected %v", tt.signature, result, tt.expected)
 			}
@@ -143,21 +141,19 @@ func TestHasValidSignature(t *testing.T) {
 func TestCacheSignature_TextHashCollisionResistance(t *testing.T) {
 	ClearSignatureCache("")
 
-	sessionID := "hash-test-session"
-
 	// Different texts should produce different hashes
 	text1 := "First thinking text"
 	text2 := "Second thinking text"
 	sig1 := "signature1_1234567890123456789012345678901234567890123456"
 	sig2 := "signature2_1234567890123456789012345678901234567890123456"
 
-	CacheSignature(sessionID, text1, sig1)
-	CacheSignature(sessionID, text2, sig2)
+	CacheSignature("test-model", text1, sig1)
+	CacheSignature("test-model", text2, sig2)
 
-	if GetCachedSignature(sessionID, text1) != sig1 {
+	if GetCachedSignature("test-model", text1) != sig1 {
 		t.Error("text1 signature mismatch")
 	}
-	if GetCachedSignature(sessionID, text2) != sig2 {
+	if GetCachedSignature("test-model", text2) != sig2 {
 		t.Error("text2 signature mismatch")
 	}
 }
@@ -165,13 +161,12 @@ func TestCacheSignature_TextHashCollisionResistance(t *testing.T) {
 func TestCacheSignature_UnicodeText(t *testing.T) {
 	ClearSignatureCache("")
 
-	sessionID := "unicode-session"
 	text := "한글 텍스트와 이모지 🎉 그리고 特殊文字"
 	sig := "unicodeSig123456789012345678901234567890123456789012345"
 
-	CacheSignature(sessionID, text, sig)
+	CacheSignature("test-model", text, sig)
 
-	if got := GetCachedSignature(sessionID, text); got != sig {
+	if got := GetCachedSignature("test-model", text); got != sig {
 		t.Errorf("Unicode text signature retrieval failed, got '%s'", got)
 	}
 }
@@ -179,15 +174,14 @@ func TestCacheSignature_UnicodeText(t *testing.T) {
 func TestCacheSignature_Overwrite(t *testing.T) {
 	ClearSignatureCache("")
 
-	sessionID := "overwrite-session"
 	text := "Same text"
 	sig1 := "firstSignature12345678901234567890123456789012345678901"
 	sig2 := "secondSignature1234567890123456789012345678901234567890"
 
-	CacheSignature(sessionID, text, sig1)
-	CacheSignature(sessionID, text, sig2) // Overwrite
+	CacheSignature("test-model", text, sig1)
+	CacheSignature("test-model", text, sig2) // Overwrite
 
-	if got := GetCachedSignature(sessionID, text); got != sig2 {
+	if got := GetCachedSignature("test-model", text); got != sig2 {
 		t.Errorf("Expected overwritten signature '%s', got '%s'", sig2, got)
 	}
 }
@@ -199,14 +193,13 @@ func TestCacheSignature_ExpirationLogic(t *testing.T) {
 
 	// This test verifies the expiration check exists
 	// In a real scenario, we'd mock time.Now()
-	sessionID := "expiration-test"
 	text := "text"
 	sig := "validSig1234567890123456789012345678901234567890123456"
 
-	CacheSignature(sessionID, text, sig)
+	CacheSignature("test-model", text, sig)
 
 	// Fresh entry should be retrievable
-	if got := GetCachedSignature(sessionID, text); got != sig {
+	if got := GetCachedSignature("test-model", text); got != sig {
 		t.Errorf("Fresh entry should be retrievable, got '%s'", got)
 	}
 

internal/cmd/login.go

@@ -118,6 +118,7 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 	}
 
 	activatedProjects := make([]string, 0, len(projectSelections))
+	seenProjects := make(map[string]bool)
 	for _, candidateID := range projectSelections {
 		log.Infof("Activating project %s", candidateID)
 		if errSetup := performGeminiCLISetup(ctx, httpClient, storage, candidateID); errSetup != nil {
@@ -134,6 +135,13 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 		if finalID == "" {
 			finalID = candidateID
 		}
+
+		// Skip duplicates
+		if seenProjects[finalID] {
+			log.Infof("Project %s already activated, skipping", finalID)
+			continue
+		}
+		seenProjects[finalID] = true
 		activatedProjects = append(activatedProjects, finalID)
 	}
 
@@ -261,7 +269,39 @@ func performGeminiCLISetup(ctx context.Context, httpClient *http.Client, storage
 			finalProjectID := projectID
 			if responseProjectID != "" {
 				if explicitProject && !strings.EqualFold(responseProjectID, projectID) {
-					log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
+					// Check if this is a free user (gen-lang-client projects or free/legacy tier)
+					isFreeUser := strings.HasPrefix(projectID, "gen-lang-client-") ||
+						strings.EqualFold(tierID, "FREE") ||
+						strings.EqualFold(tierID, "LEGACY")
+
+					if isFreeUser {
+						// Interactive prompt for free users
+						fmt.Printf("\nGoogle returned a different project ID:\n")
+						fmt.Printf("  Requested (frontend): %s\n", projectID)
+						fmt.Printf("  Returned (backend):   %s\n\n", responseProjectID)
+						fmt.Printf("  Backend project IDs have access to preview models (gemini-3-*).\n")
+						fmt.Printf("  This is normal for free tier users.\n\n")
+						fmt.Printf("Which project ID would you like to use?\n")
+						fmt.Printf("  [1] Backend (recommended): %s\n", responseProjectID)
+						fmt.Printf("  [2] Frontend: %s\n\n", projectID)
+						fmt.Printf("Enter choice [1]: ")
+
+						reader := bufio.NewReader(os.Stdin)
+						choice, _ := reader.ReadString('\n')
+						choice = strings.TrimSpace(choice)
+
+						if choice == "2" {
+							log.Infof("Using frontend project ID: %s", projectID)
+							fmt.Println(". Warning: Frontend project IDs may not have access to preview models.")
+							finalProjectID = projectID
+						} else {
+							log.Infof("Using backend project ID: %s (recommended)", responseProjectID)
+							finalProjectID = responseProjectID
+						}
+					} else {
+						// Pro users: keep requested project ID (original behavior)
+						log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
+					}
 				} else {
 					finalProjectID = responseProjectID
 				}

internal/config/config.go

@@ -248,6 +248,25 @@ type PayloadModelRule struct {
 	Protocol string `yaml:"protocol" json:"protocol"`
 }
 
+// CloakConfig configures request cloaking for non-Claude-Code clients.
+// Cloaking disguises API requests to appear as originating from the official Claude Code CLI.
+type CloakConfig struct {
+	// Mode controls cloaking behavior: "auto" (default), "always", or "never".
+	// - "auto": cloak only when client is not Claude Code (based on User-Agent)
+	// - "always": always apply cloaking regardless of client
+	// - "never": never apply cloaking
+	Mode string `yaml:"mode,omitempty" json:"mode,omitempty"`
+
+	// StrictMode controls how system prompts are handled when cloaking.
+	// - false (default): prepend Claude Code prompt to user system messages
+	// - true: strip all user system messages, keep only Claude Code prompt
+	StrictMode bool `yaml:"strict-mode,omitempty" json:"strict-mode,omitempty"`
+
+	// SensitiveWords is a list of words to obfuscate with zero-width characters.
+	// This can help bypass certain content filters.
+	SensitiveWords []string `yaml:"sensitive-words,omitempty" json:"sensitive-words,omitempty"`
+}
+
 // ClaudeKey represents the configuration for a Claude API key,
 // including the API key itself and an optional base URL for the API endpoint.
 type ClaudeKey struct {
@@ -276,6 +295,9 @@ type ClaudeKey struct {
 
 	// ExcludedModels lists model IDs that should be excluded for this provider.
 	ExcludedModels []string `yaml:"excluded-models,omitempty" json:"excluded-models,omitempty"`
+
+	// Cloak configures request cloaking for non-Claude-Code clients.
+	Cloak *CloakConfig `yaml:"cloak,omitempty" json:"cloak,omitempty"`
 }
 
 func (k ClaudeKey) GetAPIKey() string  { return k.APIKey }

internal/logging/gin_logger.go

@@ -4,6 +4,7 @@
 package logging
 
 import (
+	"errors"
 	"fmt"
 	"net/http"
 	"runtime/debug"
@@ -112,6 +113,11 @@ func isAIAPIPath(path string) bool {
 //   - gin.HandlerFunc: A middleware handler for panic recovery
 func GinLogrusRecovery() gin.HandlerFunc {
 	return gin.CustomRecovery(func(c *gin.Context, recovered interface{}) {
+		if err, ok := recovered.(error); ok && errors.Is(err, http.ErrAbortHandler) {
+			// Let net/http handle ErrAbortHandler so the connection is aborted without noisy stack logs.
+			panic(http.ErrAbortHandler)
+		}
+
 		log.WithFields(log.Fields{
 			"panic": recovered,
 			"stack": string(debug.Stack()),

internal/logging/gin_logger_test.go

@@ -0,0 +1,60 @@
+package logging
+
+import (
+	"errors"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+)
+
+func TestGinLogrusRecoveryRepanicsErrAbortHandler(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	engine := gin.New()
+	engine.Use(GinLogrusRecovery())
+	engine.GET("/abort", func(c *gin.Context) {
+		panic(http.ErrAbortHandler)
+	})
+
+	req := httptest.NewRequest(http.MethodGet, "/abort", nil)
+	recorder := httptest.NewRecorder()
+
+	defer func() {
+		recovered := recover()
+		if recovered == nil {
+			t.Fatalf("expected panic, got nil")
+		}
+		err, ok := recovered.(error)
+		if !ok {
+			t.Fatalf("expected error panic, got %T", recovered)
+		}
+		if !errors.Is(err, http.ErrAbortHandler) {
+			t.Fatalf("expected ErrAbortHandler, got %v", err)
+		}
+		if err != http.ErrAbortHandler {
+			t.Fatalf("expected exact ErrAbortHandler sentinel, got %v", err)
+		}
+	}()
+
+	engine.ServeHTTP(recorder, req)
+}
+
+func TestGinLogrusRecoveryHandlesRegularPanic(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	engine := gin.New()
+	engine.Use(GinLogrusRecovery())
+	engine.GET("/panic", func(c *gin.Context) {
+		panic("boom")
+	})
+
+	req := httptest.NewRequest(http.MethodGet, "/panic", nil)
+	recorder := httptest.NewRecorder()
+
+	engine.ServeHTTP(recorder, req)
+	if recorder.Code != http.StatusInternalServerError {
+		t.Fatalf("expected 500, got %d", recorder.Code)
+	}
+}

internal/registry/model_definitions.go

@@ -287,6 +287,67 @@ func GetGeminiVertexModels() []*ModelInfo {
 			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
 			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
 		},
+		// Imagen image generation models - use :predict action
+		{
+			ID:                         "imagen-4.0-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Generate",
+			Description:                "Imagen 4.0 image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-4.0-ultra-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-ultra-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Ultra Generate",
+			Description:                "Imagen 4.0 Ultra high-quality image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-3.0-generate-002",
+			Object:                     "model",
+			Created:                    1740000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-3.0-generate-002",
+			Version:                    "3.0",
+			DisplayName:                "Imagen 3.0 Generate",
+			Description:                "Imagen 3.0 image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-3.0-fast-generate-001",
+			Object:                     "model",
+			Created:                    1740000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-3.0-fast-generate-001",
+			Version:                    "3.0",
+			DisplayName:                "Imagen 3.0 Fast Generate",
+			Description:                "Imagen 3.0 fast image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-4.0-fast-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-fast-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Fast Generate",
+			Description:                "Imagen 4.0 fast image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
 	}
 }
 
@@ -765,21 +826,23 @@ func GetIFlowModels() []*ModelInfo {
 type AntigravityModelConfig struct {
 	Thinking            *ThinkingSupport
 	MaxCompletionTokens int
-	Name                string
 }
 
 // GetAntigravityModelConfig returns static configuration for antigravity models.
 // Keys use upstream model names returned by the Antigravity models endpoint.
 func GetAntigravityModelConfig() map[string]*AntigravityModelConfig {
 	return map[string]*AntigravityModelConfig{
-		"gemini-2.5-flash":           {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}, Name: "models/gemini-2.5-flash"},
-		"gemini-2.5-flash-lite":      {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}, Name: "models/gemini-2.5-flash-lite"},
-		"rev19-uic3-1p":              {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}, Name: "models/rev19-uic3-1p"},
-		"gemini-3-pro-high":          {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}, Name: "models/gemini-3-pro-high"},
-		"gemini-3-pro-image":         {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}, Name: "models/gemini-3-pro-image"},
-		"gemini-3-flash":             {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}, Name: "models/gemini-3-flash"},
+		"gemini-2.5-flash":           {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
+		"gemini-2.5-flash-lite":      {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
+		"rev19-uic3-1p":              {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}},
+		"gemini-3-pro-high":          {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
+		"gemini-3-pro-image":         {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
+		"gemini-3-flash":             {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}},
 		"claude-sonnet-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
 		"claude-opus-4-5-thinking":   {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
+		"claude-sonnet-4-5":          {MaxCompletionTokens: 64000},
+		"gpt-oss-120b-medium":        {},
+		"tab_flash_lite_preview":     {},
 	}
 }
 
@@ -809,10 +872,9 @@ func LookupStaticModelInfo(modelID string) *ModelInfo {
 	}
 
 	// Check Antigravity static config
-	if cfg := GetAntigravityModelConfig()[modelID]; cfg != nil && cfg.Thinking != nil {
+	if cfg := GetAntigravityModelConfig()[modelID]; cfg != nil {
 		return &ModelInfo{
 			ID:                  modelID,
-			Name:                cfg.Name,
 			Thinking:            cfg.Thinking,
 			MaxCompletionTokens: cfg.MaxCompletionTokens,
 		}

internal/registry/model_registry.go

@@ -78,6 +78,8 @@ type ThinkingSupport struct {
 type ModelRegistration struct {
 	// Info contains the model metadata
 	Info *ModelInfo
+	// InfoByProvider maps provider identifiers to specific ModelInfo to support differing capabilities.
+	InfoByProvider map[string]*ModelInfo
 	// Count is the number of active clients that can provide this model
 	Count int
 	// LastUpdated tracks when this registration was last modified
@@ -132,16 +134,19 @@ func GetGlobalRegistry() *ModelRegistry {
 	return globalRegistry
 }
 
-// LookupModelInfo searches the dynamic registry first, then falls back to static model definitions.
-//
-// This helper exists because some code paths only have a model ID and still need Thinking and
-// max completion token metadata even when the dynamic registry hasn't been populated.
-func LookupModelInfo(modelID string) *ModelInfo {
+// LookupModelInfo searches dynamic registry (provider-specific > global) then static definitions.
+func LookupModelInfo(modelID string, provider ...string) *ModelInfo {
 	modelID = strings.TrimSpace(modelID)
 	if modelID == "" {
 		return nil
 	}
-	if info := GetGlobalRegistry().GetModelInfo(modelID); info != nil {
+
+	p := ""
+	if len(provider) > 0 {
+		p = strings.ToLower(strings.TrimSpace(provider[0]))
+	}
+
+	if info := GetGlobalRegistry().GetModelInfo(modelID, p); info != nil {
 		return info
 	}
 	return LookupStaticModelInfo(modelID)
@@ -297,6 +302,9 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 				if count, okProv := reg.Providers[oldProvider]; okProv {
 					if count <= toRemove {
 						delete(reg.Providers, oldProvider)
+						if reg.InfoByProvider != nil {
+							delete(reg.InfoByProvider, oldProvider)
+						}
 					} else {
 						reg.Providers[oldProvider] = count - toRemove
 					}
@@ -346,6 +354,12 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 		model := newModels[id]
 		if reg, ok := r.models[id]; ok {
 			reg.Info = cloneModelInfo(model)
+			if provider != "" {
+				if reg.InfoByProvider == nil {
+					reg.InfoByProvider = make(map[string]*ModelInfo)
+				}
+				reg.InfoByProvider[provider] = cloneModelInfo(model)
+			}
 			reg.LastUpdated = now
 			if reg.QuotaExceededClients != nil {
 				delete(reg.QuotaExceededClients, clientID)
@@ -409,11 +423,15 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 		if existing.SuspendedClients == nil {
 			existing.SuspendedClients = make(map[string]string)
 		}
+		if existing.InfoByProvider == nil {
+			existing.InfoByProvider = make(map[string]*ModelInfo)
+		}
 		if provider != "" {
 			if existing.Providers == nil {
 				existing.Providers = make(map[string]int)
 			}
 			existing.Providers[provider]++
+			existing.InfoByProvider[provider] = cloneModelInfo(model)
 		}
 		log.Debugf("Incremented count for model %s, now %d clients", modelID, existing.Count)
 		return
@@ -421,6 +439,7 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 
 	registration := &ModelRegistration{
 		Info:                 cloneModelInfo(model),
+		InfoByProvider:       make(map[string]*ModelInfo),
 		Count:                1,
 		LastUpdated:          now,
 		QuotaExceededClients: make(map[string]*time.Time),
@@ -428,6 +447,7 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 	}
 	if provider != "" {
 		registration.Providers = map[string]int{provider: 1}
+		registration.InfoByProvider[provider] = cloneModelInfo(model)
 	}
 	r.models[modelID] = registration
 	log.Debugf("Registered new model %s from provider %s", modelID, provider)
@@ -453,6 +473,9 @@ func (r *ModelRegistry) removeModelRegistration(clientID, modelID, provider stri
 		if count, ok := registration.Providers[provider]; ok {
 			if count <= 1 {
 				delete(registration.Providers, provider)
+				if registration.InfoByProvider != nil {
+					delete(registration.InfoByProvider, provider)
+				}
 			} else {
 				registration.Providers[provider] = count - 1
 			}
@@ -534,6 +557,9 @@ func (r *ModelRegistry) unregisterClientInternal(clientID string) {
 				if count, ok := registration.Providers[provider]; ok {
 					if count <= 1 {
 						delete(registration.Providers, provider)
+						if registration.InfoByProvider != nil {
+							delete(registration.InfoByProvider, provider)
+						}
 					} else {
 						registration.Providers[provider] = count - 1
 					}
@@ -940,12 +966,22 @@ func (r *ModelRegistry) GetModelProviders(modelID string) []string {
 	return result
 }
 
-// GetModelInfo returns the registered ModelInfo for the given model ID, if present.
-// Returns nil if the model is unknown to the registry.
-func (r *ModelRegistry) GetModelInfo(modelID string) *ModelInfo {
+// GetModelInfo returns ModelInfo, prioritizing provider-specific definition if available.
+func (r *ModelRegistry) GetModelInfo(modelID, provider string) *ModelInfo {
 	r.mutex.RLock()
 	defer r.mutex.RUnlock()
 	if reg, ok := r.models[modelID]; ok && reg != nil {
+		// Try provider specific definition first
+		if provider != "" && reg.InfoByProvider != nil {
+			if reg.Providers != nil {
+				if count, ok := reg.Providers[provider]; ok && count > 0 {
+					if info, ok := reg.InfoByProvider[provider]; ok && info != nil {
+						return info
+					}
+				}
+			}
+		}
+		// Fallback to global info (last registered)
 		return reg.Info
 	}
 	return nil

internal/runtime/executor/aistudio_executor.go

@@ -393,7 +393,7 @@ func (e *AIStudioExecutor) translateRequest(req cliproxyexecutor.Request, opts c
 	}
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, stream)
 	payload := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), stream)
-	payload, err := thinking.ApplyThinking(payload, req.Model, from.String(), to.String())
+	payload, err := thinking.ApplyThinking(payload, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, translatedPayload{}, err
 	}

internal/runtime/executor/antigravity_executor.go

@@ -137,7 +137,7 @@ func (e *AntigravityExecutor) Execute(ctx context.Context, auth *cliproxyauth.Au
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
 	translated := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -256,7 +256,7 @@ func (e *AntigravityExecutor) executeClaudeNonStream(ctx context.Context, auth *
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	translated := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -622,7 +622,7 @@ func (e *AntigravityExecutor) ExecuteStream(ctx context.Context, auth *cliproxya
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	translated := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -802,7 +802,7 @@ func (e *AntigravityExecutor) CountTokens(ctx context.Context, auth *cliproxyaut
 	// Prepare payload once (doesn't depend on baseURL)
 	payload := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	payload, err := thinking.ApplyThinking(payload, req.Model, from.String(), to.String())
+	payload, err := thinking.ApplyThinking(payload, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}
@@ -1005,9 +1005,6 @@ func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *c
 			}
 			modelCfg := modelConfig[modelID]
 			modelName := modelID
-			if modelCfg != nil && modelCfg.Name != "" {
-				modelName = modelCfg.Name
-			}
 			modelInfo := &registry.ModelInfo{
 				ID:          modelID,
 				Name:        modelName,
@@ -1205,7 +1202,7 @@ func (e *AntigravityExecutor) buildRequest(ctx context.Context, auth *cliproxyau
 	payload = geminiToAntigravity(modelName, payload, projectID)
 	payload, _ = sjson.SetBytes(payload, "model", modelName)
 
-	if strings.Contains(modelName, "claude") {
+	if strings.Contains(modelName, "claude") || strings.Contains(modelName, "gemini-3-pro-high") {
 		strJSON := string(payload)
 		paths := make([]string, 0)
 		util.Walk(gjson.ParseBytes(payload), "", "parametersJsonSchema", &paths)
@@ -1408,16 +1405,9 @@ func geminiToAntigravity(modelName string, payload []byte, projectID string) []b
 	template, _ = sjson.Set(template, "request.sessionId", generateStableSessionID(payload))
 
 	template, _ = sjson.Delete(template, "request.safetySettings")
-	template, _ = sjson.Set(template, "request.toolConfig.functionCallingConfig.mode", "VALIDATED")
+	//	template, _ = sjson.Set(template, "request.toolConfig.functionCallingConfig.mode", "VALIDATED")
 
-	if !strings.HasPrefix(modelName, "gemini-3-") {
-		if thinkingLevel := gjson.Get(template, "request.generationConfig.thinkingConfig.thinkingLevel"); thinkingLevel.Exists() {
-			template, _ = sjson.Delete(template, "request.generationConfig.thinkingConfig.thinkingLevel")
-			template, _ = sjson.Set(template, "request.generationConfig.thinkingConfig.thinkingBudget", -1)
-		}
-	}
-
-	if strings.Contains(modelName, "claude") {
+	if strings.Contains(modelName, "claude") || strings.Contains(modelName, "gemini-3-pro-high") {
 		gjson.Get(template, "request.tools").ForEach(func(key, tool gjson.Result) bool {
 			tool.Get("functionDeclarations").ForEach(func(funKey, funcDecl gjson.Result) bool {
 				if funcDecl.Get("parametersJsonSchema").Exists() {
@@ -1429,7 +1419,9 @@ func geminiToAntigravity(modelName string, payload []byte, projectID string) []b
 			})
 			return true
 		})
-	} else {
+	}
+
+	if !strings.Contains(modelName, "claude") {
 		template, _ = sjson.Delete(template, "request.generationConfig.maxOutputTokens")
 	}
 

internal/runtime/executor/claude_executor.go

@@ -17,7 +17,6 @@ import (
 	claudeauth "github.com/router-for-me/CLIProxyAPI/v6/internal/auth/claude"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/misc"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	cliproxyauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
@@ -106,22 +105,20 @@ func (e *ClaudeExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, r
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), stream)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
 
-	if !strings.HasPrefix(baseModel, "claude-3-5-haiku") {
-		body = checkSystemInstructions(body)
-	}
+	// Apply cloaking (system prompt injection, fake user ID, sensitive word obfuscation)
+	// based on client type and configuration.
+	body = applyCloaking(ctx, e.cfg, auth, body, baseModel)
+
 	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
 
 	// Disable thinking if tool_choice forces tool use (Anthropic API constraint)
 	body = disableThinkingIfToolChoiceForced(body)
 
-	// Ensure max_tokens > thinking.budget_tokens when thinking is enabled
-	body = ensureMaxTokensForThinking(baseModel, body)
-
 	// Extract betas from body and convert to header
 	var extraBetas []string
 	extraBetas, body = extractAndRemoveBetas(body)
@@ -239,20 +236,20 @@ func (e *ClaudeExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.A
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
 
-	body = checkSystemInstructions(body)
+	// Apply cloaking (system prompt injection, fake user ID, sensitive word obfuscation)
+	// based on client type and configuration.
+	body = applyCloaking(ctx, e.cfg, auth, body, baseModel)
+
 	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
 
 	// Disable thinking if tool_choice forces tool use (Anthropic API constraint)
 	body = disableThinkingIfToolChoiceForced(body)
 
-	// Ensure max_tokens > thinking.budget_tokens when thinking is enabled
-	body = ensureMaxTokensForThinking(baseModel, body)
-
 	// Extract betas from body and convert to header
 	var extraBetas []string
 	extraBetas, body = extractAndRemoveBetas(body)
@@ -541,81 +538,6 @@ func disableThinkingIfToolChoiceForced(body []byte) []byte {
 	return body
 }
 
-// ensureMaxTokensForThinking ensures max_tokens > thinking.budget_tokens when thinking is enabled.
-// Anthropic API requires this constraint; violating it returns a 400 error.
-// This function should be called after all thinking configuration is finalized.
-// It looks up the model's MaxCompletionTokens from the registry to use as the cap.
-func ensureMaxTokensForThinking(modelName string, body []byte) []byte {
-	thinkingType := gjson.GetBytes(body, "thinking.type").String()
-	if thinkingType != "enabled" {
-		return body
-	}
-
-	budgetTokens := gjson.GetBytes(body, "thinking.budget_tokens").Int()
-	if budgetTokens <= 0 {
-		return body
-	}
-
-	maxTokens := gjson.GetBytes(body, "max_tokens").Int()
-
-	// Look up the model's max completion tokens from the registry
-	maxCompletionTokens := 0
-	if modelInfo := registry.LookupModelInfo(modelName); modelInfo != nil {
-		maxCompletionTokens = modelInfo.MaxCompletionTokens
-	}
-
-	// Fall back to budget + buffer if registry lookup fails or returns 0
-	const fallbackBuffer = 4000
-	requiredMaxTokens := budgetTokens + fallbackBuffer
-	if maxCompletionTokens > 0 {
-		requiredMaxTokens = int64(maxCompletionTokens)
-	}
-
-	if maxTokens < requiredMaxTokens {
-		body, _ = sjson.SetBytes(body, "max_tokens", requiredMaxTokens)
-	}
-	return body
-}
-
-func (e *ClaudeExecutor) resolveClaudeConfig(auth *cliproxyauth.Auth) *config.ClaudeKey {
-	if auth == nil || e.cfg == nil {
-		return nil
-	}
-	var attrKey, attrBase string
-	if auth.Attributes != nil {
-		attrKey = strings.TrimSpace(auth.Attributes["api_key"])
-		attrBase = strings.TrimSpace(auth.Attributes["base_url"])
-	}
-	for i := range e.cfg.ClaudeKey {
-		entry := &e.cfg.ClaudeKey[i]
-		cfgKey := strings.TrimSpace(entry.APIKey)
-		cfgBase := strings.TrimSpace(entry.BaseURL)
-		if attrKey != "" && attrBase != "" {
-			if strings.EqualFold(cfgKey, attrKey) && strings.EqualFold(cfgBase, attrBase) {
-				return entry
-			}
-			continue
-		}
-		if attrKey != "" && strings.EqualFold(cfgKey, attrKey) {
-			if cfgBase == "" || strings.EqualFold(cfgBase, attrBase) {
-				return entry
-			}
-		}
-		if attrKey == "" && attrBase != "" && strings.EqualFold(cfgBase, attrBase) {
-			return entry
-		}
-	}
-	if attrKey != "" {
-		for i := range e.cfg.ClaudeKey {
-			entry := &e.cfg.ClaudeKey[i]
-			if strings.EqualFold(strings.TrimSpace(entry.APIKey), attrKey) {
-				return entry
-			}
-		}
-	}
-	return nil
-}
-
 type compositeReadCloser struct {
 	io.Reader
 	closers []func() error
@@ -901,3 +823,163 @@ func stripClaudeToolPrefixFromStreamLine(line []byte, prefix string) []byte {
 	}
 	return updated
 }
+
+// getClientUserAgent extracts the client User-Agent from the gin context.
+func getClientUserAgent(ctx context.Context) string {
+	if ginCtx, ok := ctx.Value("gin").(*gin.Context); ok && ginCtx != nil && ginCtx.Request != nil {
+		return ginCtx.GetHeader("User-Agent")
+	}
+	return ""
+}
+
+// getCloakConfigFromAuth extracts cloak configuration from auth attributes.
+// Returns (cloakMode, strictMode, sensitiveWords).
+func getCloakConfigFromAuth(auth *cliproxyauth.Auth) (string, bool, []string) {
+	if auth == nil || auth.Attributes == nil {
+		return "auto", false, nil
+	}
+
+	cloakMode := auth.Attributes["cloak_mode"]
+	if cloakMode == "" {
+		cloakMode = "auto"
+	}
+
+	strictMode := strings.ToLower(auth.Attributes["cloak_strict_mode"]) == "true"
+
+	var sensitiveWords []string
+	if wordsStr := auth.Attributes["cloak_sensitive_words"]; wordsStr != "" {
+		sensitiveWords = strings.Split(wordsStr, ",")
+		for i := range sensitiveWords {
+			sensitiveWords[i] = strings.TrimSpace(sensitiveWords[i])
+		}
+	}
+
+	return cloakMode, strictMode, sensitiveWords
+}
+
+// resolveClaudeKeyCloakConfig finds the matching ClaudeKey config and returns its CloakConfig.
+func resolveClaudeKeyCloakConfig(cfg *config.Config, auth *cliproxyauth.Auth) *config.CloakConfig {
+	if cfg == nil || auth == nil {
+		return nil
+	}
+
+	apiKey, baseURL := claudeCreds(auth)
+	if apiKey == "" {
+		return nil
+	}
+
+	for i := range cfg.ClaudeKey {
+		entry := &cfg.ClaudeKey[i]
+		cfgKey := strings.TrimSpace(entry.APIKey)
+		cfgBase := strings.TrimSpace(entry.BaseURL)
+
+		// Match by API key
+		if strings.EqualFold(cfgKey, apiKey) {
+			// If baseURL is specified, also check it
+			if baseURL != "" && cfgBase != "" && !strings.EqualFold(cfgBase, baseURL) {
+				continue
+			}
+			return entry.Cloak
+		}
+	}
+
+	return nil
+}
+
+// injectFakeUserID generates and injects a fake user ID into the request metadata.
+func injectFakeUserID(payload []byte) []byte {
+	metadata := gjson.GetBytes(payload, "metadata")
+	if !metadata.Exists() {
+		payload, _ = sjson.SetBytes(payload, "metadata.user_id", generateFakeUserID())
+		return payload
+	}
+
+	existingUserID := gjson.GetBytes(payload, "metadata.user_id").String()
+	if existingUserID == "" || !isValidUserID(existingUserID) {
+		payload, _ = sjson.SetBytes(payload, "metadata.user_id", generateFakeUserID())
+	}
+	return payload
+}
+
+// checkSystemInstructionsWithMode injects Claude Code system prompt.
+// In strict mode, it replaces all user system messages.
+// In non-strict mode (default), it prepends to existing system messages.
+func checkSystemInstructionsWithMode(payload []byte, strictMode bool) []byte {
+	system := gjson.GetBytes(payload, "system")
+	claudeCodeInstructions := `[{"type":"text","text":"You are Claude Code, Anthropic's official CLI for Claude."}]`
+
+	if strictMode {
+		// Strict mode: replace all system messages with Claude Code prompt only
+		payload, _ = sjson.SetRawBytes(payload, "system", []byte(claudeCodeInstructions))
+		return payload
+	}
+
+	// Non-strict mode (default): prepend Claude Code prompt to existing system messages
+	if system.IsArray() {
+		if gjson.GetBytes(payload, "system.0.text").String() != "You are Claude Code, Anthropic's official CLI for Claude." {
+			system.ForEach(func(_, part gjson.Result) bool {
+				if part.Get("type").String() == "text" {
+					claudeCodeInstructions, _ = sjson.SetRaw(claudeCodeInstructions, "-1", part.Raw)
+				}
+				return true
+			})
+			payload, _ = sjson.SetRawBytes(payload, "system", []byte(claudeCodeInstructions))
+		}
+	} else {
+		payload, _ = sjson.SetRawBytes(payload, "system", []byte(claudeCodeInstructions))
+	}
+	return payload
+}
+
+// applyCloaking applies cloaking transformations to the payload based on config and client.
+// Cloaking includes: system prompt injection, fake user ID, and sensitive word obfuscation.
+func applyCloaking(ctx context.Context, cfg *config.Config, auth *cliproxyauth.Auth, payload []byte, model string) []byte {
+	clientUserAgent := getClientUserAgent(ctx)
+
+	// Get cloak config from ClaudeKey configuration
+	cloakCfg := resolveClaudeKeyCloakConfig(cfg, auth)
+
+	// Determine cloak settings
+	var cloakMode string
+	var strictMode bool
+	var sensitiveWords []string
+
+	if cloakCfg != nil {
+		cloakMode = cloakCfg.Mode
+		strictMode = cloakCfg.StrictMode
+		sensitiveWords = cloakCfg.SensitiveWords
+	}
+
+	// Fallback to auth attributes if no config found
+	if cloakMode == "" {
+		attrMode, attrStrict, attrWords := getCloakConfigFromAuth(auth)
+		cloakMode = attrMode
+		if !strictMode {
+			strictMode = attrStrict
+		}
+		if len(sensitiveWords) == 0 {
+			sensitiveWords = attrWords
+		}
+	}
+
+	// Determine if cloaking should be applied
+	if !shouldCloak(cloakMode, clientUserAgent) {
+		return payload
+	}
+
+	// Skip system instructions for claude-3-5-haiku models
+	if !strings.HasPrefix(model, "claude-3-5-haiku") {
+		payload = checkSystemInstructionsWithMode(payload, strictMode)
+	}
+
+	// Inject fake user ID
+	payload = injectFakeUserID(payload)
+
+	// Apply sensitive word obfuscation
+	if len(sensitiveWords) > 0 {
+		matcher := buildSensitiveWordMatcher(sensitiveWords)
+		payload = obfuscateSensitiveWords(payload, matcher)
+	}
+
+	return payload
+}

internal/runtime/executor/cloak_obfuscate.go

@@ -0,0 +1,176 @@
+package executor
+
+import (
+	"regexp"
+	"sort"
+	"strings"
+	"unicode/utf8"
+
+	"github.com/tidwall/gjson"
+	"github.com/tidwall/sjson"
+)
+
+// zeroWidthSpace is the Unicode zero-width space character used for obfuscation.
+const zeroWidthSpace = "\u200B"
+
+// SensitiveWordMatcher holds the compiled regex for matching sensitive words.
+type SensitiveWordMatcher struct {
+	regex *regexp.Regexp
+}
+
+// buildSensitiveWordMatcher compiles a regex from the word list.
+// Words are sorted by length (longest first) for proper matching.
+func buildSensitiveWordMatcher(words []string) *SensitiveWordMatcher {
+	if len(words) == 0 {
+		return nil
+	}
+
+	// Filter and normalize words
+	var validWords []string
+	for _, w := range words {
+		w = strings.TrimSpace(w)
+		if utf8.RuneCountInString(w) >= 2 && !strings.Contains(w, zeroWidthSpace) {
+			validWords = append(validWords, w)
+		}
+	}
+
+	if len(validWords) == 0 {
+		return nil
+	}
+
+	// Sort by length (longest first) for proper matching
+	sort.Slice(validWords, func(i, j int) bool {
+		return len(validWords[i]) > len(validWords[j])
+	})
+
+	// Escape and join
+	escaped := make([]string, len(validWords))
+	for i, w := range validWords {
+		escaped[i] = regexp.QuoteMeta(w)
+	}
+
+	pattern := "(?i)" + strings.Join(escaped, "|")
+	re, err := regexp.Compile(pattern)
+	if err != nil {
+		return nil
+	}
+
+	return &SensitiveWordMatcher{regex: re}
+}
+
+// obfuscateWord inserts a zero-width space after the first grapheme.
+func obfuscateWord(word string) string {
+	if strings.Contains(word, zeroWidthSpace) {
+		return word
+	}
+
+	// Get first rune
+	r, size := utf8.DecodeRuneInString(word)
+	if r == utf8.RuneError || size >= len(word) {
+		return word
+	}
+
+	return string(r) + zeroWidthSpace + word[size:]
+}
+
+// obfuscateText replaces all sensitive words in the text.
+func (m *SensitiveWordMatcher) obfuscateText(text string) string {
+	if m == nil || m.regex == nil {
+		return text
+	}
+	return m.regex.ReplaceAllStringFunc(text, obfuscateWord)
+}
+
+// obfuscateSensitiveWords processes the payload and obfuscates sensitive words
+// in system blocks and message content.
+func obfuscateSensitiveWords(payload []byte, matcher *SensitiveWordMatcher) []byte {
+	if matcher == nil || matcher.regex == nil {
+		return payload
+	}
+
+	// Obfuscate in system blocks
+	payload = obfuscateSystemBlocks(payload, matcher)
+
+	// Obfuscate in messages
+	payload = obfuscateMessages(payload, matcher)
+
+	return payload
+}
+
+// obfuscateSystemBlocks obfuscates sensitive words in system blocks.
+func obfuscateSystemBlocks(payload []byte, matcher *SensitiveWordMatcher) []byte {
+	system := gjson.GetBytes(payload, "system")
+	if !system.Exists() {
+		return payload
+	}
+
+	if system.IsArray() {
+		modified := false
+		system.ForEach(func(key, value gjson.Result) bool {
+			if value.Get("type").String() == "text" {
+				text := value.Get("text").String()
+				obfuscated := matcher.obfuscateText(text)
+				if obfuscated != text {
+					path := "system." + key.String() + ".text"
+					payload, _ = sjson.SetBytes(payload, path, obfuscated)
+					modified = true
+				}
+			}
+			return true
+		})
+		if modified {
+			return payload
+		}
+	} else if system.Type == gjson.String {
+		text := system.String()
+		obfuscated := matcher.obfuscateText(text)
+		if obfuscated != text {
+			payload, _ = sjson.SetBytes(payload, "system", obfuscated)
+		}
+	}
+
+	return payload
+}
+
+// obfuscateMessages obfuscates sensitive words in message content.
+func obfuscateMessages(payload []byte, matcher *SensitiveWordMatcher) []byte {
+	messages := gjson.GetBytes(payload, "messages")
+	if !messages.Exists() || !messages.IsArray() {
+		return payload
+	}
+
+	messages.ForEach(func(msgKey, msg gjson.Result) bool {
+		content := msg.Get("content")
+		if !content.Exists() {
+			return true
+		}
+
+		msgPath := "messages." + msgKey.String()
+
+		if content.Type == gjson.String {
+			// Simple string content
+			text := content.String()
+			obfuscated := matcher.obfuscateText(text)
+			if obfuscated != text {
+				payload, _ = sjson.SetBytes(payload, msgPath+".content", obfuscated)
+			}
+		} else if content.IsArray() {
+			// Array of content blocks
+			content.ForEach(func(blockKey, block gjson.Result) bool {
+				if block.Get("type").String() == "text" {
+					text := block.Get("text").String()
+					obfuscated := matcher.obfuscateText(text)
+					if obfuscated != text {
+						path := msgPath + ".content." + blockKey.String() + ".text"
+						payload, _ = sjson.SetBytes(payload, path, obfuscated)
+					}
+				}
+				return true
+			})
+		}
+
+		return true
+	})
+
+	return payload
+}

internal/runtime/executor/cloak_utils.go

@@ -0,0 +1,47 @@
+package executor
+
+import (
+	"crypto/rand"
+	"encoding/hex"
+	"regexp"
+	"strings"
+
+	"github.com/google/uuid"
+)
+
+// userIDPattern matches Claude Code format: user_[64-hex]_account__session_[uuid-v4]
+var userIDPattern = regexp.MustCompile(`^user_[a-fA-F0-9]{64}_account__session_[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$`)
+
+// generateFakeUserID generates a fake user ID in Claude Code format.
+// Format: user_[64-hex-chars]_account__session_[UUID-v4]
+func generateFakeUserID() string {
+	hexBytes := make([]byte, 32)
+	_, _ = rand.Read(hexBytes)
+	hexPart := hex.EncodeToString(hexBytes)
+	uuidPart := uuid.New().String()
+	return "user_" + hexPart + "_account__session_" + uuidPart
+}
+
+// isValidUserID checks if a user ID matches Claude Code format.
+func isValidUserID(userID string) bool {
+	return userIDPattern.MatchString(userID)
+}
+
+// shouldCloak determines if request should be cloaked based on config and client User-Agent.
+// Returns true if cloaking should be applied.
+func shouldCloak(cloakMode string, userAgent string) bool {
+	switch strings.ToLower(cloakMode) {
+	case "always":
+		return true
+	case "never":
+		return false
+	default: // "auto" or empty
+		// If client is Claude Code, don't cloak
+		return !strings.HasPrefix(userAgent, "claude-cli")
+	}
+}
+
+// isClaudeCodeClient checks if the User-Agent indicates a Claude Code client.
+func isClaudeCodeClient(userAgent string) bool {
+	return strings.HasPrefix(userAgent, "claude-cli")
+}

internal/runtime/executor/codex_executor.go

@@ -96,7 +96,7 @@ func (e *CodexExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, re
 	body = sdktranslator.TranslateRequest(from, to, baseModel, body, false)
 	body = misc.StripCodexUserAgent(body)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -208,7 +208,7 @@ func (e *CodexExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Au
 	body = sdktranslator.TranslateRequest(from, to, baseModel, body, true)
 	body = misc.StripCodexUserAgent(body)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -316,7 +316,7 @@ func (e *CodexExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.Auth
 	body = sdktranslator.TranslateRequest(from, to, baseModel, body, false)
 	body = misc.StripCodexUserAgent(body)
 
-	body, err := thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err := thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}

internal/runtime/executor/gemini_cli_executor.go

@@ -123,7 +123,7 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
 	basePayload := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	basePayload, err = thinking.ApplyThinking(basePayload, req.Model, from.String(), to.String())
+	basePayload, err = thinking.ApplyThinking(basePayload, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -272,7 +272,7 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	basePayload := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	basePayload, err = thinking.ApplyThinking(basePayload, req.Model, from.String(), to.String())
+	basePayload, err = thinking.ApplyThinking(basePayload, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -479,7 +479,7 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
 	for range models {
 		payload := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-		payload, err = thinking.ApplyThinking(payload, req.Model, from.String(), to.String())
+		payload, err = thinking.ApplyThinking(payload, req.Model, from.String(), to.String(), e.Identifier())
 		if err != nil {
 			return cliproxyexecutor.Response{}, err
 		}

internal/runtime/executor/gemini_executor.go

@@ -120,7 +120,7 @@ func (e *GeminiExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, r
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -222,7 +222,7 @@ func (e *GeminiExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.A
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -338,7 +338,7 @@ func (e *GeminiExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.Aut
 	to := sdktranslator.FromString("gemini")
 	translatedReq := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String())
+	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}

internal/runtime/executor/gemini_vertex_executor.go

@@ -12,6 +12,7 @@ import (
 	"io"
 	"net/http"
 	"strings"
+	"time"
 
 	vertexauth "github.com/router-for-me/CLIProxyAPI/v6/internal/auth/vertex"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
@@ -31,6 +32,143 @@ const (
 	vertexAPIVersion = "v1"
 )
 
+// isImagenModel checks if the model name is an Imagen image generation model.
+// Imagen models use the :predict action instead of :generateContent.
+func isImagenModel(model string) bool {
+	lowerModel := strings.ToLower(model)
+	return strings.Contains(lowerModel, "imagen")
+}
+
+// getVertexAction returns the appropriate action for the given model.
+// Imagen models use "predict", while Gemini models use "generateContent".
+func getVertexAction(model string, isStream bool) string {
+	if isImagenModel(model) {
+		return "predict"
+	}
+	if isStream {
+		return "streamGenerateContent"
+	}
+	return "generateContent"
+}
+
+// convertImagenToGeminiResponse converts Imagen API response to Gemini format
+// so it can be processed by the standard translation pipeline.
+// This ensures Imagen models return responses in the same format as gemini-3-pro-image-preview.
+func convertImagenToGeminiResponse(data []byte, model string) []byte {
+	predictions := gjson.GetBytes(data, "predictions")
+	if !predictions.Exists() || !predictions.IsArray() {
+		return data
+	}
+
+	// Build Gemini-compatible response with inlineData
+	parts := make([]map[string]any, 0)
+	for _, pred := range predictions.Array() {
+		imageData := pred.Get("bytesBase64Encoded").String()
+		mimeType := pred.Get("mimeType").String()
+		if mimeType == "" {
+			mimeType = "image/png"
+		}
+		if imageData != "" {
+			parts = append(parts, map[string]any{
+				"inlineData": map[string]any{
+					"mimeType": mimeType,
+					"data":     imageData,
+				},
+			})
+		}
+	}
+
+	// Generate unique response ID using timestamp
+	responseId := fmt.Sprintf("imagen-%d", time.Now().UnixNano())
+
+	response := map[string]any{
+		"candidates": []map[string]any{{
+			"content": map[string]any{
+				"parts": parts,
+				"role":  "model",
+			},
+			"finishReason": "STOP",
+		}},
+		"responseId":   responseId,
+		"modelVersion": model,
+		// Imagen API doesn't return token counts, set to 0 for tracking purposes
+		"usageMetadata": map[string]any{
+			"promptTokenCount":     0,
+			"candidatesTokenCount": 0,
+			"totalTokenCount":      0,
+		},
+	}
+
+	result, err := json.Marshal(response)
+	if err != nil {
+		return data
+	}
+	return result
+}
+
+// convertToImagenRequest converts a Gemini-style request to Imagen API format.
+// Imagen API uses a different structure: instances[].prompt instead of contents[].
+func convertToImagenRequest(payload []byte) ([]byte, error) {
+	// Extract prompt from Gemini-style contents
+	prompt := ""
+
+	// Try to get prompt from contents[0].parts[0].text
+	contentsText := gjson.GetBytes(payload, "contents.0.parts.0.text")
+	if contentsText.Exists() {
+		prompt = contentsText.String()
+	}
+
+	// If no contents, try messages format (OpenAI-compatible)
+	if prompt == "" {
+		messagesText := gjson.GetBytes(payload, "messages.#.content")
+		if messagesText.Exists() && messagesText.IsArray() {
+			for _, msg := range messagesText.Array() {
+				if msg.String() != "" {
+					prompt = msg.String()
+					break
+				}
+			}
+		}
+	}
+
+	// If still no prompt, try direct prompt field
+	if prompt == "" {
+		directPrompt := gjson.GetBytes(payload, "prompt")
+		if directPrompt.Exists() {
+			prompt = directPrompt.String()
+		}
+	}
+
+	if prompt == "" {
+		return nil, fmt.Errorf("imagen: no prompt found in request")
+	}
+
+	// Build Imagen API request
+	imagenReq := map[string]any{
+		"instances": []map[string]any{
+			{
+				"prompt": prompt,
+			},
+		},
+		"parameters": map[string]any{
+			"sampleCount": 1,
+		},
+	}
+
+	// Extract optional parameters
+	if aspectRatio := gjson.GetBytes(payload, "aspectRatio"); aspectRatio.Exists() {
+		imagenReq["parameters"].(map[string]any)["aspectRatio"] = aspectRatio.String()
+	}
+	if sampleCount := gjson.GetBytes(payload, "sampleCount"); sampleCount.Exists() {
+		imagenReq["parameters"].(map[string]any)["sampleCount"] = int(sampleCount.Int())
+	}
+	if negativePrompt := gjson.GetBytes(payload, "negativePrompt"); negativePrompt.Exists() {
+		imagenReq["instances"].([]map[string]any)[0]["negativePrompt"] = negativePrompt.String()
+	}
+
+	return json.Marshal(imagenReq)
+}
+
 // GeminiVertexExecutor sends requests to Vertex AI Gemini endpoints using service account credentials.
 type GeminiVertexExecutor struct {
 	cfg *config.Config
@@ -160,26 +298,38 @@ func (e *GeminiVertexExecutor) executeWithServiceAccount(ctx context.Context, au
 	reporter := newUsageReporter(ctx, e.Identifier(), baseModel, auth)
 	defer reporter.trackFailure(ctx, &err)
 
-	from := opts.SourceFormat
-	to := sdktranslator.FromString("gemini")
+	var body []byte
 
-	originalPayload := bytes.Clone(req.Payload)
-	if len(opts.OriginalRequest) > 0 {
-		originalPayload = bytes.Clone(opts.OriginalRequest)
-	}
-	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
-	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
+	// Handle Imagen models with special request format
+	if isImagenModel(baseModel) {
+		imagenBody, errImagen := convertToImagenRequest(req.Payload)
+		if errImagen != nil {
+			return resp, errImagen
+		}
+		body = imagenBody
+	} else {
+		// Standard Gemini translation flow
+		from := opts.SourceFormat
+		to := sdktranslator.FromString("gemini")
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
-	if err != nil {
-		return resp, err
+		originalPayload := bytes.Clone(req.Payload)
+		if len(opts.OriginalRequest) > 0 {
+			originalPayload = bytes.Clone(opts.OriginalRequest)
+		}
+		originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
+		body = sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
+
+		body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
+		if err != nil {
+			return resp, err
+		}
+
+		body = fixGeminiImageAspectRatio(baseModel, body)
+		body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
+		body, _ = sjson.SetBytes(body, "model", baseModel)
 	}
 
-	body = fixGeminiImageAspectRatio(baseModel, body)
-	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
-	body, _ = sjson.SetBytes(body, "model", baseModel)
-
-	action := "generateContent"
+	action := getVertexAction(baseModel, false)
 	if req.Metadata != nil {
 		if a, _ := req.Metadata["action"].(string); a == "countTokens" {
 			action = "countTokens"
@@ -249,6 +399,16 @@ func (e *GeminiVertexExecutor) executeWithServiceAccount(ctx context.Context, au
 	}
 	appendAPIResponseChunk(ctx, e.cfg, data)
 	reporter.publish(ctx, parseGeminiUsage(data))
+
+	// For Imagen models, convert response to Gemini format before translation
+	// This ensures Imagen responses use the same format as gemini-3-pro-image-preview
+	if isImagenModel(baseModel) {
+		data = convertImagenToGeminiResponse(data, baseModel)
+	}
+
+	// Standard Gemini translation (works for both Gemini and converted Imagen responses)
+	from := opts.SourceFormat
+	to := sdktranslator.FromString("gemini")
 	var param any
 	out := sdktranslator.TranslateNonStream(ctx, to, from, req.Model, bytes.Clone(opts.OriginalRequest), body, data, &param)
 	resp = cliproxyexecutor.Response{Payload: []byte(out)}
@@ -272,7 +432,7 @@ func (e *GeminiVertexExecutor) executeWithAPIKey(ctx context.Context, auth *clip
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, false)
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -281,7 +441,7 @@ func (e *GeminiVertexExecutor) executeWithAPIKey(ctx context.Context, auth *clip
 	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	action := "generateContent"
+	action := getVertexAction(baseModel, false)
 	if req.Metadata != nil {
 		if a, _ := req.Metadata["action"].(string); a == "countTokens" {
 			action = "countTokens"
@@ -375,7 +535,7 @@ func (e *GeminiVertexExecutor) executeStreamWithServiceAccount(ctx context.Conte
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -384,12 +544,16 @@ func (e *GeminiVertexExecutor) executeStreamWithServiceAccount(ctx context.Conte
 	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
+	action := getVertexAction(baseModel, true)
 	baseURL := vertexBaseURL(location)
-	url := fmt.Sprintf("%s/%s/projects/%s/locations/%s/publishers/google/models/%s:%s", baseURL, vertexAPIVersion, projectID, location, baseModel, "streamGenerateContent")
-	if opts.Alt == "" {
-		url = url + "?alt=sse"
-	} else {
-		url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+	url := fmt.Sprintf("%s/%s/projects/%s/locations/%s/publishers/google/models/%s:%s", baseURL, vertexAPIVersion, projectID, location, baseModel, action)
+	// Imagen models don't support streaming, skip SSE params
+	if !isImagenModel(baseModel) {
+		if opts.Alt == "" {
+			url = url + "?alt=sse"
+		} else {
+			url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+		}
 	}
 	body, _ = sjson.DeleteBytes(body, "session_id")
 
@@ -494,7 +658,7 @@ func (e *GeminiVertexExecutor) executeStreamWithAPIKey(ctx context.Context, auth
 	originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -503,15 +667,19 @@ func (e *GeminiVertexExecutor) executeStreamWithAPIKey(ctx context.Context, auth
 	body = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
+	action := getVertexAction(baseModel, true)
 	// For API key auth, use simpler URL format without project/location
 	if baseURL == "" {
 		baseURL = "https://generativelanguage.googleapis.com"
 	}
-	url := fmt.Sprintf("%s/%s/publishers/google/models/%s:%s", baseURL, vertexAPIVersion, baseModel, "streamGenerateContent")
-	if opts.Alt == "" {
-		url = url + "?alt=sse"
-	} else {
-		url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+	url := fmt.Sprintf("%s/%s/publishers/google/models/%s:%s", baseURL, vertexAPIVersion, baseModel, action)
+	// Imagen models don't support streaming, skip SSE params
+	if !isImagenModel(baseModel) {
+		if opts.Alt == "" {
+			url = url + "?alt=sse"
+		} else {
+			url = url + fmt.Sprintf("?$alt=%s", opts.Alt)
+		}
 	}
 	body, _ = sjson.DeleteBytes(body, "session_id")
 
@@ -605,7 +773,7 @@ func (e *GeminiVertexExecutor) countTokensWithServiceAccount(ctx context.Context
 
 	translatedReq := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String())
+	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}
@@ -689,7 +857,7 @@ func (e *GeminiVertexExecutor) countTokensWithAPIKey(ctx context.Context, auth *
 
 	translatedReq := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 
-	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String())
+	translatedReq, err := thinking.ApplyThinking(translatedReq, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}

internal/runtime/executor/iflow_executor.go

@@ -92,7 +92,7 @@ func (e *IFlowExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, re
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), "iflow")
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), "iflow", e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -190,7 +190,7 @@ func (e *IFlowExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Au
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), "iflow")
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), "iflow", e.Identifier())
 	if err != nil {
 		return nil, err
 	}

internal/runtime/executor/openai_compat_executor.go

@@ -92,7 +92,7 @@ func (e *OpenAICompatExecutor) Execute(ctx context.Context, auth *cliproxyauth.A
 	translated := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), opts.Stream)
 	translated = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated)
 
-	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -187,7 +187,7 @@ func (e *OpenAICompatExecutor) ExecuteStream(ctx context.Context, auth *cliproxy
 	translated := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 	translated = applyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated)
 
-	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}
@@ -297,7 +297,7 @@ func (e *OpenAICompatExecutor) CountTokens(ctx context.Context, auth *cliproxyau
 
 	modelForCounting := baseModel
 
-	translated, err := thinking.ApplyThinking(translated, req.Model, from.String(), to.String())
+	translated, err := thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return cliproxyexecutor.Response{}, err
 	}

internal/runtime/executor/qwen_executor.go

@@ -86,7 +86,7 @@ func (e *QwenExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, req
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), false)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return resp, err
 	}
@@ -172,7 +172,7 @@ func (e *QwenExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Aut
 	body := sdktranslator.TranslateRequest(from, to, baseModel, bytes.Clone(req.Payload), true)
 	body, _ = sjson.SetBytes(body, "model", baseModel)
 
-	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String())
+	body, err = thinking.ApplyThinking(body, req.Model, from.String(), to.String(), e.Identifier())
 	if err != nil {
 		return nil, err
 	}

internal/thinking/apply.go

@@ -63,6 +63,7 @@ func IsUserDefinedModel(modelInfo *registry.ModelInfo) bool {
 //   - model: Model name, optionally with thinking suffix (e.g., "claude-sonnet-4-5(16384)")
 //   - fromFormat: Source request format (e.g., openai, codex, gemini)
 //   - toFormat: Target provider format for the request body (gemini, gemini-cli, antigravity, claude, openai, codex, iflow)
+//   - providerKey: Provider identifier used for registry model lookups (may differ from toFormat, e.g., openrouter -> openai)
 //
 // Returns:
 //   - Modified request body JSON with thinking configuration applied
@@ -79,12 +80,16 @@ func IsUserDefinedModel(modelInfo *registry.ModelInfo) bool {
 // Example:
 //
 //	// With suffix - suffix config takes priority
-//	result, err := thinking.ApplyThinking(body, "gemini-2.5-pro(8192)", "gemini", "gemini")
+//	result, err := thinking.ApplyThinking(body, "gemini-2.5-pro(8192)", "gemini", "gemini", "gemini")
 //
 //	// Without suffix - uses body config
-//	result, err := thinking.ApplyThinking(body, "gemini-2.5-pro", "gemini", "gemini")
-func ApplyThinking(body []byte, model string, fromFormat string, toFormat string) ([]byte, error) {
+//	result, err := thinking.ApplyThinking(body, "gemini-2.5-pro", "gemini", "gemini", "gemini")
+func ApplyThinking(body []byte, model string, fromFormat string, toFormat string, providerKey string) ([]byte, error) {
 	providerFormat := strings.ToLower(strings.TrimSpace(toFormat))
+	providerKey = strings.ToLower(strings.TrimSpace(providerKey))
+	if providerKey == "" {
+		providerKey = providerFormat
+	}
 	fromFormat = strings.ToLower(strings.TrimSpace(fromFormat))
 	if fromFormat == "" {
 		fromFormat = providerFormat
@@ -102,7 +107,8 @@ func ApplyThinking(body []byte, model string, fromFormat string, toFormat string
 	// 2. Parse suffix and get modelInfo
 	suffixResult := ParseSuffix(model)
 	baseModel := suffixResult.ModelName
-	modelInfo := registry.LookupModelInfo(baseModel)
+	// Use provider-specific lookup to handle capability differences across providers.
+	modelInfo := registry.LookupModelInfo(baseModel, providerKey)
 
 	// 3. Model capability check
 	// Unknown models are treated as user-defined so thinking config can still be applied.

internal/thinking/provider/claude/apply.go

@@ -80,9 +80,66 @@ func (a *Applier) Apply(body []byte, config thinking.ThinkingConfig, modelInfo *
 
 	result, _ := sjson.SetBytes(body, "thinking.type", "enabled")
 	result, _ = sjson.SetBytes(result, "thinking.budget_tokens", config.Budget)
+
+	// Ensure max_tokens > thinking.budget_tokens (Anthropic API constraint)
+	result = a.normalizeClaudeBudget(result, config.Budget, modelInfo)
 	return result, nil
 }
 
+// normalizeClaudeBudget applies Claude-specific constraints to ensure max_tokens > budget_tokens.
+// Anthropic API requires this constraint; violating it returns a 400 error.
+func (a *Applier) normalizeClaudeBudget(body []byte, budgetTokens int, modelInfo *registry.ModelInfo) []byte {
+	if budgetTokens <= 0 {
+		return body
+	}
+
+	// Ensure the request satisfies Claude constraints:
+	//  1) Determine effective max_tokens (request overrides model default)
+	//  2) If budget_tokens >= max_tokens, reduce budget_tokens to max_tokens-1
+	//  3) If the adjusted budget falls below the model minimum, leave the request unchanged
+	//  4) If max_tokens came from model default, write it back into the request
+
+	effectiveMax, setDefaultMax := a.effectiveMaxTokens(body, modelInfo)
+	if setDefaultMax && effectiveMax > 0 {
+		body, _ = sjson.SetBytes(body, "max_tokens", effectiveMax)
+	}
+
+	// Compute the budget we would apply after enforcing budget_tokens < max_tokens.
+	adjustedBudget := budgetTokens
+	if effectiveMax > 0 && adjustedBudget >= effectiveMax {
+		adjustedBudget = effectiveMax - 1
+	}
+
+	minBudget := 0
+	if modelInfo != nil && modelInfo.Thinking != nil {
+		minBudget = modelInfo.Thinking.Min
+	}
+	if minBudget > 0 && adjustedBudget > 0 && adjustedBudget < minBudget {
+		// If enforcing the max_tokens constraint would push the budget below the model minimum,
+		// leave the request unchanged.
+		return body
+	}
+
+	if adjustedBudget != budgetTokens {
+		body, _ = sjson.SetBytes(body, "thinking.budget_tokens", adjustedBudget)
+	}
+
+	return body
+}
+
+// effectiveMaxTokens returns the max tokens to cap thinking:
+// prefer request-provided max_tokens; otherwise fall back to model default.
+// The boolean indicates whether the value came from the model default (and thus should be written back).
+func (a *Applier) effectiveMaxTokens(body []byte, modelInfo *registry.ModelInfo) (max int, fromModel bool) {
+	if maxTok := gjson.GetBytes(body, "max_tokens"); maxTok.Exists() && maxTok.Int() > 0 {
+		return int(maxTok.Int()), false
+	}
+	if modelInfo != nil && modelInfo.MaxCompletionTokens > 0 {
+		return modelInfo.MaxCompletionTokens, true
+	}
+	return 0, false
+}
+
 func applyCompatibleClaude(body []byte, config thinking.ThinkingConfig) ([]byte, error) {
 	if config.Mode != thinking.ModeBudget && config.Mode != thinking.ModeNone && config.Mode != thinking.ModeAuto {
 		return body, nil

internal/translator/antigravity/claude/antigravity_claude_request.go

@@ -7,8 +7,6 @@ package claude
 
 import (
 	"bytes"
-	"crypto/sha256"
-	"encoding/hex"
 	"strings"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/cache"
@@ -19,29 +17,6 @@ import (
 	"github.com/tidwall/sjson"
 )
 
-// deriveSessionID generates a stable session ID from the request.
-// Uses the hash of the first user message to identify the conversation.
-func deriveSessionID(rawJSON []byte) string {
-	messages := gjson.GetBytes(rawJSON, "messages")
-	if !messages.IsArray() {
-		return ""
-	}
-	for _, msg := range messages.Array() {
-		if msg.Get("role").String() == "user" {
-			content := msg.Get("content").String()
-			if content == "" {
-				// Try to get text from content array
-				content = msg.Get("content.0.text").String()
-			}
-			if content != "" {
-				h := sha256.Sum256([]byte(content))
-				return hex.EncodeToString(h[:16])
-			}
-		}
-	}
-	return ""
-}
-
 // ConvertClaudeRequestToAntigravity parses and transforms a Claude Code API request into Gemini CLI API format.
 // It extracts the model name, system instruction, message contents, and tool declarations
 // from the raw JSON request and returns them in the format expected by the Gemini CLI API.
@@ -61,11 +36,9 @@ func deriveSessionID(rawJSON []byte) string {
 // Returns:
 //   - []byte: The transformed request data in Gemini CLI API format
 func ConvertClaudeRequestToAntigravity(modelName string, inputRawJSON []byte, _ bool) []byte {
+	enableThoughtTranslate := true
 	rawJSON := bytes.Clone(inputRawJSON)
 
-	// Derive session ID for signature caching
-	sessionID := deriveSessionID(rawJSON)
-
 	// system instruction
 	systemInstructionJSON := ""
 	hasSystemInstruction := false
@@ -124,41 +97,49 @@ func ConvertClaudeRequestToAntigravity(modelName string, inputRawJSON []byte, _
 					if contentTypeResult.Type == gjson.String && contentTypeResult.String() == "thinking" {
 						// Use GetThinkingText to handle wrapped thinking objects
 						thinkingText := thinking.GetThinkingText(contentResult)
-						signatureResult := contentResult.Get("signature")
-						clientSignature := ""
-						if signatureResult.Exists() && signatureResult.String() != "" {
-							clientSignature = signatureResult.String()
-						}
 
 						// Always try cached signature first (more reliable than client-provided)
 						// Client may send stale or invalid signatures from different sessions
 						signature := ""
-						if sessionID != "" && thinkingText != "" {
-							if cachedSig := cache.GetCachedSignature(sessionID, thinkingText); cachedSig != "" {
+						if thinkingText != "" {
+							if cachedSig := cache.GetCachedSignature(modelName, thinkingText); cachedSig != "" {
 								signature = cachedSig
 								// log.Debugf("Using cached signature for thinking block")
 							}
 						}
 
 						// Fallback to client signature only if cache miss and client signature is valid
-						if signature == "" && cache.HasValidSignature(clientSignature) {
-							signature = clientSignature
+						if signature == "" {
+							signatureResult := contentResult.Get("signature")
+							clientSignature := ""
+							if signatureResult.Exists() && signatureResult.String() != "" {
+								arrayClientSignatures := strings.SplitN(signatureResult.String(), "#", 2)
+								if len(arrayClientSignatures) == 2 {
+									if modelName == arrayClientSignatures[0] {
+										clientSignature = arrayClientSignatures[1]
+									}
+								}
+							}
+							if cache.HasValidSignature(modelName, clientSignature) {
+								signature = clientSignature
+							}
 							// log.Debugf("Using client-provided signature for thinking block")
 						}
 
 						// Store for subsequent tool_use in the same message
-						if cache.HasValidSignature(signature) {
+						if cache.HasValidSignature(modelName, signature) {
 							currentMessageThinkingSignature = signature
 						}
 
 						// Skip trailing unsigned thinking blocks on last assistant message
-						isUnsigned := !cache.HasValidSignature(signature)
+						isUnsigned := !cache.HasValidSignature(modelName, signature)
 
 						// If unsigned, skip entirely (don't convert to text)
 						// Claude requires assistant messages to start with thinking blocks when thinking is enabled
 						// Converting to text would break this requirement
 						if isUnsigned {
 							// log.Debugf("Dropping unsigned thinking block (no valid signature)")
+							enableThoughtTranslate = false
 							continue
 						}
 
@@ -206,7 +187,7 @@ func ConvertClaudeRequestToAntigravity(modelName string, inputRawJSON []byte, _
 							// This is the approach used in opencode-google-antigravity-auth for Gemini
 							// and also works for Claude through Antigravity API
 							const skipSentinel = "skip_thought_signature_validator"
-							if cache.HasValidSignature(currentMessageThinkingSignature) {
+							if cache.HasValidSignature(modelName, currentMessageThinkingSignature) {
 								partJSON, _ = sjson.Set(partJSON, "thoughtSignature", currentMessageThinkingSignature)
 							} else {
 								// No valid signature - use skip sentinel to bypass validation
@@ -386,7 +367,7 @@ func ConvertClaudeRequestToAntigravity(modelName string, inputRawJSON []byte, _
 	}
 
 	// Map Anthropic thinking -> Gemini thinkingBudget/include_thoughts when type==enabled
-	if t := gjson.GetBytes(rawJSON, "thinking"); t.Exists() && t.IsObject() {
+	if t := gjson.GetBytes(rawJSON, "thinking"); enableThoughtTranslate && t.Exists() && t.IsObject() {
 		if t.Get("type").String() == "enabled" {
 			if b := t.Get("budget_tokens"); b.Exists() && b.Type == gjson.Number {
 				budget := int(b.Int())

internal/translator/antigravity/claude/antigravity_claude_request_test.go

@@ -4,6 +4,7 @@ import (
 	"strings"
 	"testing"
 
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/cache"
 	"github.com/tidwall/gjson"
 )
 
@@ -75,28 +76,39 @@ func TestConvertClaudeRequestToAntigravity_RoleMapping(t *testing.T) {
 func TestConvertClaudeRequestToAntigravity_ThinkingBlocks(t *testing.T) {
 	// Valid signature must be at least 50 characters
 	validSignature := "abc123validSignature1234567890123456789012345678901234567890"
+	thinkingText := "Let me think..."
+
+	// Pre-cache the signature (simulating a response from the same session)
+	// The session ID is derived from the first user message hash
+	// Since there's no user message in this test, we need to add one
 	inputJSON := []byte(`{
 		"model": "claude-sonnet-4-5-thinking",
 		"messages": [
+			{
+				"role": "user",
+				"content": [{"type": "text", "text": "Test user message"}]
+			},
 			{
 				"role": "assistant",
 				"content": [
-					{"type": "thinking", "thinking": "Let me think...", "signature": "` + validSignature + `"},
+					{"type": "thinking", "thinking": "` + thinkingText + `", "signature": "` + validSignature + `"},
 					{"type": "text", "text": "Answer"}
 				]
 			}
 		]
 	}`)
 
+	cache.CacheSignature("claude-sonnet-4-5-thinking", thinkingText, validSignature)
+
 	output := ConvertClaudeRequestToAntigravity("claude-sonnet-4-5-thinking", inputJSON, false)
 	outputStr := string(output)
 
-	// Check thinking block conversion
-	firstPart := gjson.Get(outputStr, "request.contents.0.parts.0")
+	// Check thinking block conversion (now in contents.1 due to user message)
+	firstPart := gjson.Get(outputStr, "request.contents.1.parts.0")
 	if !firstPart.Get("thought").Bool() {
 		t.Error("thinking block should have thought: true")
 	}
-	if firstPart.Get("text").String() != "Let me think..." {
+	if firstPart.Get("text").String() != thinkingText {
 		t.Error("thinking text mismatch")
 	}
 	if firstPart.Get("thoughtSignature").String() != validSignature {
@@ -227,13 +239,19 @@ func TestConvertClaudeRequestToAntigravity_ToolUse(t *testing.T) {
 
 func TestConvertClaudeRequestToAntigravity_ToolUse_WithSignature(t *testing.T) {
 	validSignature := "abc123validSignature1234567890123456789012345678901234567890"
+	thinkingText := "Let me think..."
+
 	inputJSON := []byte(`{
 		"model": "claude-sonnet-4-5-thinking",
 		"messages": [
+			{
+				"role": "user",
+				"content": [{"type": "text", "text": "Test user message"}]
+			},
 			{
 				"role": "assistant",
 				"content": [
-					{"type": "thinking", "thinking": "Let me think...", "signature": "` + validSignature + `"},
+					{"type": "thinking", "thinking": "` + thinkingText + `", "signature": "` + validSignature + `"},
 					{
 						"type": "tool_use",
 						"id": "call_123",
@@ -245,11 +263,13 @@ func TestConvertClaudeRequestToAntigravity_ToolUse_WithSignature(t *testing.T) {
 		]
 	}`)
 
+	cache.CacheSignature("claude-sonnet-4-5-thinking", thinkingText, validSignature)
+
 	output := ConvertClaudeRequestToAntigravity("claude-sonnet-4-5-thinking", inputJSON, false)
 	outputStr := string(output)
 
-	// Check function call has the signature from the preceding thinking block
-	part := gjson.Get(outputStr, "request.contents.0.parts.1")
+	// Check function call has the signature from the preceding thinking block (now in contents.1)
+	part := gjson.Get(outputStr, "request.contents.1.parts.1")
 	if part.Get("functionCall.name").String() != "get_weather" {
 		t.Errorf("Expected functionCall, got %s", part.Raw)
 	}
@@ -261,24 +281,32 @@ func TestConvertClaudeRequestToAntigravity_ToolUse_WithSignature(t *testing.T) {
 func TestConvertClaudeRequestToAntigravity_ReorderThinking(t *testing.T) {
 	// Case: text block followed by thinking block -> should be reordered to thinking first
 	validSignature := "abc123validSignature1234567890123456789012345678901234567890"
+	thinkingText := "Planning..."
+
 	inputJSON := []byte(`{
 		"model": "claude-sonnet-4-5-thinking",
 		"messages": [
+			{
+				"role": "user",
+				"content": [{"type": "text", "text": "Test user message"}]
+			},
 			{
 				"role": "assistant",
 				"content": [
 					{"type": "text", "text": "Here is the plan."},
-					{"type": "thinking", "thinking": "Planning...", "signature": "` + validSignature + `"}
+					{"type": "thinking", "thinking": "` + thinkingText + `", "signature": "` + validSignature + `"}
 				]
 			}
 		]
 	}`)
 
+	cache.CacheSignature("claude-sonnet-4-5-thinking", thinkingText, validSignature)
+
 	output := ConvertClaudeRequestToAntigravity("claude-sonnet-4-5-thinking", inputJSON, false)
 	outputStr := string(output)
 
-	// Verify order: Thinking block MUST be first
-	parts := gjson.Get(outputStr, "request.contents.0.parts").Array()
+	// Verify order: Thinking block MUST be first (now in contents.1 due to user message)
+	parts := gjson.Get(outputStr, "request.contents.1.parts").Array()
 	if len(parts) != 2 {
 		t.Fatalf("Expected 2 parts, got %d", len(parts))
 	}
@@ -460,6 +488,9 @@ func TestConvertClaudeRequestToAntigravity_TrailingUnsignedThinking_Removed(t *t
 
 func TestConvertClaudeRequestToAntigravity_TrailingSignedThinking_Kept(t *testing.T) {
 	// Last assistant message ends with signed thinking block - should be kept
+	validSignature := "abc123validSignature1234567890123456789012345678901234567890"
+	thinkingText := "Valid thinking..."
+
 	inputJSON := []byte(`{
 		"model": "claude-sonnet-4-5-thinking",
 		"messages": [
@@ -471,12 +502,14 @@ func TestConvertClaudeRequestToAntigravity_TrailingSignedThinking_Kept(t *testin
 				"role": "assistant",
 				"content": [
 					{"type": "text", "text": "Here is my answer"},
-					{"type": "thinking", "thinking": "Valid thinking...", "signature": "abc123validSignature1234567890123456789012345678901234567890"}
+					{"type": "thinking", "thinking": "` + thinkingText + `", "signature": "` + validSignature + `"}
 				]
 			}
 		]
 	}`)
 
+	cache.CacheSignature("claude-sonnet-4-5-thinking", thinkingText, validSignature)
+
 	output := ConvertClaudeRequestToAntigravity("claude-sonnet-4-5-thinking", inputJSON, false)
 	outputStr := string(output)
 

internal/translator/antigravity/claude/antigravity_claude_response.go

@@ -41,7 +41,6 @@ type Params struct {
 	HasContent           bool   // Tracks whether any content (text, thinking, or tool use) has been output
 
 	// Signature caching support
-	SessionID           string          // Session ID derived from request for signature caching
 	CurrentThinkingText strings.Builder // Accumulates thinking text for signature caching
 }
 
@@ -70,9 +69,9 @@ func ConvertAntigravityResponseToClaude(_ context.Context, _ string, originalReq
 			HasFirstResponse: false,
 			ResponseType:     0,
 			ResponseIndex:    0,
-			SessionID:        deriveSessionID(originalRequestRawJSON),
 		}
 	}
+	modelName := gjson.GetBytes(requestRawJSON, "model").String()
 
 	params := (*param).(*Params)
 
@@ -138,14 +137,14 @@ func ConvertAntigravityResponseToClaude(_ context.Context, _ string, originalReq
 					if thoughtSignature := partResult.Get("thoughtSignature"); thoughtSignature.Exists() && thoughtSignature.String() != "" {
 						// log.Debug("Branch: signature_delta")
 
-						if params.SessionID != "" && params.CurrentThinkingText.Len() > 0 {
-							cache.CacheSignature(params.SessionID, params.CurrentThinkingText.String(), thoughtSignature.String())
+						if params.CurrentThinkingText.Len() > 0 {
+							cache.CacheSignature(modelName, params.CurrentThinkingText.String(), thoughtSignature.String())
 							// log.Debugf("Cached signature for thinking block (sessionID=%s, textLen=%d)", params.SessionID, params.CurrentThinkingText.Len())
 							params.CurrentThinkingText.Reset()
 						}
 
 						output = output + "event: content_block_delta\n"
-						data, _ := sjson.Set(fmt.Sprintf(`{"type":"content_block_delta","index":%d,"delta":{"type":"signature_delta","signature":""}}`, params.ResponseIndex), "delta.signature", thoughtSignature.String())
+						data, _ := sjson.Set(fmt.Sprintf(`{"type":"content_block_delta","index":%d,"delta":{"type":"signature_delta","signature":""}}`, params.ResponseIndex), "delta.signature", fmt.Sprintf("%s#%s", cache.GetModelGroup(modelName), thoughtSignature.String()))
 						output = output + fmt.Sprintf("data: %s\n\n\n", data)
 						params.HasContent = true
 					} else if params.ResponseType == 2 { // Continue existing thinking block if already in thinking state
@@ -372,7 +371,7 @@ func resolveStopReason(params *Params) string {
 //   - string: A Claude-compatible JSON response.
 func ConvertAntigravityResponseToClaudeNonStream(_ context.Context, _ string, originalRequestRawJSON, requestRawJSON, rawJSON []byte, _ *any) string {
 	_ = originalRequestRawJSON
-	_ = requestRawJSON
+	modelName := gjson.GetBytes(requestRawJSON, "model").String()
 
 	root := gjson.ParseBytes(rawJSON)
 	promptTokens := root.Get("response.usageMetadata.promptTokenCount").Int()
@@ -437,7 +436,7 @@ func ConvertAntigravityResponseToClaudeNonStream(_ context.Context, _ string, or
 		block := `{"type":"thinking","thinking":""}`
 		block, _ = sjson.Set(block, "thinking", thinkingBuilder.String())
 		if thinkingSignature != "" {
-			block, _ = sjson.Set(block, "signature", thinkingSignature)
+			block, _ = sjson.Set(block, "signature", fmt.Sprintf("%s#%s", cache.GetModelGroup(modelName), thinkingSignature))
 		}
 		responseJSON, _ = sjson.SetRaw(responseJSON, "content.-1", block)
 		thinkingBuilder.Reset()

internal/translator/antigravity/claude/antigravity_claude_response_test.go

@@ -97,6 +97,7 @@ func TestConvertAntigravityResponseToClaude_SignatureCached(t *testing.T) {
 	cache.ClearSignatureCache("")
 
 	requestJSON := []byte(`{
+		"model": "claude-sonnet-4-5-thinking",
 		"messages": [{"role": "user", "content": [{"type": "text", "text": "Cache test"}]}]
 	}`)
 
@@ -143,7 +144,7 @@ func TestConvertAntigravityResponseToClaude_SignatureCached(t *testing.T) {
 	ConvertAntigravityResponseToClaude(ctx, "claude-sonnet-4-5-thinking", requestJSON, requestJSON, signatureChunk, &param)
 
 	// Verify signature was cached
-	cachedSig := cache.GetCachedSignature(sessionID, thinkingText)
+	cachedSig := cache.GetCachedSignature("claude-sonnet-4-5-thinking", thinkingText)
 	if cachedSig != validSignature {
 		t.Errorf("Expected cached signature '%s', got '%s'", validSignature, cachedSig)
 	}
@@ -158,6 +159,7 @@ func TestConvertAntigravityResponseToClaude_MultipleThinkingBlocks(t *testing.T)
 	cache.ClearSignatureCache("")
 
 	requestJSON := []byte(`{
+		"model": "claude-sonnet-4-5-thinking",
 		"messages": [{"role": "user", "content": [{"type": "text", "text": "Multi block test"}]}]
 	}`)
 
@@ -221,13 +223,12 @@ func TestConvertAntigravityResponseToClaude_MultipleThinkingBlocks(t *testing.T)
 	// Process first thinking block
 	ConvertAntigravityResponseToClaude(ctx, "claude-sonnet-4-5-thinking", requestJSON, requestJSON, block1Thinking, &param)
 	params := param.(*Params)
-	sessionID := params.SessionID
 	firstThinkingText := params.CurrentThinkingText.String()
 
 	ConvertAntigravityResponseToClaude(ctx, "claude-sonnet-4-5-thinking", requestJSON, requestJSON, block1Sig, &param)
 
 	// Verify first signature cached
-	if cache.GetCachedSignature(sessionID, firstThinkingText) != validSig1 {
+	if cache.GetCachedSignature("claude-sonnet-4-5-thinking", firstThinkingText) != validSig1 {
 		t.Error("First thinking block signature should be cached")
 	}
 
@@ -241,76 +242,7 @@ func TestConvertAntigravityResponseToClaude_MultipleThinkingBlocks(t *testing.T)
 	ConvertAntigravityResponseToClaude(ctx, "claude-sonnet-4-5-thinking", requestJSON, requestJSON, block2Sig, &param)
 
 	// Verify second signature cached
-	if cache.GetCachedSignature(sessionID, secondThinkingText) != validSig2 {
+	if cache.GetCachedSignature("claude-sonnet-4-5-thinking", secondThinkingText) != validSig2 {
 		t.Error("Second thinking block signature should be cached")
 	}
 }
-
-func TestDeriveSessionIDFromRequest(t *testing.T) {
-	tests := []struct {
-		name      string
-		input     []byte
-		wantEmpty bool
-	}{
-		{
-			name:      "valid user message",
-			input:     []byte(`{"messages": [{"role": "user", "content": "Hello"}]}`),
-			wantEmpty: false,
-		},
-		{
-			name:      "user message with content array",
-			input:     []byte(`{"messages": [{"role": "user", "content": [{"type": "text", "text": "Hello"}]}]}`),
-			wantEmpty: false,
-		},
-		{
-			name:      "no user message",
-			input:     []byte(`{"messages": [{"role": "assistant", "content": "Hi"}]}`),
-			wantEmpty: true,
-		},
-		{
-			name:      "empty messages",
-			input:     []byte(`{"messages": []}`),
-			wantEmpty: true,
-		},
-		{
-			name:      "no messages field",
-			input:     []byte(`{}`),
-			wantEmpty: true,
-		},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			result := deriveSessionID(tt.input)
-			if tt.wantEmpty && result != "" {
-				t.Errorf("Expected empty session ID, got '%s'", result)
-			}
-			if !tt.wantEmpty && result == "" {
-				t.Error("Expected non-empty session ID")
-			}
-		})
-	}
-}
-
-func TestDeriveSessionIDFromRequest_Deterministic(t *testing.T) {
-	input := []byte(`{"messages": [{"role": "user", "content": "Same message"}]}`)
-
-	id1 := deriveSessionID(input)
-	id2 := deriveSessionID(input)
-
-	if id1 != id2 {
-		t.Errorf("Session ID should be deterministic: '%s' != '%s'", id1, id2)
-	}
-}
-
-func TestDeriveSessionIDFromRequest_DifferentMessages(t *testing.T) {
-	input1 := []byte(`{"messages": [{"role": "user", "content": "Message A"}]}`)
-	input2 := []byte(`{"messages": [{"role": "user", "content": "Message B"}]}`)
-
-	id1 := deriveSessionID(input1)
-	id2 := deriveSessionID(input2)
-
-	if id1 == id2 {
-		t.Error("Different messages should produce different session IDs")
-	}
-}

internal/translator/antigravity/gemini/antigravity_gemini_request.go

@@ -8,6 +8,7 @@ package gemini
 import (
 	"bytes"
 	"fmt"
+	"strings"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/translator/gemini/common"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
@@ -32,12 +33,12 @@ import (
 //
 // Returns:
 //   - []byte: The transformed request data in Gemini API format
-func ConvertGeminiRequestToAntigravity(_ string, inputRawJSON []byte, _ bool) []byte {
+func ConvertGeminiRequestToAntigravity(modelName string, inputRawJSON []byte, _ bool) []byte {
 	rawJSON := bytes.Clone(inputRawJSON)
 	template := ""
 	template = `{"project":"","request":{},"model":""}`
 	template, _ = sjson.SetRaw(template, "request", string(rawJSON))
-	template, _ = sjson.Set(template, "model", gjson.Get(template, "request.model").String())
+	template, _ = sjson.Set(template, "model", modelName)
 	template, _ = sjson.Delete(template, "request.model")
 
 	template, errFixCLIToolResponse := fixCLIToolResponse(template)
@@ -97,37 +98,40 @@ func ConvertGeminiRequestToAntigravity(_ string, inputRawJSON []byte, _ bool) []
 		}
 	}
 
-	// Gemini-specific handling: add skip_thought_signature_validator to functionCall parts
-	// and remove thinking blocks entirely (Gemini doesn't need to preserve them)
-	const skipSentinel = "skip_thought_signature_validator"
+	// Gemini-specific handling for non-Claude models:
+	// - Add skip_thought_signature_validator to functionCall parts so upstream can bypass signature validation.
+	// - Also mark thinking parts with the same sentinel when present (we keep the parts; we only annotate them).
+	if !strings.Contains(modelName, "claude") {
+		const skipSentinel = "skip_thought_signature_validator"
 
-	gjson.GetBytes(rawJSON, "request.contents").ForEach(func(contentIdx, content gjson.Result) bool {
-		if content.Get("role").String() == "model" {
-			// First pass: collect indices of thinking parts to remove
-			var thinkingIndicesToRemove []int64
-			content.Get("parts").ForEach(func(partIdx, part gjson.Result) bool {
-				// Mark thinking blocks for removal
-				if part.Get("thought").Bool() {
-					thinkingIndicesToRemove = append(thinkingIndicesToRemove, partIdx.Int())
-				}
-				// Add skip sentinel to functionCall parts
-				if part.Get("functionCall").Exists() {
-					existingSig := part.Get("thoughtSignature").String()
-					if existingSig == "" || len(existingSig) < 50 {
-						rawJSON, _ = sjson.SetBytes(rawJSON, fmt.Sprintf("request.contents.%d.parts.%d.thoughtSignature", contentIdx.Int(), partIdx.Int()), skipSentinel)
+		gjson.GetBytes(rawJSON, "request.contents").ForEach(func(contentIdx, content gjson.Result) bool {
+			if content.Get("role").String() == "model" {
+				// First pass: collect indices of thinking parts to mark with skip sentinel
+				var thinkingIndicesToSkipSignature []int64
+				content.Get("parts").ForEach(func(partIdx, part gjson.Result) bool {
+					// Collect indices of thinking blocks to mark with skip sentinel
+					if part.Get("thought").Bool() {
+						thinkingIndicesToSkipSignature = append(thinkingIndicesToSkipSignature, partIdx.Int())
 					}
-				}
-				return true
-			})
+					// Add skip sentinel to functionCall parts
+					if part.Get("functionCall").Exists() {
+						existingSig := part.Get("thoughtSignature").String()
+						if existingSig == "" || len(existingSig) < 50 {
+							rawJSON, _ = sjson.SetBytes(rawJSON, fmt.Sprintf("request.contents.%d.parts.%d.thoughtSignature", contentIdx.Int(), partIdx.Int()), skipSentinel)
+						}
+					}
+					return true
+				})
 
-			// Remove thinking blocks in reverse order to preserve indices
-			for i := len(thinkingIndicesToRemove) - 1; i >= 0; i-- {
-				idx := thinkingIndicesToRemove[i]
-				rawJSON, _ = sjson.DeleteBytes(rawJSON, fmt.Sprintf("request.contents.%d.parts.%d", contentIdx.Int(), idx))
+				// Add skip_thought_signature_validator sentinel to thinking blocks in reverse order to preserve indices
+				for i := len(thinkingIndicesToSkipSignature) - 1; i >= 0; i-- {
+					idx := thinkingIndicesToSkipSignature[i]
+					rawJSON, _ = sjson.SetBytes(rawJSON, fmt.Sprintf("request.contents.%d.parts.%d.thoughtSignature", contentIdx.Int(), idx), skipSentinel)
+				}
 			}
-		}
-		return true
-	})
+			return true
+		})
+	}
 
 	return common.AttachDefaultSafetySettings(rawJSON, "request.safetySettings")
 }

internal/translator/antigravity/openai/chat-completions/antigravity_openai_request.go

@@ -66,6 +66,13 @@ func ConvertOpenAIRequestToAntigravity(modelName string, inputRawJSON []byte, _
 		out, _ = sjson.SetBytes(out, "request.generationConfig.maxOutputTokens", maxTok.Num)
 	}
 
+	// Candidate count (OpenAI 'n' parameter)
+	if n := gjson.GetBytes(rawJSON, "n"); n.Exists() && n.Type == gjson.Number {
+		if val := n.Int(); val > 1 {
+			out, _ = sjson.SetBytes(out, "request.generationConfig.candidateCount", val)
+		}
+	}
+
 	// Map OpenAI modalities -> Gemini CLI request.generationConfig.responseModalities
 	// e.g. "modalities": ["image", "text"] -> ["IMAGE", "TEXT"]
 	if mods := gjson.GetBytes(rawJSON, "modalities"); mods.Exists() && mods.IsArray() {

internal/translator/codex/claude/codex_claude_response.go

@@ -117,8 +117,12 @@ func ConvertCodexResponseToClaude(_ context.Context, _ string, originalRequestRa
 		} else {
 			template, _ = sjson.Set(template, "delta.stop_reason", "end_turn")
 		}
-		template, _ = sjson.Set(template, "usage.input_tokens", rootResult.Get("response.usage.input_tokens").Int())
-		template, _ = sjson.Set(template, "usage.output_tokens", rootResult.Get("response.usage.output_tokens").Int())
+		inputTokens, outputTokens, cachedTokens := extractResponsesUsage(rootResult.Get("response.usage"))
+		template, _ = sjson.Set(template, "usage.input_tokens", inputTokens)
+		template, _ = sjson.Set(template, "usage.output_tokens", outputTokens)
+		if cachedTokens > 0 {
+			template, _ = sjson.Set(template, "usage.cache_read_input_tokens", cachedTokens)
+		}
 
 		output = "event: message_delta\n"
 		output += fmt.Sprintf("data: %s\n\n", template)
@@ -204,8 +208,12 @@ func ConvertCodexResponseToClaudeNonStream(_ context.Context, _ string, original
 	out := `{"id":"","type":"message","role":"assistant","model":"","content":[],"stop_reason":null,"stop_sequence":null,"usage":{"input_tokens":0,"output_tokens":0}}`
 	out, _ = sjson.Set(out, "id", responseData.Get("id").String())
 	out, _ = sjson.Set(out, "model", responseData.Get("model").String())
-	out, _ = sjson.Set(out, "usage.input_tokens", responseData.Get("usage.input_tokens").Int())
-	out, _ = sjson.Set(out, "usage.output_tokens", responseData.Get("usage.output_tokens").Int())
+	inputTokens, outputTokens, cachedTokens := extractResponsesUsage(responseData.Get("usage"))
+	out, _ = sjson.Set(out, "usage.input_tokens", inputTokens)
+	out, _ = sjson.Set(out, "usage.output_tokens", outputTokens)
+	if cachedTokens > 0 {
+		out, _ = sjson.Set(out, "usage.cache_read_input_tokens", cachedTokens)
+	}
 
 	hasToolCall := false
 
@@ -308,12 +316,27 @@ func ConvertCodexResponseToClaudeNonStream(_ context.Context, _ string, original
 		out, _ = sjson.SetRaw(out, "stop_sequence", stopSequence.Raw)
 	}
 
-	if responseData.Get("usage.input_tokens").Exists() || responseData.Get("usage.output_tokens").Exists() {
-		out, _ = sjson.Set(out, "usage.input_tokens", responseData.Get("usage.input_tokens").Int())
-		out, _ = sjson.Set(out, "usage.output_tokens", responseData.Get("usage.output_tokens").Int())
+	return out
+}
+
+func extractResponsesUsage(usage gjson.Result) (int64, int64, int64) {
+	if !usage.Exists() || usage.Type == gjson.Null {
+		return 0, 0, 0
 	}
 
-	return out
+	inputTokens := usage.Get("input_tokens").Int()
+	outputTokens := usage.Get("output_tokens").Int()
+	cachedTokens := usage.Get("input_tokens_details.cached_tokens").Int()
+
+	if cachedTokens > 0 {
+		if inputTokens >= cachedTokens {
+			inputTokens -= cachedTokens
+		} else {
+			inputTokens = 0
+		}
+	}
+
+	return inputTokens, outputTokens, cachedTokens
 }
 
 // buildReverseMapFromClaudeOriginalShortToOriginal builds a map[short]original from original Claude request tools.

internal/translator/gemini-cli/openai/chat-completions/gemini-cli_openai_request.go

@@ -63,6 +63,13 @@ func ConvertOpenAIRequestToGeminiCLI(modelName string, inputRawJSON []byte, _ bo
 		out, _ = sjson.SetBytes(out, "request.generationConfig.topK", tkr.Num)
 	}
 
+	// Candidate count (OpenAI 'n' parameter)
+	if n := gjson.GetBytes(rawJSON, "n"); n.Exists() && n.Type == gjson.Number {
+		if val := n.Int(); val > 1 {
+			out, _ = sjson.SetBytes(out, "request.generationConfig.candidateCount", val)
+		}
+	}
+
 	// Map OpenAI modalities -> Gemini CLI request.generationConfig.responseModalities
 	// e.g. "modalities": ["image", "text"] -> ["IMAGE", "TEXT"]
 	if mods := gjson.GetBytes(rawJSON, "modalities"); mods.Exists() && mods.IsArray() {

internal/translator/gemini/openai/chat-completions/gemini_openai_request.go

@@ -63,6 +63,13 @@ func ConvertOpenAIRequestToGemini(modelName string, inputRawJSON []byte, _ bool)
 		out, _ = sjson.SetBytes(out, "generationConfig.topK", tkr.Num)
 	}
 
+	// Candidate count (OpenAI 'n' parameter)
+	if n := gjson.GetBytes(rawJSON, "n"); n.Exists() && n.Type == gjson.Number {
+		if val := n.Int(); val > 1 {
+			out, _ = sjson.SetBytes(out, "generationConfig.candidateCount", val)
+		}
+	}
+
 	// Map OpenAI modalities -> Gemini generationConfig.responseModalities
 	// e.g. "modalities": ["image", "text"] -> ["IMAGE", "TEXT"]
 	if mods := gjson.GetBytes(rawJSON, "modalities"); mods.Exists() && mods.IsArray() {

internal/translator/gemini/openai/chat-completions/gemini_openai_response.go

@@ -21,7 +21,8 @@ import (
 // convertGeminiResponseToOpenAIChatParams holds parameters for response conversion.
 type convertGeminiResponseToOpenAIChatParams struct {
 	UnixTimestamp int64
-	FunctionIndex int
+	// FunctionIndex tracks tool call indices per candidate index to support multiple candidates.
+	FunctionIndex map[int]int
 }
 
 // functionCallIDCounter provides a process-wide unique counter for function call identifiers.
@@ -42,13 +43,20 @@ var functionCallIDCounter uint64
 // Returns:
 //   - []string: A slice of strings, each containing an OpenAI-compatible JSON response
 func ConvertGeminiResponseToOpenAI(_ context.Context, _ string, originalRequestRawJSON, requestRawJSON, rawJSON []byte, param *any) []string {
+	// Initialize parameters if nil.
 	if *param == nil {
 		*param = &convertGeminiResponseToOpenAIChatParams{
 			UnixTimestamp: 0,
-			FunctionIndex: 0,
+			FunctionIndex: make(map[int]int),
 		}
 	}
 
+	// Ensure the Map is initialized (handling cases where param might be reused from older context).
+	p := (*param).(*convertGeminiResponseToOpenAIChatParams)
+	if p.FunctionIndex == nil {
+		p.FunctionIndex = make(map[int]int)
+	}
+
 	if bytes.HasPrefix(rawJSON, []byte("data:")) {
 		rawJSON = bytes.TrimSpace(rawJSON[5:])
 	}
@@ -57,151 +65,179 @@ func ConvertGeminiResponseToOpenAI(_ context.Context, _ string, originalRequestR
 		return []string{}
 	}
 
-	// Initialize the OpenAI SSE template.
-	template := `{"id":"","object":"chat.completion.chunk","created":12345,"model":"model","choices":[{"index":0,"delta":{"role":null,"content":null,"reasoning_content":null,"tool_calls":null},"finish_reason":null,"native_finish_reason":null}]}`
+	// Initialize the OpenAI SSE base template.
+	// We use a base template and clone it for each candidate to support multiple candidates.
+	baseTemplate := `{"id":"","object":"chat.completion.chunk","created":12345,"model":"model","choices":[{"index":0,"delta":{"role":null,"content":null,"reasoning_content":null,"tool_calls":null},"finish_reason":null,"native_finish_reason":null}]}`
 
 	// Extract and set the model version.
 	if modelVersionResult := gjson.GetBytes(rawJSON, "modelVersion"); modelVersionResult.Exists() {
-		template, _ = sjson.Set(template, "model", modelVersionResult.String())
+		baseTemplate, _ = sjson.Set(baseTemplate, "model", modelVersionResult.String())
 	}
 
 	// Extract and set the creation timestamp.
 	if createTimeResult := gjson.GetBytes(rawJSON, "createTime"); createTimeResult.Exists() {
 		t, err := time.Parse(time.RFC3339Nano, createTimeResult.String())
 		if err == nil {
-			(*param).(*convertGeminiResponseToOpenAIChatParams).UnixTimestamp = t.Unix()
+			p.UnixTimestamp = t.Unix()
 		}
-		template, _ = sjson.Set(template, "created", (*param).(*convertGeminiResponseToOpenAIChatParams).UnixTimestamp)
+		baseTemplate, _ = sjson.Set(baseTemplate, "created", p.UnixTimestamp)
 	} else {
-		template, _ = sjson.Set(template, "created", (*param).(*convertGeminiResponseToOpenAIChatParams).UnixTimestamp)
+		baseTemplate, _ = sjson.Set(baseTemplate, "created", p.UnixTimestamp)
 	}
 
 	// Extract and set the response ID.
 	if responseIDResult := gjson.GetBytes(rawJSON, "responseId"); responseIDResult.Exists() {
-		template, _ = sjson.Set(template, "id", responseIDResult.String())
-	}
-
-	// Extract and set the finish reason.
-	if finishReasonResult := gjson.GetBytes(rawJSON, "candidates.0.finishReason"); finishReasonResult.Exists() {
-		template, _ = sjson.Set(template, "choices.0.finish_reason", strings.ToLower(finishReasonResult.String()))
-		template, _ = sjson.Set(template, "choices.0.native_finish_reason", strings.ToLower(finishReasonResult.String()))
+		baseTemplate, _ = sjson.Set(baseTemplate, "id", responseIDResult.String())
 	}
 
 	// Extract and set usage metadata (token counts).
+	// Usage is applied to the base template so it appears in the chunks.
 	if usageResult := gjson.GetBytes(rawJSON, "usageMetadata"); usageResult.Exists() {
 		cachedTokenCount := usageResult.Get("cachedContentTokenCount").Int()
 		if candidatesTokenCountResult := usageResult.Get("candidatesTokenCount"); candidatesTokenCountResult.Exists() {
-			template, _ = sjson.Set(template, "usage.completion_tokens", candidatesTokenCountResult.Int())
+			baseTemplate, _ = sjson.Set(baseTemplate, "usage.completion_tokens", candidatesTokenCountResult.Int())
 		}
 		if totalTokenCountResult := usageResult.Get("totalTokenCount"); totalTokenCountResult.Exists() {
-			template, _ = sjson.Set(template, "usage.total_tokens", totalTokenCountResult.Int())
+			baseTemplate, _ = sjson.Set(baseTemplate, "usage.total_tokens", totalTokenCountResult.Int())
 		}
 		promptTokenCount := usageResult.Get("promptTokenCount").Int() - cachedTokenCount
 		thoughtsTokenCount := usageResult.Get("thoughtsTokenCount").Int()
-		template, _ = sjson.Set(template, "usage.prompt_tokens", promptTokenCount+thoughtsTokenCount)
+		baseTemplate, _ = sjson.Set(baseTemplate, "usage.prompt_tokens", promptTokenCount+thoughtsTokenCount)
 		if thoughtsTokenCount > 0 {
-			template, _ = sjson.Set(template, "usage.completion_tokens_details.reasoning_tokens", thoughtsTokenCount)
+			baseTemplate, _ = sjson.Set(baseTemplate, "usage.completion_tokens_details.reasoning_tokens", thoughtsTokenCount)
 		}
 		// Include cached token count if present (indicates prompt caching is working)
 		if cachedTokenCount > 0 {
 			var err error
-			template, err = sjson.Set(template, "usage.prompt_tokens_details.cached_tokens", cachedTokenCount)
+			baseTemplate, err = sjson.Set(baseTemplate, "usage.prompt_tokens_details.cached_tokens", cachedTokenCount)
 			if err != nil {
 				log.Warnf("gemini openai response: failed to set cached_tokens in streaming: %v", err)
 			}
 		}
 	}
 
-	// Process the main content part of the response.
-	partsResult := gjson.GetBytes(rawJSON, "candidates.0.content.parts")
-	hasFunctionCall := false
-	if partsResult.IsArray() {
-		partResults := partsResult.Array()
-		for i := 0; i < len(partResults); i++ {
-			partResult := partResults[i]
-			partTextResult := partResult.Get("text")
-			functionCallResult := partResult.Get("functionCall")
-			inlineDataResult := partResult.Get("inlineData")
-			if !inlineDataResult.Exists() {
-				inlineDataResult = partResult.Get("inline_data")
-			}
-			thoughtSignatureResult := partResult.Get("thoughtSignature")
-			if !thoughtSignatureResult.Exists() {
-				thoughtSignatureResult = partResult.Get("thought_signature")
+	var responseStrings []string
+	candidates := gjson.GetBytes(rawJSON, "candidates")
+
+	// Iterate over all candidates to support candidate_count > 1.
+	if candidates.IsArray() {
+		candidates.ForEach(func(_, candidate gjson.Result) bool {
+			// Clone the template for the current candidate.
+			template := baseTemplate
+
+			// Set the specific index for this candidate.
+			candidateIndex := int(candidate.Get("index").Int())
+			template, _ = sjson.Set(template, "choices.0.index", candidateIndex)
+
+			// Extract and set the finish reason.
+			if finishReasonResult := candidate.Get("finishReason"); finishReasonResult.Exists() {
+				template, _ = sjson.Set(template, "choices.0.finish_reason", strings.ToLower(finishReasonResult.String()))
+				template, _ = sjson.Set(template, "choices.0.native_finish_reason", strings.ToLower(finishReasonResult.String()))
 			}
 
-			hasThoughtSignature := thoughtSignatureResult.Exists() && thoughtSignatureResult.String() != ""
-			hasContentPayload := partTextResult.Exists() || functionCallResult.Exists() || inlineDataResult.Exists()
+			partsResult := candidate.Get("content.parts")
+			hasFunctionCall := false
 
-			// Skip pure thoughtSignature parts but keep any actual payload in the same part.
-			if hasThoughtSignature && !hasContentPayload {
-				continue
+			if partsResult.IsArray() {
+				partResults := partsResult.Array()
+				for i := 0; i < len(partResults); i++ {
+					partResult := partResults[i]
+					partTextResult := partResult.Get("text")
+					functionCallResult := partResult.Get("functionCall")
+					inlineDataResult := partResult.Get("inlineData")
+					if !inlineDataResult.Exists() {
+						inlineDataResult = partResult.Get("inline_data")
+					}
+					thoughtSignatureResult := partResult.Get("thoughtSignature")
+					if !thoughtSignatureResult.Exists() {
+						thoughtSignatureResult = partResult.Get("thought_signature")
+					}
+
+					hasThoughtSignature := thoughtSignatureResult.Exists() && thoughtSignatureResult.String() != ""
+					hasContentPayload := partTextResult.Exists() || functionCallResult.Exists() || inlineDataResult.Exists()
+
+					// Skip pure thoughtSignature parts but keep any actual payload in the same part.
+					if hasThoughtSignature && !hasContentPayload {
+						continue
+					}
+
+					if partTextResult.Exists() {
+						text := partTextResult.String()
+						// Handle text content, distinguishing between regular content and reasoning/thoughts.
+						if partResult.Get("thought").Bool() {
+							template, _ = sjson.Set(template, "choices.0.delta.reasoning_content", text)
+						} else {
+							template, _ = sjson.Set(template, "choices.0.delta.content", text)
+						}
+						template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
+					} else if functionCallResult.Exists() {
+						// Handle function call content.
+						hasFunctionCall = true
+						toolCallsResult := gjson.Get(template, "choices.0.delta.tool_calls")
+
+						// Retrieve the function index for this specific candidate.
+						functionCallIndex := p.FunctionIndex[candidateIndex]
+						p.FunctionIndex[candidateIndex]++
+
+						if toolCallsResult.Exists() && toolCallsResult.IsArray() {
+							functionCallIndex = len(toolCallsResult.Array())
+						} else {
+							template, _ = sjson.SetRaw(template, "choices.0.delta.tool_calls", `[]`)
+						}
+
+						functionCallTemplate := `{"id": "","index": 0,"type": "function","function": {"name": "","arguments": ""}}`
+						fcName := functionCallResult.Get("name").String()
+						functionCallTemplate, _ = sjson.Set(functionCallTemplate, "id", fmt.Sprintf("%s-%d-%d", fcName, time.Now().UnixNano(), atomic.AddUint64(&functionCallIDCounter, 1)))
+						functionCallTemplate, _ = sjson.Set(functionCallTemplate, "index", functionCallIndex)
+						functionCallTemplate, _ = sjson.Set(functionCallTemplate, "function.name", fcName)
+						if fcArgsResult := functionCallResult.Get("args"); fcArgsResult.Exists() {
+							functionCallTemplate, _ = sjson.Set(functionCallTemplate, "function.arguments", fcArgsResult.Raw)
+						}
+						template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
+						template, _ = sjson.SetRaw(template, "choices.0.delta.tool_calls.-1", functionCallTemplate)
+					} else if inlineDataResult.Exists() {
+						data := inlineDataResult.Get("data").String()
+						if data == "" {
+							continue
+						}
+						mimeType := inlineDataResult.Get("mimeType").String()
+						if mimeType == "" {
+							mimeType = inlineDataResult.Get("mime_type").String()
+						}
+						if mimeType == "" {
+							mimeType = "image/png"
+						}
+						imageURL := fmt.Sprintf("data:%s;base64,%s", mimeType, data)
+						imagesResult := gjson.Get(template, "choices.0.delta.images")
+						if !imagesResult.Exists() || !imagesResult.IsArray() {
+							template, _ = sjson.SetRaw(template, "choices.0.delta.images", `[]`)
+						}
+						imageIndex := len(gjson.Get(template, "choices.0.delta.images").Array())
+						imagePayload := `{"type":"image_url","image_url":{"url":""}}`
+						imagePayload, _ = sjson.Set(imagePayload, "index", imageIndex)
+						imagePayload, _ = sjson.Set(imagePayload, "image_url.url", imageURL)
+						template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
+						template, _ = sjson.SetRaw(template, "choices.0.delta.images.-1", imagePayload)
+					}
+				}
 			}
 
-			if partTextResult.Exists() {
-				text := partTextResult.String()
-				// Handle text content, distinguishing between regular content and reasoning/thoughts.
-				if partResult.Get("thought").Bool() {
-					template, _ = sjson.Set(template, "choices.0.delta.reasoning_content", text)
-				} else {
-					template, _ = sjson.Set(template, "choices.0.delta.content", text)
-				}
-				template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
-			} else if functionCallResult.Exists() {
-				// Handle function call content.
-				hasFunctionCall = true
-				toolCallsResult := gjson.Get(template, "choices.0.delta.tool_calls")
-				functionCallIndex := (*param).(*convertGeminiResponseToOpenAIChatParams).FunctionIndex
-				(*param).(*convertGeminiResponseToOpenAIChatParams).FunctionIndex++
-				if toolCallsResult.Exists() && toolCallsResult.IsArray() {
-					functionCallIndex = len(toolCallsResult.Array())
-				} else {
-					template, _ = sjson.SetRaw(template, "choices.0.delta.tool_calls", `[]`)
-				}
-
-				functionCallTemplate := `{"id": "","index": 0,"type": "function","function": {"name": "","arguments": ""}}`
-				fcName := functionCallResult.Get("name").String()
-				functionCallTemplate, _ = sjson.Set(functionCallTemplate, "id", fmt.Sprintf("%s-%d-%d", fcName, time.Now().UnixNano(), atomic.AddUint64(&functionCallIDCounter, 1)))
-				functionCallTemplate, _ = sjson.Set(functionCallTemplate, "index", functionCallIndex)
-				functionCallTemplate, _ = sjson.Set(functionCallTemplate, "function.name", fcName)
-				if fcArgsResult := functionCallResult.Get("args"); fcArgsResult.Exists() {
-					functionCallTemplate, _ = sjson.Set(functionCallTemplate, "function.arguments", fcArgsResult.Raw)
-				}
-				template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
-				template, _ = sjson.SetRaw(template, "choices.0.delta.tool_calls.-1", functionCallTemplate)
-			} else if inlineDataResult.Exists() {
-				data := inlineDataResult.Get("data").String()
-				if data == "" {
-					continue
-				}
-				mimeType := inlineDataResult.Get("mimeType").String()
-				if mimeType == "" {
-					mimeType = inlineDataResult.Get("mime_type").String()
-				}
-				if mimeType == "" {
-					mimeType = "image/png"
-				}
-				imageURL := fmt.Sprintf("data:%s;base64,%s", mimeType, data)
-				imagesResult := gjson.Get(template, "choices.0.delta.images")
-				if !imagesResult.Exists() || !imagesResult.IsArray() {
-					template, _ = sjson.SetRaw(template, "choices.0.delta.images", `[]`)
-				}
-				imageIndex := len(gjson.Get(template, "choices.0.delta.images").Array())
-				imagePayload := `{"type":"image_url","image_url":{"url":""}}`
-				imagePayload, _ = sjson.Set(imagePayload, "index", imageIndex)
-				imagePayload, _ = sjson.Set(imagePayload, "image_url.url", imageURL)
-				template, _ = sjson.Set(template, "choices.0.delta.role", "assistant")
-				template, _ = sjson.SetRaw(template, "choices.0.delta.images.-1", imagePayload)
+			if hasFunctionCall {
+				template, _ = sjson.Set(template, "choices.0.finish_reason", "tool_calls")
+				template, _ = sjson.Set(template, "choices.0.native_finish_reason", "tool_calls")
 			}
+
+			responseStrings = append(responseStrings, template)
+			return true // continue loop
+		})
+	} else {
+		// If there are no candidates (e.g., a pure usageMetadata chunk), return the usage chunk if present.
+		if gjson.GetBytes(rawJSON, "usageMetadata").Exists() && len(responseStrings) == 0 {
+			responseStrings = append(responseStrings, baseTemplate)
 		}
 	}
 
-	if hasFunctionCall {
-		template, _ = sjson.Set(template, "choices.0.finish_reason", "tool_calls")
-		template, _ = sjson.Set(template, "choices.0.native_finish_reason", "tool_calls")
-	}
-
-	return []string{template}
+	return responseStrings
 }
 
 // ConvertGeminiResponseToOpenAINonStream converts a non-streaming Gemini response to a non-streaming OpenAI response.
@@ -219,7 +255,9 @@ func ConvertGeminiResponseToOpenAI(_ context.Context, _ string, originalRequestR
 //   - string: An OpenAI-compatible JSON response containing all message content and metadata
 func ConvertGeminiResponseToOpenAINonStream(_ context.Context, _ string, originalRequestRawJSON, requestRawJSON, rawJSON []byte, _ *any) string {
 	var unixTimestamp int64
-	template := `{"id":"","object":"chat.completion","created":123456,"model":"model","choices":[{"index":0,"message":{"role":"assistant","content":null,"reasoning_content":null,"tool_calls":null},"finish_reason":null,"native_finish_reason":null}]}`
+	// Initialize template with an empty choices array to support multiple candidates.
+	template := `{"id":"","object":"chat.completion","created":123456,"model":"model","choices":[]}`
+
 	if modelVersionResult := gjson.GetBytes(rawJSON, "modelVersion"); modelVersionResult.Exists() {
 		template, _ = sjson.Set(template, "model", modelVersionResult.String())
 	}
@@ -238,11 +276,6 @@ func ConvertGeminiResponseToOpenAINonStream(_ context.Context, _ string, origina
 		template, _ = sjson.Set(template, "id", responseIDResult.String())
 	}
 
-	if finishReasonResult := gjson.GetBytes(rawJSON, "candidates.0.finishReason"); finishReasonResult.Exists() {
-		template, _ = sjson.Set(template, "choices.0.finish_reason", strings.ToLower(finishReasonResult.String()))
-		template, _ = sjson.Set(template, "choices.0.native_finish_reason", strings.ToLower(finishReasonResult.String()))
-	}
-
 	if usageResult := gjson.GetBytes(rawJSON, "usageMetadata"); usageResult.Exists() {
 		if candidatesTokenCountResult := usageResult.Get("candidatesTokenCount"); candidatesTokenCountResult.Exists() {
 			template, _ = sjson.Set(template, "usage.completion_tokens", candidatesTokenCountResult.Int())
@@ -267,74 +300,96 @@ func ConvertGeminiResponseToOpenAINonStream(_ context.Context, _ string, origina
 		}
 	}
 
-	// Process the main content part of the response.
-	partsResult := gjson.GetBytes(rawJSON, "candidates.0.content.parts")
-	hasFunctionCall := false
-	if partsResult.IsArray() {
-		partsResults := partsResult.Array()
-		for i := 0; i < len(partsResults); i++ {
-			partResult := partsResults[i]
-			partTextResult := partResult.Get("text")
-			functionCallResult := partResult.Get("functionCall")
-			inlineDataResult := partResult.Get("inlineData")
-			if !inlineDataResult.Exists() {
-				inlineDataResult = partResult.Get("inline_data")
+	// Process the main content part of the response for all candidates.
+	candidates := gjson.GetBytes(rawJSON, "candidates")
+	if candidates.IsArray() {
+		candidates.ForEach(func(_, candidate gjson.Result) bool {
+			// Construct a single Choice object.
+			choiceTemplate := `{"index":0,"message":{"role":"assistant","content":null,"reasoning_content":null,"tool_calls":null},"finish_reason":null,"native_finish_reason":null}`
+
+			// Set the index for this choice.
+			choiceTemplate, _ = sjson.Set(choiceTemplate, "index", candidate.Get("index").Int())
+
+			// Set finish reason.
+			if finishReasonResult := candidate.Get("finishReason"); finishReasonResult.Exists() {
+				choiceTemplate, _ = sjson.Set(choiceTemplate, "finish_reason", strings.ToLower(finishReasonResult.String()))
+				choiceTemplate, _ = sjson.Set(choiceTemplate, "native_finish_reason", strings.ToLower(finishReasonResult.String()))
 			}
 
-			if partTextResult.Exists() {
-				// Append text content, distinguishing between regular content and reasoning.
-				if partResult.Get("thought").Bool() {
-					template, _ = sjson.Set(template, "choices.0.message.reasoning_content", partTextResult.String())
-				} else {
-					template, _ = sjson.Set(template, "choices.0.message.content", partTextResult.String())
-				}
-				template, _ = sjson.Set(template, "choices.0.message.role", "assistant")
-			} else if functionCallResult.Exists() {
-				// Append function call content to the tool_calls array.
-				hasFunctionCall = true
-				toolCallsResult := gjson.Get(template, "choices.0.message.tool_calls")
-				if !toolCallsResult.Exists() || !toolCallsResult.IsArray() {
-					template, _ = sjson.SetRaw(template, "choices.0.message.tool_calls", `[]`)
-				}
-				functionCallItemTemplate := `{"id": "","type": "function","function": {"name": "","arguments": ""}}`
-				fcName := functionCallResult.Get("name").String()
-				functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "id", fmt.Sprintf("%s-%d-%d", fcName, time.Now().UnixNano(), atomic.AddUint64(&functionCallIDCounter, 1)))
-				functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "function.name", fcName)
-				if fcArgsResult := functionCallResult.Get("args"); fcArgsResult.Exists() {
-					functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "function.arguments", fcArgsResult.Raw)
-				}
-				template, _ = sjson.Set(template, "choices.0.message.role", "assistant")
-				template, _ = sjson.SetRaw(template, "choices.0.message.tool_calls.-1", functionCallItemTemplate)
-			} else if inlineDataResult.Exists() {
-				data := inlineDataResult.Get("data").String()
-				if data == "" {
-					continue
-				}
-				mimeType := inlineDataResult.Get("mimeType").String()
-				if mimeType == "" {
-					mimeType = inlineDataResult.Get("mime_type").String()
-				}
-				if mimeType == "" {
-					mimeType = "image/png"
-				}
-				imageURL := fmt.Sprintf("data:%s;base64,%s", mimeType, data)
-				imagesResult := gjson.Get(template, "choices.0.message.images")
-				if !imagesResult.Exists() || !imagesResult.IsArray() {
-					template, _ = sjson.SetRaw(template, "choices.0.message.images", `[]`)
-				}
-				imageIndex := len(gjson.Get(template, "choices.0.message.images").Array())
-				imagePayload := `{"type":"image_url","image_url":{"url":""}}`
-				imagePayload, _ = sjson.Set(imagePayload, "index", imageIndex)
-				imagePayload, _ = sjson.Set(imagePayload, "image_url.url", imageURL)
-				template, _ = sjson.Set(template, "choices.0.message.role", "assistant")
-				template, _ = sjson.SetRaw(template, "choices.0.message.images.-1", imagePayload)
-			}
-		}
-	}
+			partsResult := candidate.Get("content.parts")
+			hasFunctionCall := false
+			if partsResult.IsArray() {
+				partsResults := partsResult.Array()
+				for i := 0; i < len(partsResults); i++ {
+					partResult := partsResults[i]
+					partTextResult := partResult.Get("text")
+					functionCallResult := partResult.Get("functionCall")
+					inlineDataResult := partResult.Get("inlineData")
+					if !inlineDataResult.Exists() {
+						inlineDataResult = partResult.Get("inline_data")
+					}
 
-	if hasFunctionCall {
-		template, _ = sjson.Set(template, "choices.0.finish_reason", "tool_calls")
-		template, _ = sjson.Set(template, "choices.0.native_finish_reason", "tool_calls")
+					if partTextResult.Exists() {
+						// Append text content, distinguishing between regular content and reasoning.
+						if partResult.Get("thought").Bool() {
+							oldVal := gjson.Get(choiceTemplate, "message.reasoning_content").String()
+							choiceTemplate, _ = sjson.Set(choiceTemplate, "message.reasoning_content", oldVal+partTextResult.String())
+						} else {
+							oldVal := gjson.Get(choiceTemplate, "message.content").String()
+							choiceTemplate, _ = sjson.Set(choiceTemplate, "message.content", oldVal+partTextResult.String())
+						}
+						choiceTemplate, _ = sjson.Set(choiceTemplate, "message.role", "assistant")
+					} else if functionCallResult.Exists() {
+						// Append function call content to the tool_calls array.
+						hasFunctionCall = true
+						toolCallsResult := gjson.Get(choiceTemplate, "message.tool_calls")
+						if !toolCallsResult.Exists() || !toolCallsResult.IsArray() {
+							choiceTemplate, _ = sjson.SetRaw(choiceTemplate, "message.tool_calls", `[]`)
+						}
+						functionCallItemTemplate := `{"id": "","type": "function","function": {"name": "","arguments": ""}}`
+						fcName := functionCallResult.Get("name").String()
+						functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "id", fmt.Sprintf("%s-%d-%d", fcName, time.Now().UnixNano(), atomic.AddUint64(&functionCallIDCounter, 1)))
+						functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "function.name", fcName)
+						if fcArgsResult := functionCallResult.Get("args"); fcArgsResult.Exists() {
+							functionCallItemTemplate, _ = sjson.Set(functionCallItemTemplate, "function.arguments", fcArgsResult.Raw)
+						}
+						choiceTemplate, _ = sjson.Set(choiceTemplate, "message.role", "assistant")
+						choiceTemplate, _ = sjson.SetRaw(choiceTemplate, "message.tool_calls.-1", functionCallItemTemplate)
+					} else if inlineDataResult.Exists() {
+						data := inlineDataResult.Get("data").String()
+						if data != "" {
+							mimeType := inlineDataResult.Get("mimeType").String()
+							if mimeType == "" {
+								mimeType = inlineDataResult.Get("mime_type").String()
+							}
+							if mimeType == "" {
+								mimeType = "image/png"
+							}
+							imageURL := fmt.Sprintf("data:%s;base64,%s", mimeType, data)
+							imagesResult := gjson.Get(choiceTemplate, "message.images")
+							if !imagesResult.Exists() || !imagesResult.IsArray() {
+								choiceTemplate, _ = sjson.SetRaw(choiceTemplate, "message.images", `[]`)
+							}
+							imageIndex := len(gjson.Get(choiceTemplate, "message.images").Array())
+							imagePayload := `{"type":"image_url","image_url":{"url":""}}`
+							imagePayload, _ = sjson.Set(imagePayload, "index", imageIndex)
+							imagePayload, _ = sjson.Set(imagePayload, "image_url.url", imageURL)
+							choiceTemplate, _ = sjson.Set(choiceTemplate, "message.role", "assistant")
+							choiceTemplate, _ = sjson.SetRaw(choiceTemplate, "message.images.-1", imagePayload)
+						}
+					}
+				}
+			}
+
+			if hasFunctionCall {
+				choiceTemplate, _ = sjson.Set(choiceTemplate, "finish_reason", "tool_calls")
+				choiceTemplate, _ = sjson.Set(choiceTemplate, "native_finish_reason", "tool_calls")
+			}
+
+			// Append the constructed choice to the main choices array.
+			template, _ = sjson.SetRaw(template, "choices.-1", choiceTemplate)
+			return true
+		})
 	}
 
 	return template

internal/translator/gemini/openai/responses/gemini_openai-responses_request.go

@@ -298,6 +298,15 @@ func ConvertOpenAIResponsesRequestToGemini(modelName string, inputRawJSON []byte
 				}
 				functionContent, _ = sjson.SetRaw(functionContent, "parts.-1", functionResponse)
 				out, _ = sjson.SetRaw(out, "contents.-1", functionContent)
+
+			case "reasoning":
+				thoughtContent := `{"role":"model","parts":[]}`
+				thought := `{"text":"","thoughtSignature":"","thought":true}`
+				thought, _ = sjson.Set(thought, "text", item.Get("summary.0.text").String())
+				thought, _ = sjson.Set(thought, "thoughtSignature", item.Get("encrypted_content").String())
+
+				thoughtContent, _ = sjson.SetRaw(thoughtContent, "parts.-1", thought)
+				out, _ = sjson.SetRaw(out, "contents.-1", thoughtContent)
 			}
 		}
 	} else if input.Exists() && input.Type == gjson.String {

internal/translator/gemini/openai/responses/gemini_openai-responses_response.go

@@ -20,6 +20,7 @@ type geminiToResponsesState struct {
 
 	// message aggregation
 	MsgOpened    bool
+	MsgClosed    bool
 	MsgIndex     int
 	CurrentMsgID string
 	TextBuf      strings.Builder
@@ -29,6 +30,7 @@ type geminiToResponsesState struct {
 	ReasoningOpened bool
 	ReasoningIndex  int
 	ReasoningItemID string
+	ReasoningEnc    string
 	ReasoningBuf    strings.Builder
 	ReasoningClosed bool
 
@@ -37,6 +39,7 @@ type geminiToResponsesState struct {
 	FuncArgsBuf map[int]*strings.Builder
 	FuncNames   map[int]string
 	FuncCallIDs map[int]string
+	FuncDone    map[int]bool
 }
 
 // responseIDCounter provides a process-wide unique counter for synthesized response identifiers.
@@ -45,6 +48,39 @@ var responseIDCounter uint64
 // funcCallIDCounter provides a process-wide unique counter for function call identifiers.
 var funcCallIDCounter uint64
 
+func pickRequestJSON(originalRequestRawJSON, requestRawJSON []byte) []byte {
+	if len(originalRequestRawJSON) > 0 && gjson.ValidBytes(originalRequestRawJSON) {
+		return originalRequestRawJSON
+	}
+	if len(requestRawJSON) > 0 && gjson.ValidBytes(requestRawJSON) {
+		return requestRawJSON
+	}
+	return nil
+}
+
+func unwrapRequestRoot(root gjson.Result) gjson.Result {
+	req := root.Get("request")
+	if !req.Exists() {
+		return root
+	}
+	if req.Get("model").Exists() || req.Get("input").Exists() || req.Get("instructions").Exists() {
+		return req
+	}
+	return root
+}
+
+func unwrapGeminiResponseRoot(root gjson.Result) gjson.Result {
+	resp := root.Get("response")
+	if !resp.Exists() {
+		return root
+	}
+	// Vertex-style Gemini responses wrap the actual payload in a "response" object.
+	if resp.Get("candidates").Exists() || resp.Get("responseId").Exists() || resp.Get("usageMetadata").Exists() {
+		return resp
+	}
+	return root
+}
+
 func emitEvent(event string, payload string) string {
 	return fmt.Sprintf("event: %s\ndata: %s", event, payload)
 }
@@ -56,18 +92,37 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 			FuncArgsBuf: make(map[int]*strings.Builder),
 			FuncNames:   make(map[int]string),
 			FuncCallIDs: make(map[int]string),
+			FuncDone:    make(map[int]bool),
 		}
 	}
 	st := (*param).(*geminiToResponsesState)
+	if st.FuncArgsBuf == nil {
+		st.FuncArgsBuf = make(map[int]*strings.Builder)
+	}
+	if st.FuncNames == nil {
+		st.FuncNames = make(map[int]string)
+	}
+	if st.FuncCallIDs == nil {
+		st.FuncCallIDs = make(map[int]string)
+	}
+	if st.FuncDone == nil {
+		st.FuncDone = make(map[int]bool)
+	}
 
 	if bytes.HasPrefix(rawJSON, []byte("data:")) {
 		rawJSON = bytes.TrimSpace(rawJSON[5:])
 	}
 
+	rawJSON = bytes.TrimSpace(rawJSON)
+	if len(rawJSON) == 0 || bytes.Equal(rawJSON, []byte("[DONE]")) {
+		return []string{}
+	}
+
 	root := gjson.ParseBytes(rawJSON)
 	if !root.Exists() {
 		return []string{}
 	}
+	root = unwrapGeminiResponseRoot(root)
 
 	var out []string
 	nextSeq := func() int { st.Seq++; return st.Seq }
@@ -98,19 +153,54 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 		itemDone, _ = sjson.Set(itemDone, "sequence_number", nextSeq())
 		itemDone, _ = sjson.Set(itemDone, "item.id", st.ReasoningItemID)
 		itemDone, _ = sjson.Set(itemDone, "output_index", st.ReasoningIndex)
+		itemDone, _ = sjson.Set(itemDone, "item.encrypted_content", st.ReasoningEnc)
 		itemDone, _ = sjson.Set(itemDone, "item.summary.0.text", full)
 		out = append(out, emitEvent("response.output_item.done", itemDone))
 
 		st.ReasoningClosed = true
 	}
 
+	// Helper to finalize the assistant message in correct order.
+	// It emits response.output_text.done, response.content_part.done,
+	// and response.output_item.done exactly once.
+	finalizeMessage := func() {
+		if !st.MsgOpened || st.MsgClosed {
+			return
+		}
+		fullText := st.ItemTextBuf.String()
+		done := `{"type":"response.output_text.done","sequence_number":0,"item_id":"","output_index":0,"content_index":0,"text":"","logprobs":[]}`
+		done, _ = sjson.Set(done, "sequence_number", nextSeq())
+		done, _ = sjson.Set(done, "item_id", st.CurrentMsgID)
+		done, _ = sjson.Set(done, "output_index", st.MsgIndex)
+		done, _ = sjson.Set(done, "text", fullText)
+		out = append(out, emitEvent("response.output_text.done", done))
+		partDone := `{"type":"response.content_part.done","sequence_number":0,"item_id":"","output_index":0,"content_index":0,"part":{"type":"output_text","annotations":[],"logprobs":[],"text":""}}`
+		partDone, _ = sjson.Set(partDone, "sequence_number", nextSeq())
+		partDone, _ = sjson.Set(partDone, "item_id", st.CurrentMsgID)
+		partDone, _ = sjson.Set(partDone, "output_index", st.MsgIndex)
+		partDone, _ = sjson.Set(partDone, "part.text", fullText)
+		out = append(out, emitEvent("response.content_part.done", partDone))
+		final := `{"type":"response.output_item.done","sequence_number":0,"output_index":0,"item":{"id":"","type":"message","status":"completed","content":[{"type":"output_text","text":""}],"role":"assistant"}}`
+		final, _ = sjson.Set(final, "sequence_number", nextSeq())
+		final, _ = sjson.Set(final, "output_index", st.MsgIndex)
+		final, _ = sjson.Set(final, "item.id", st.CurrentMsgID)
+		final, _ = sjson.Set(final, "item.content.0.text", fullText)
+		out = append(out, emitEvent("response.output_item.done", final))
+
+		st.MsgClosed = true
+	}
+
 	// Initialize per-response fields and emit created/in_progress once
 	if !st.Started {
-		if v := root.Get("responseId"); v.Exists() {
-			st.ResponseID = v.String()
+		st.ResponseID = root.Get("responseId").String()
+		if st.ResponseID == "" {
+			st.ResponseID = fmt.Sprintf("resp_%x_%d", time.Now().UnixNano(), atomic.AddUint64(&responseIDCounter, 1))
+		}
+		if !strings.HasPrefix(st.ResponseID, "resp_") {
+			st.ResponseID = fmt.Sprintf("resp_%s", st.ResponseID)
 		}
 		if v := root.Get("createTime"); v.Exists() {
-			if t, err := time.Parse(time.RFC3339Nano, v.String()); err == nil {
+			if t, errParseCreateTime := time.Parse(time.RFC3339Nano, v.String()); errParseCreateTime == nil {
 				st.CreatedAt = t.Unix()
 			}
 		}
@@ -143,15 +233,21 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 					// Ignore any late thought chunks after reasoning is finalized.
 					return true
 				}
+				if sig := part.Get("thoughtSignature"); sig.Exists() && sig.String() != "" && sig.String() != geminiResponsesThoughtSignature {
+					st.ReasoningEnc = sig.String()
+				} else if sig = part.Get("thought_signature"); sig.Exists() && sig.String() != "" && sig.String() != geminiResponsesThoughtSignature {
+					st.ReasoningEnc = sig.String()
+				}
 				if !st.ReasoningOpened {
 					st.ReasoningOpened = true
 					st.ReasoningIndex = st.NextIndex
 					st.NextIndex++
 					st.ReasoningItemID = fmt.Sprintf("rs_%s_%d", st.ResponseID, st.ReasoningIndex)
-					item := `{"type":"response.output_item.added","sequence_number":0,"output_index":0,"item":{"id":"","type":"reasoning","status":"in_progress","summary":[]}}`
+					item := `{"type":"response.output_item.added","sequence_number":0,"output_index":0,"item":{"id":"","type":"reasoning","status":"in_progress","encrypted_content":"","summary":[]}}`
 					item, _ = sjson.Set(item, "sequence_number", nextSeq())
 					item, _ = sjson.Set(item, "output_index", st.ReasoningIndex)
 					item, _ = sjson.Set(item, "item.id", st.ReasoningItemID)
+					item, _ = sjson.Set(item, "item.encrypted_content", st.ReasoningEnc)
 					out = append(out, emitEvent("response.output_item.added", item))
 					partAdded := `{"type":"response.reasoning_summary_part.added","sequence_number":0,"item_id":"","output_index":0,"summary_index":0,"part":{"type":"summary_text","text":""}}`
 					partAdded, _ = sjson.Set(partAdded, "sequence_number", nextSeq())
@@ -191,9 +287,9 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 					partAdded, _ = sjson.Set(partAdded, "output_index", st.MsgIndex)
 					out = append(out, emitEvent("response.content_part.added", partAdded))
 					st.ItemTextBuf.Reset()
-					st.ItemTextBuf.WriteString(t.String())
 				}
 				st.TextBuf.WriteString(t.String())
+				st.ItemTextBuf.WriteString(t.String())
 				msg := `{"type":"response.output_text.delta","sequence_number":0,"item_id":"","output_index":0,"content_index":0,"delta":"","logprobs":[]}`
 				msg, _ = sjson.Set(msg, "sequence_number", nextSeq())
 				msg, _ = sjson.Set(msg, "item_id", st.CurrentMsgID)
@@ -205,8 +301,10 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 
 			// Function call
 			if fc := part.Get("functionCall"); fc.Exists() {
-				// Before emitting function-call outputs, finalize reasoning if open.
+				// Before emitting function-call outputs, finalize reasoning and the message (if open).
+				// Responses streaming requires message done events before the next output_item.added.
 				finalizeReasoning()
+				finalizeMessage()
 				name := fc.Get("name").String()
 				idx := st.NextIndex
 				st.NextIndex++
@@ -219,6 +317,14 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 				}
 				st.FuncNames[idx] = name
 
+				argsJSON := "{}"
+				if args := fc.Get("args"); args.Exists() {
+					argsJSON = args.Raw
+				}
+				if st.FuncArgsBuf[idx].Len() == 0 && argsJSON != "" {
+					st.FuncArgsBuf[idx].WriteString(argsJSON)
+				}
+
 				// Emit item.added for function call
 				item := `{"type":"response.output_item.added","sequence_number":0,"output_index":0,"item":{"id":"","type":"function_call","status":"in_progress","arguments":"","call_id":"","name":""}}`
 				item, _ = sjson.Set(item, "sequence_number", nextSeq())
@@ -228,10 +334,9 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 				item, _ = sjson.Set(item, "item.name", name)
 				out = append(out, emitEvent("response.output_item.added", item))
 
-				// Emit arguments delta (full args in one chunk)
-				if args := fc.Get("args"); args.Exists() {
-					argsJSON := args.Raw
-					st.FuncArgsBuf[idx].WriteString(argsJSON)
+				// Emit arguments delta (full args in one chunk).
+				// When Gemini omits args, emit "{}" to keep Responses streaming event order consistent.
+				if argsJSON != "" {
 					ad := `{"type":"response.function_call_arguments.delta","sequence_number":0,"item_id":"","output_index":0,"delta":""}`
 					ad, _ = sjson.Set(ad, "sequence_number", nextSeq())
 					ad, _ = sjson.Set(ad, "item_id", fmt.Sprintf("fc_%s", st.FuncCallIDs[idx]))
@@ -240,6 +345,27 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 					out = append(out, emitEvent("response.function_call_arguments.delta", ad))
 				}
 
+				// Gemini emits the full function call payload at once, so we can finalize it immediately.
+				if !st.FuncDone[idx] {
+					fcDone := `{"type":"response.function_call_arguments.done","sequence_number":0,"item_id":"","output_index":0,"arguments":""}`
+					fcDone, _ = sjson.Set(fcDone, "sequence_number", nextSeq())
+					fcDone, _ = sjson.Set(fcDone, "item_id", fmt.Sprintf("fc_%s", st.FuncCallIDs[idx]))
+					fcDone, _ = sjson.Set(fcDone, "output_index", idx)
+					fcDone, _ = sjson.Set(fcDone, "arguments", argsJSON)
+					out = append(out, emitEvent("response.function_call_arguments.done", fcDone))
+
+					itemDone := `{"type":"response.output_item.done","sequence_number":0,"output_index":0,"item":{"id":"","type":"function_call","status":"completed","arguments":"","call_id":"","name":""}}`
+					itemDone, _ = sjson.Set(itemDone, "sequence_number", nextSeq())
+					itemDone, _ = sjson.Set(itemDone, "output_index", idx)
+					itemDone, _ = sjson.Set(itemDone, "item.id", fmt.Sprintf("fc_%s", st.FuncCallIDs[idx]))
+					itemDone, _ = sjson.Set(itemDone, "item.arguments", argsJSON)
+					itemDone, _ = sjson.Set(itemDone, "item.call_id", st.FuncCallIDs[idx])
+					itemDone, _ = sjson.Set(itemDone, "item.name", st.FuncNames[idx])
+					out = append(out, emitEvent("response.output_item.done", itemDone))
+
+					st.FuncDone[idx] = true
+				}
+
 				return true
 			}
 
@@ -251,28 +377,7 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 	if fr := root.Get("candidates.0.finishReason"); fr.Exists() && fr.String() != "" {
 		// Finalize reasoning first to keep ordering tight with last delta
 		finalizeReasoning()
-		// Close message output if opened
-		if st.MsgOpened {
-			fullText := st.ItemTextBuf.String()
-			done := `{"type":"response.output_text.done","sequence_number":0,"item_id":"","output_index":0,"content_index":0,"text":"","logprobs":[]}`
-			done, _ = sjson.Set(done, "sequence_number", nextSeq())
-			done, _ = sjson.Set(done, "item_id", st.CurrentMsgID)
-			done, _ = sjson.Set(done, "output_index", st.MsgIndex)
-			done, _ = sjson.Set(done, "text", fullText)
-			out = append(out, emitEvent("response.output_text.done", done))
-			partDone := `{"type":"response.content_part.done","sequence_number":0,"item_id":"","output_index":0,"content_index":0,"part":{"type":"output_text","annotations":[],"logprobs":[],"text":""}}`
-			partDone, _ = sjson.Set(partDone, "sequence_number", nextSeq())
-			partDone, _ = sjson.Set(partDone, "item_id", st.CurrentMsgID)
-			partDone, _ = sjson.Set(partDone, "output_index", st.MsgIndex)
-			partDone, _ = sjson.Set(partDone, "part.text", fullText)
-			out = append(out, emitEvent("response.content_part.done", partDone))
-			final := `{"type":"response.output_item.done","sequence_number":0,"output_index":0,"item":{"id":"","type":"message","status":"completed","content":[{"type":"output_text","text":""}],"role":"assistant"}}`
-			final, _ = sjson.Set(final, "sequence_number", nextSeq())
-			final, _ = sjson.Set(final, "output_index", st.MsgIndex)
-			final, _ = sjson.Set(final, "item.id", st.CurrentMsgID)
-			final, _ = sjson.Set(final, "item.content.0.text", fullText)
-			out = append(out, emitEvent("response.output_item.done", final))
-		}
+		finalizeMessage()
 
 		// Close function calls
 		if len(st.FuncArgsBuf) > 0 {
@@ -289,6 +394,9 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 				}
 			}
 			for _, idx := range idxs {
+				if st.FuncDone[idx] {
+					continue
+				}
 				args := "{}"
 				if b := st.FuncArgsBuf[idx]; b != nil && b.Len() > 0 {
 					args = b.String()
@@ -308,6 +416,8 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 				itemDone, _ = sjson.Set(itemDone, "item.call_id", st.FuncCallIDs[idx])
 				itemDone, _ = sjson.Set(itemDone, "item.name", st.FuncNames[idx])
 				out = append(out, emitEvent("response.output_item.done", itemDone))
+
+				st.FuncDone[idx] = true
 			}
 		}
 
@@ -319,8 +429,8 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 		completed, _ = sjson.Set(completed, "response.id", st.ResponseID)
 		completed, _ = sjson.Set(completed, "response.created_at", st.CreatedAt)
 
-		if requestRawJSON != nil {
-			req := gjson.ParseBytes(requestRawJSON)
+		if reqJSON := pickRequestJSON(originalRequestRawJSON, requestRawJSON); len(reqJSON) > 0 {
+			req := unwrapRequestRoot(gjson.ParseBytes(reqJSON))
 			if v := req.Get("instructions"); v.Exists() {
 				completed, _ = sjson.Set(completed, "response.instructions", v.String())
 			}
@@ -383,41 +493,34 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 			}
 		}
 
-		// Compose outputs in encountered order: reasoning, message, function_calls
+		// Compose outputs in output_index order.
 		outputsWrapper := `{"arr":[]}`
-		if st.ReasoningOpened {
-			item := `{"id":"","type":"reasoning","summary":[{"type":"summary_text","text":""}]}`
-			item, _ = sjson.Set(item, "id", st.ReasoningItemID)
-			item, _ = sjson.Set(item, "summary.0.text", st.ReasoningBuf.String())
-			outputsWrapper, _ = sjson.SetRaw(outputsWrapper, "arr.-1", item)
-		}
-		if st.MsgOpened {
-			item := `{"id":"","type":"message","status":"completed","content":[{"type":"output_text","annotations":[],"logprobs":[],"text":""}],"role":"assistant"}`
-			item, _ = sjson.Set(item, "id", st.CurrentMsgID)
-			item, _ = sjson.Set(item, "content.0.text", st.TextBuf.String())
-			outputsWrapper, _ = sjson.SetRaw(outputsWrapper, "arr.-1", item)
-		}
-		if len(st.FuncArgsBuf) > 0 {
-			idxs := make([]int, 0, len(st.FuncArgsBuf))
-			for idx := range st.FuncArgsBuf {
-				idxs = append(idxs, idx)
+		for idx := 0; idx < st.NextIndex; idx++ {
+			if st.ReasoningOpened && idx == st.ReasoningIndex {
+				item := `{"id":"","type":"reasoning","encrypted_content":"","summary":[{"type":"summary_text","text":""}]}`
+				item, _ = sjson.Set(item, "id", st.ReasoningItemID)
+				item, _ = sjson.Set(item, "encrypted_content", st.ReasoningEnc)
+				item, _ = sjson.Set(item, "summary.0.text", st.ReasoningBuf.String())
+				outputsWrapper, _ = sjson.SetRaw(outputsWrapper, "arr.-1", item)
+				continue
 			}
-			for i := 0; i < len(idxs); i++ {
-				for j := i + 1; j < len(idxs); j++ {
-					if idxs[j] < idxs[i] {
-						idxs[i], idxs[j] = idxs[j], idxs[i]
-					}
-				}
+			if st.MsgOpened && idx == st.MsgIndex {
+				item := `{"id":"","type":"message","status":"completed","content":[{"type":"output_text","annotations":[],"logprobs":[],"text":""}],"role":"assistant"}`
+				item, _ = sjson.Set(item, "id", st.CurrentMsgID)
+				item, _ = sjson.Set(item, "content.0.text", st.TextBuf.String())
+				outputsWrapper, _ = sjson.SetRaw(outputsWrapper, "arr.-1", item)
+				continue
 			}
-			for _, idx := range idxs {
-				args := ""
-				if b := st.FuncArgsBuf[idx]; b != nil {
+
+			if callID, ok := st.FuncCallIDs[idx]; ok && callID != "" {
+				args := "{}"
+				if b := st.FuncArgsBuf[idx]; b != nil && b.Len() > 0 {
 					args = b.String()
 				}
 				item := `{"id":"","type":"function_call","status":"completed","arguments":"","call_id":"","name":""}`
-				item, _ = sjson.Set(item, "id", fmt.Sprintf("fc_%s", st.FuncCallIDs[idx]))
+				item, _ = sjson.Set(item, "id", fmt.Sprintf("fc_%s", callID))
 				item, _ = sjson.Set(item, "arguments", args)
-				item, _ = sjson.Set(item, "call_id", st.FuncCallIDs[idx])
+				item, _ = sjson.Set(item, "call_id", callID)
 				item, _ = sjson.Set(item, "name", st.FuncNames[idx])
 				outputsWrapper, _ = sjson.SetRaw(outputsWrapper, "arr.-1", item)
 			}
@@ -431,8 +534,8 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 			// input tokens = prompt + thoughts
 			input := um.Get("promptTokenCount").Int() + um.Get("thoughtsTokenCount").Int()
 			completed, _ = sjson.Set(completed, "response.usage.input_tokens", input)
-			// cached_tokens not provided by Gemini; default to 0 for structure compatibility
-			completed, _ = sjson.Set(completed, "response.usage.input_tokens_details.cached_tokens", 0)
+			// cached token details: align with OpenAI "cached_tokens" semantics.
+			completed, _ = sjson.Set(completed, "response.usage.input_tokens_details.cached_tokens", um.Get("cachedContentTokenCount").Int())
 			// output tokens
 			if v := um.Get("candidatesTokenCount"); v.Exists() {
 				completed, _ = sjson.Set(completed, "response.usage.output_tokens", v.Int())
@@ -460,6 +563,7 @@ func ConvertGeminiResponseToOpenAIResponses(_ context.Context, modelName string,
 // ConvertGeminiResponseToOpenAIResponsesNonStream aggregates Gemini response JSON into a single OpenAI Responses JSON object.
 func ConvertGeminiResponseToOpenAIResponsesNonStream(_ context.Context, _ string, originalRequestRawJSON, requestRawJSON, rawJSON []byte, _ *any) string {
 	root := gjson.ParseBytes(rawJSON)
+	root = unwrapGeminiResponseRoot(root)
 
 	// Base response scaffold
 	resp := `{"id":"","object":"response","created_at":0,"status":"completed","background":false,"error":null,"incomplete_details":null}`
@@ -478,15 +582,15 @@ func ConvertGeminiResponseToOpenAIResponsesNonStream(_ context.Context, _ string
 	// created_at: map from createTime if available
 	createdAt := time.Now().Unix()
 	if v := root.Get("createTime"); v.Exists() {
-		if t, err := time.Parse(time.RFC3339Nano, v.String()); err == nil {
+		if t, errParseCreateTime := time.Parse(time.RFC3339Nano, v.String()); errParseCreateTime == nil {
 			createdAt = t.Unix()
 		}
 	}
 	resp, _ = sjson.Set(resp, "created_at", createdAt)
 
 	// Echo request fields when present; fallback model from response modelVersion
-	if len(requestRawJSON) > 0 {
-		req := gjson.ParseBytes(requestRawJSON)
+	if reqJSON := pickRequestJSON(originalRequestRawJSON, requestRawJSON); len(reqJSON) > 0 {
+		req := unwrapRequestRoot(gjson.ParseBytes(reqJSON))
 		if v := req.Get("instructions"); v.Exists() {
 			resp, _ = sjson.Set(resp, "instructions", v.String())
 		}
@@ -636,8 +740,8 @@ func ConvertGeminiResponseToOpenAIResponsesNonStream(_ context.Context, _ string
 		// input tokens = prompt + thoughts
 		input := um.Get("promptTokenCount").Int() + um.Get("thoughtsTokenCount").Int()
 		resp, _ = sjson.Set(resp, "usage.input_tokens", input)
-		// cached_tokens not provided by Gemini; default to 0 for structure compatibility
-		resp, _ = sjson.Set(resp, "usage.input_tokens_details.cached_tokens", 0)
+		// cached token details: align with OpenAI "cached_tokens" semantics.
+		resp, _ = sjson.Set(resp, "usage.input_tokens_details.cached_tokens", um.Get("cachedContentTokenCount").Int())
 		// output tokens
 		if v := um.Get("candidatesTokenCount"); v.Exists() {
 			resp, _ = sjson.Set(resp, "usage.output_tokens", v.Int())

internal/translator/gemini/openai/responses/gemini_openai-responses_response_test.go

@@ -0,0 +1,353 @@
+package responses
+
+import (
+	"context"
+	"strings"
+	"testing"
+
+	"github.com/tidwall/gjson"
+)
+
+func parseSSEEvent(t *testing.T, chunk string) (string, gjson.Result) {
+	t.Helper()
+
+	lines := strings.Split(chunk, "\n")
+	if len(lines) < 2 {
+		t.Fatalf("unexpected SSE chunk: %q", chunk)
+	}
+
+	event := strings.TrimSpace(strings.TrimPrefix(lines[0], "event:"))
+	dataLine := strings.TrimSpace(strings.TrimPrefix(lines[1], "data:"))
+	if !gjson.Valid(dataLine) {
+		t.Fatalf("invalid SSE data JSON: %q", dataLine)
+	}
+	return event, gjson.Parse(dataLine)
+}
+
+func TestConvertGeminiResponseToOpenAIResponses_UnwrapAndAggregateText(t *testing.T) {
+	// Vertex-style Gemini stream wraps the actual response payload under "response".
+	// This test ensures we unwrap and that output_text.done contains the full text.
+	in := []string{
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":""}]}}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":"让"}]}}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":"我先"}]}}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":"了解"}]}}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"functionCall":{"name":"mcp__serena__list_dir","args":{"recursive":false,"relative_path":"internal"},"id":"toolu_1"}}]}}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":""}]},"finishReason":"STOP"}],"usageMetadata":{"promptTokenCount":10,"candidatesTokenCount":5,"totalTokenCount":15,"cachedContentTokenCount":2},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+	}
+
+	originalReq := []byte(`{"instructions":"test instructions","model":"gpt-5","max_output_tokens":123}`)
+
+	var param any
+	var out []string
+	for _, line := range in {
+		out = append(out, ConvertGeminiResponseToOpenAIResponses(context.Background(), "test-model", originalReq, nil, []byte(line), &param)...)
+	}
+
+	var (
+		gotTextDone     bool
+		gotMessageDone  bool
+		gotResponseDone bool
+		gotFuncDone     bool
+
+		textDone     string
+		messageText  string
+		responseID   string
+		instructions string
+		cachedTokens int64
+
+		funcName string
+		funcArgs string
+
+		posTextDone    = -1
+		posPartDone    = -1
+		posMessageDone = -1
+		posFuncAdded   = -1
+	)
+
+	for i, chunk := range out {
+		ev, data := parseSSEEvent(t, chunk)
+		switch ev {
+		case "response.output_text.done":
+			gotTextDone = true
+			if posTextDone == -1 {
+				posTextDone = i
+			}
+			textDone = data.Get("text").String()
+		case "response.content_part.done":
+			if posPartDone == -1 {
+				posPartDone = i
+			}
+		case "response.output_item.done":
+			switch data.Get("item.type").String() {
+			case "message":
+				gotMessageDone = true
+				if posMessageDone == -1 {
+					posMessageDone = i
+				}
+				messageText = data.Get("item.content.0.text").String()
+			case "function_call":
+				gotFuncDone = true
+				funcName = data.Get("item.name").String()
+				funcArgs = data.Get("item.arguments").String()
+			}
+		case "response.output_item.added":
+			if data.Get("item.type").String() == "function_call" && posFuncAdded == -1 {
+				posFuncAdded = i
+			}
+		case "response.completed":
+			gotResponseDone = true
+			responseID = data.Get("response.id").String()
+			instructions = data.Get("response.instructions").String()
+			cachedTokens = data.Get("response.usage.input_tokens_details.cached_tokens").Int()
+		}
+	}
+
+	if !gotTextDone {
+		t.Fatalf("missing response.output_text.done event")
+	}
+	if posTextDone == -1 || posPartDone == -1 || posMessageDone == -1 || posFuncAdded == -1 {
+		t.Fatalf("missing ordering events: textDone=%d partDone=%d messageDone=%d funcAdded=%d", posTextDone, posPartDone, posMessageDone, posFuncAdded)
+	}
+	if !(posTextDone < posPartDone && posPartDone < posMessageDone && posMessageDone < posFuncAdded) {
+		t.Fatalf("unexpected message/function ordering: textDone=%d partDone=%d messageDone=%d funcAdded=%d", posTextDone, posPartDone, posMessageDone, posFuncAdded)
+	}
+	if !gotMessageDone {
+		t.Fatalf("missing message response.output_item.done event")
+	}
+	if !gotFuncDone {
+		t.Fatalf("missing function_call response.output_item.done event")
+	}
+	if !gotResponseDone {
+		t.Fatalf("missing response.completed event")
+	}
+
+	if textDone != "让我先了解" {
+		t.Fatalf("unexpected output_text.done text: got %q", textDone)
+	}
+	if messageText != "让我先了解" {
+		t.Fatalf("unexpected message done text: got %q", messageText)
+	}
+
+	if responseID != "resp_req_vrtx_1" {
+		t.Fatalf("unexpected response id: got %q", responseID)
+	}
+	if instructions != "test instructions" {
+		t.Fatalf("unexpected instructions echo: got %q", instructions)
+	}
+	if cachedTokens != 2 {
+		t.Fatalf("unexpected cached token count: got %d", cachedTokens)
+	}
+
+	if funcName != "mcp__serena__list_dir" {
+		t.Fatalf("unexpected function name: got %q", funcName)
+	}
+	if !gjson.Valid(funcArgs) {
+		t.Fatalf("invalid function arguments JSON: %q", funcArgs)
+	}
+	if gjson.Get(funcArgs, "recursive").Bool() != false {
+		t.Fatalf("unexpected recursive arg: %v", gjson.Get(funcArgs, "recursive").Value())
+	}
+	if gjson.Get(funcArgs, "relative_path").String() != "internal" {
+		t.Fatalf("unexpected relative_path arg: %q", gjson.Get(funcArgs, "relative_path").String())
+	}
+}
+
+func TestConvertGeminiResponseToOpenAIResponses_ReasoningEncryptedContent(t *testing.T) {
+	sig := "RXE0RENrZ0lDeEFDR0FJcVFOZDdjUzlleGFuRktRdFcvSzNyZ2MvWDNCcDQ4RmxSbGxOWUlOVU5kR1l1UHMrMGdkMVp0Vkg3ekdKU0g4YVljc2JjN3lNK0FrdGpTNUdqamI4T3Z0VVNETzdQd3pmcFhUOGl3U3hXUEJvTVFRQ09mWTFyMEtTWGZxUUlJakFqdmFGWk83RW1XRlBKckJVOVpkYzdDKw=="
+	in := []string{
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"thought":true,"thoughtSignature":"` + sig + `","text":""}]}}],"modelVersion":"test-model","responseId":"req_vrtx_sig"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"thought":true,"text":"a"}]}}],"modelVersion":"test-model","responseId":"req_vrtx_sig"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":"hello"}]}}],"modelVersion":"test-model","responseId":"req_vrtx_sig"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":""}]},"finishReason":"STOP"}],"modelVersion":"test-model","responseId":"req_vrtx_sig"},"traceId":"t1"}`,
+	}
+
+	var param any
+	var out []string
+	for _, line := range in {
+		out = append(out, ConvertGeminiResponseToOpenAIResponses(context.Background(), "test-model", nil, nil, []byte(line), &param)...)
+	}
+
+	var (
+		addedEnc string
+		doneEnc  string
+	)
+	for _, chunk := range out {
+		ev, data := parseSSEEvent(t, chunk)
+		switch ev {
+		case "response.output_item.added":
+			if data.Get("item.type").String() == "reasoning" {
+				addedEnc = data.Get("item.encrypted_content").String()
+			}
+		case "response.output_item.done":
+			if data.Get("item.type").String() == "reasoning" {
+				doneEnc = data.Get("item.encrypted_content").String()
+			}
+		}
+	}
+
+	if addedEnc != sig {
+		t.Fatalf("unexpected encrypted_content in response.output_item.added: got %q", addedEnc)
+	}
+	if doneEnc != sig {
+		t.Fatalf("unexpected encrypted_content in response.output_item.done: got %q", doneEnc)
+	}
+}
+
+func TestConvertGeminiResponseToOpenAIResponses_FunctionCallEventOrder(t *testing.T) {
+	in := []string{
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"functionCall":{"name":"tool0"}}]}}],"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"functionCall":{"name":"tool1"}}]}}],"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"functionCall":{"name":"tool2","args":{"a":1}}}]}}],"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":""}]},"finishReason":"STOP"}],"usageMetadata":{"promptTokenCount":10,"candidatesTokenCount":5,"totalTokenCount":15,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_1"},"traceId":"t1"}`,
+	}
+
+	var param any
+	var out []string
+	for _, line := range in {
+		out = append(out, ConvertGeminiResponseToOpenAIResponses(context.Background(), "test-model", nil, nil, []byte(line), &param)...)
+	}
+
+	posAdded := []int{-1, -1, -1}
+	posArgsDelta := []int{-1, -1, -1}
+	posArgsDone := []int{-1, -1, -1}
+	posItemDone := []int{-1, -1, -1}
+	posCompleted := -1
+	deltaByIndex := map[int]string{}
+
+	for i, chunk := range out {
+		ev, data := parseSSEEvent(t, chunk)
+		switch ev {
+		case "response.output_item.added":
+			if data.Get("item.type").String() != "function_call" {
+				continue
+			}
+			idx := int(data.Get("output_index").Int())
+			if idx >= 0 && idx < len(posAdded) {
+				posAdded[idx] = i
+			}
+		case "response.function_call_arguments.delta":
+			idx := int(data.Get("output_index").Int())
+			if idx >= 0 && idx < len(posArgsDelta) {
+				posArgsDelta[idx] = i
+				deltaByIndex[idx] = data.Get("delta").String()
+			}
+		case "response.function_call_arguments.done":
+			idx := int(data.Get("output_index").Int())
+			if idx >= 0 && idx < len(posArgsDone) {
+				posArgsDone[idx] = i
+			}
+		case "response.output_item.done":
+			if data.Get("item.type").String() != "function_call" {
+				continue
+			}
+			idx := int(data.Get("output_index").Int())
+			if idx >= 0 && idx < len(posItemDone) {
+				posItemDone[idx] = i
+			}
+		case "response.completed":
+			posCompleted = i
+
+			output := data.Get("response.output")
+			if !output.Exists() || !output.IsArray() {
+				t.Fatalf("missing response.output in response.completed")
+			}
+			if len(output.Array()) != 3 {
+				t.Fatalf("unexpected response.output length: got %d", len(output.Array()))
+			}
+			if data.Get("response.output.0.name").String() != "tool0" || data.Get("response.output.0.arguments").String() != "{}" {
+				t.Fatalf("unexpected output[0]: %s", data.Get("response.output.0").Raw)
+			}
+			if data.Get("response.output.1.name").String() != "tool1" || data.Get("response.output.1.arguments").String() != "{}" {
+				t.Fatalf("unexpected output[1]: %s", data.Get("response.output.1").Raw)
+			}
+			if data.Get("response.output.2.name").String() != "tool2" {
+				t.Fatalf("unexpected output[2] name: %s", data.Get("response.output.2").Raw)
+			}
+			if !gjson.Valid(data.Get("response.output.2.arguments").String()) {
+				t.Fatalf("unexpected output[2] arguments: %q", data.Get("response.output.2.arguments").String())
+			}
+		}
+	}
+
+	if posCompleted == -1 {
+		t.Fatalf("missing response.completed event")
+	}
+	for idx := 0; idx < 3; idx++ {
+		if posAdded[idx] == -1 || posArgsDelta[idx] == -1 || posArgsDone[idx] == -1 || posItemDone[idx] == -1 {
+			t.Fatalf("missing function call events for output_index %d: added=%d argsDelta=%d argsDone=%d itemDone=%d", idx, posAdded[idx], posArgsDelta[idx], posArgsDone[idx], posItemDone[idx])
+		}
+		if !(posAdded[idx] < posArgsDelta[idx] && posArgsDelta[idx] < posArgsDone[idx] && posArgsDone[idx] < posItemDone[idx]) {
+			t.Fatalf("unexpected ordering for output_index %d: added=%d argsDelta=%d argsDone=%d itemDone=%d", idx, posAdded[idx], posArgsDelta[idx], posArgsDone[idx], posItemDone[idx])
+		}
+		if idx > 0 && !(posItemDone[idx-1] < posAdded[idx]) {
+			t.Fatalf("function call events overlap between %d and %d: prevDone=%d nextAdded=%d", idx-1, idx, posItemDone[idx-1], posAdded[idx])
+		}
+	}
+
+	if deltaByIndex[0] != "{}" {
+		t.Fatalf("unexpected delta for output_index 0: got %q", deltaByIndex[0])
+	}
+	if deltaByIndex[1] != "{}" {
+		t.Fatalf("unexpected delta for output_index 1: got %q", deltaByIndex[1])
+	}
+	if deltaByIndex[2] == "" || !gjson.Valid(deltaByIndex[2]) || gjson.Get(deltaByIndex[2], "a").Int() != 1 {
+		t.Fatalf("unexpected delta for output_index 2: got %q", deltaByIndex[2])
+	}
+	if !(posItemDone[2] < posCompleted) {
+		t.Fatalf("response.completed should be after last output_item.done: last=%d completed=%d", posItemDone[2], posCompleted)
+	}
+}
+
+func TestConvertGeminiResponseToOpenAIResponses_ResponseOutputOrdering(t *testing.T) {
+	in := []string{
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"functionCall":{"name":"tool0","args":{"x":"y"}}}]}}],"modelVersion":"test-model","responseId":"req_vrtx_2"},"traceId":"t2"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":"hi"}]}}],"modelVersion":"test-model","responseId":"req_vrtx_2"},"traceId":"t2"}`,
+		`data: {"response":{"candidates":[{"content":{"role":"model","parts":[{"text":""}]},"finishReason":"STOP"}],"usageMetadata":{"promptTokenCount":1,"candidatesTokenCount":1,"totalTokenCount":2,"cachedContentTokenCount":0},"modelVersion":"test-model","responseId":"req_vrtx_2"},"traceId":"t2"}`,
+	}
+
+	var param any
+	var out []string
+	for _, line := range in {
+		out = append(out, ConvertGeminiResponseToOpenAIResponses(context.Background(), "test-model", nil, nil, []byte(line), &param)...)
+	}
+
+	posFuncDone := -1
+	posMsgAdded := -1
+	posCompleted := -1
+
+	for i, chunk := range out {
+		ev, data := parseSSEEvent(t, chunk)
+		switch ev {
+		case "response.output_item.done":
+			if data.Get("item.type").String() == "function_call" && data.Get("output_index").Int() == 0 {
+				posFuncDone = i
+			}
+		case "response.output_item.added":
+			if data.Get("item.type").String() == "message" && data.Get("output_index").Int() == 1 {
+				posMsgAdded = i
+			}
+		case "response.completed":
+			posCompleted = i
+			if data.Get("response.output.0.type").String() != "function_call" {
+				t.Fatalf("expected response.output[0] to be function_call: %s", data.Get("response.output.0").Raw)
+			}
+			if data.Get("response.output.1.type").String() != "message" {
+				t.Fatalf("expected response.output[1] to be message: %s", data.Get("response.output.1").Raw)
+			}
+			if data.Get("response.output.1.content.0.text").String() != "hi" {
+				t.Fatalf("unexpected message text in response.output[1]: %s", data.Get("response.output.1").Raw)
+			}
+		}
+	}
+
+	if posFuncDone == -1 || posMsgAdded == -1 || posCompleted == -1 {
+		t.Fatalf("missing required events: funcDone=%d msgAdded=%d completed=%d", posFuncDone, posMsgAdded, posCompleted)
+	}
+	if !(posFuncDone < posMsgAdded) {
+		t.Fatalf("expected function_call to complete before message is added: funcDone=%d msgAdded=%d", posFuncDone, posMsgAdded)
+	}
+	if !(posMsgAdded < posCompleted) {
+		t.Fatalf("expected response.completed after message added: msgAdded=%d completed=%d", posMsgAdded, posCompleted)
+	}
+}

internal/translator/openai/claude/openai_claude_request.go

@@ -88,7 +88,7 @@ func ConvertClaudeRequestToOpenAI(modelName string, inputRawJSON []byte, stream
 	var messagesJSON = "[]"
 
 	// Handle system message first
-	systemMsgJSON := `{"role":"system","content":[{"type":"text","text":"Use ANY tool, the parameters MUST accord with RFC 8259 (The JavaScript Object Notation (JSON) Data Interchange Format), the keys and value MUST be enclosed in double quotes."}]}`
+	systemMsgJSON := `{"role":"system","content":[]}`
 	if system := root.Get("system"); system.Exists() {
 		if system.Type == gjson.String {
 			if system.String() != "" {

internal/translator/openai/claude/openai_claude_response.go

@@ -289,21 +289,17 @@ func convertOpenAIStreamingChunkToAnthropic(rawJSON []byte, param *ConvertOpenAI
 	// Only process if usage has actual values (not null)
 	if param.FinishReason != "" {
 		usage := root.Get("usage")
-		var inputTokens, outputTokens int64
+		var inputTokens, outputTokens, cachedTokens int64
 		if usage.Exists() && usage.Type != gjson.Null {
-			// Check if usage has actual token counts
-			promptTokens := usage.Get("prompt_tokens")
-			completionTokens := usage.Get("completion_tokens")
-
-			if promptTokens.Exists() && completionTokens.Exists() {
-				inputTokens = promptTokens.Int()
-				outputTokens = completionTokens.Int()
-			}
+			inputTokens, outputTokens, cachedTokens = extractOpenAIUsage(usage)
 			// Send message_delta with usage
 			messageDeltaJSON := `{"type":"message_delta","delta":{"stop_reason":"","stop_sequence":null},"usage":{"input_tokens":0,"output_tokens":0}}`
 			messageDeltaJSON, _ = sjson.Set(messageDeltaJSON, "delta.stop_reason", mapOpenAIFinishReasonToAnthropic(param.FinishReason))
 			messageDeltaJSON, _ = sjson.Set(messageDeltaJSON, "usage.input_tokens", inputTokens)
 			messageDeltaJSON, _ = sjson.Set(messageDeltaJSON, "usage.output_tokens", outputTokens)
+			if cachedTokens > 0 {
+				messageDeltaJSON, _ = sjson.Set(messageDeltaJSON, "usage.cache_read_input_tokens", cachedTokens)
+			}
 			results = append(results, "event: message_delta\ndata: "+messageDeltaJSON+"\n\n")
 			param.MessageDeltaSent = true
 
@@ -423,13 +419,12 @@ func convertOpenAINonStreamingToAnthropic(rawJSON []byte) []string {
 
 	// Set usage information
 	if usage := root.Get("usage"); usage.Exists() {
-		out, _ = sjson.Set(out, "usage.input_tokens", usage.Get("prompt_tokens").Int())
-		out, _ = sjson.Set(out, "usage.output_tokens", usage.Get("completion_tokens").Int())
-		reasoningTokens := int64(0)
-		if v := usage.Get("completion_tokens_details.reasoning_tokens"); v.Exists() {
-			reasoningTokens = v.Int()
+		inputTokens, outputTokens, cachedTokens := extractOpenAIUsage(usage)
+		out, _ = sjson.Set(out, "usage.input_tokens", inputTokens)
+		out, _ = sjson.Set(out, "usage.output_tokens", outputTokens)
+		if cachedTokens > 0 {
+			out, _ = sjson.Set(out, "usage.cache_read_input_tokens", cachedTokens)
 		}
-		out, _ = sjson.Set(out, "usage.reasoning_tokens", reasoningTokens)
 	}
 
 	return []string{out}
@@ -674,8 +669,12 @@ func ConvertOpenAIResponseToClaudeNonStream(_ context.Context, _ string, origina
 	}
 
 	if respUsage := root.Get("usage"); respUsage.Exists() {
-		out, _ = sjson.Set(out, "usage.input_tokens", respUsage.Get("prompt_tokens").Int())
-		out, _ = sjson.Set(out, "usage.output_tokens", respUsage.Get("completion_tokens").Int())
+		inputTokens, outputTokens, cachedTokens := extractOpenAIUsage(respUsage)
+		out, _ = sjson.Set(out, "usage.input_tokens", inputTokens)
+		out, _ = sjson.Set(out, "usage.output_tokens", outputTokens)
+		if cachedTokens > 0 {
+			out, _ = sjson.Set(out, "usage.cache_read_input_tokens", cachedTokens)
+		}
 	}
 
 	if !stopReasonSet {
@@ -692,3 +691,23 @@ func ConvertOpenAIResponseToClaudeNonStream(_ context.Context, _ string, origina
 func ClaudeTokenCount(ctx context.Context, count int64) string {
 	return fmt.Sprintf(`{"input_tokens":%d}`, count)
 }
+
+func extractOpenAIUsage(usage gjson.Result) (int64, int64, int64) {
+	if !usage.Exists() || usage.Type == gjson.Null {
+		return 0, 0, 0
+	}
+
+	inputTokens := usage.Get("prompt_tokens").Int()
+	outputTokens := usage.Get("completion_tokens").Int()
+	cachedTokens := usage.Get("prompt_tokens_details.cached_tokens").Int()
+
+	if cachedTokens > 0 {
+		if inputTokens >= cachedTokens {
+			inputTokens -= cachedTokens
+		} else {
+			inputTokens = 0
+		}
+	}
+
+	return inputTokens, outputTokens, cachedTokens
+}

internal/translator/openai/gemini/openai_gemini_request.go

@@ -77,7 +77,13 @@ func ConvertGeminiRequestToOpenAI(modelName string, inputRawJSON []byte, stream
 			}
 		}
 
+		// Candidate count (OpenAI 'n' parameter)
+		if candidateCount := genConfig.Get("candidateCount"); candidateCount.Exists() {
+			out, _ = sjson.Set(out, "n", candidateCount.Int())
+		}
+
 		// Map Gemini thinkingConfig to OpenAI reasoning_effort.
+		// Always perform conversion to support allowCompat models that may not be in registry
 		if thinkingConfig := genConfig.Get("thinkingConfig"); thinkingConfig.Exists() && thinkingConfig.IsObject() {
 			if thinkingLevel := thinkingConfig.Get("thinkingLevel"); thinkingLevel.Exists() {
 				effort := strings.ToLower(strings.TrimSpace(thinkingLevel.String()))

sdk/auth/codex.go

@@ -2,6 +2,8 @@ package auth
 
 import (
 	"context"
+	"crypto/sha256"
+	"encoding/hex"
 	"fmt"
 	"net/http"
 	"strings"
@@ -191,7 +193,19 @@ waitForCallback:
 		return nil, fmt.Errorf("codex token storage missing account information")
 	}
 
-	fileName := fmt.Sprintf("codex-%s.json", tokenStorage.Email)
+	planType := ""
+	hashAccountID := ""
+	if tokenStorage.IDToken != "" {
+		if claims, errParse := codex.ParseJWTToken(tokenStorage.IDToken); errParse == nil && claims != nil {
+			planType = strings.TrimSpace(claims.CodexAuthInfo.ChatgptPlanType)
+			accountID := strings.TrimSpace(claims.CodexAuthInfo.ChatgptAccountID)
+			if accountID != "" {
+				digest := sha256.Sum256([]byte(accountID))
+				hashAccountID = hex.EncodeToString(digest[:])[:8]
+			}
+		}
+	}
+	fileName := codex.CredentialFileName(tokenStorage.Email, planType, hashAccountID, true)
 	metadata := map[string]any{
 		"email": tokenStorage.Email,
 	}

sdk/auth/filestore.go

@@ -8,7 +8,6 @@ import (
 	"net/http"
 	"os"
 	"path/filepath"
-	"reflect"
 	"strings"
 	"sync"
 	"time"
@@ -76,7 +75,7 @@ func (s *FileTokenStore) Save(ctx context.Context, auth *cliproxyauth.Auth) (str
 		if existing, errRead := os.ReadFile(path); errRead == nil {
 			// Use metadataEqualIgnoringTimestamps to skip writes when only timestamp fields change.
 			// This prevents the token refresh loop caused by timestamp/expired/expires_in changes.
-			if metadataEqualIgnoringTimestamps(existing, raw) {
+			if metadataEqualIgnoringTimestamps(existing, raw, auth.Provider) {
 				return path, nil
 			}
 			file, errOpen := os.OpenFile(path, os.O_WRONLY|os.O_TRUNC, 0o600)
@@ -300,28 +299,101 @@ func (s *FileTokenStore) baseDirSnapshot() string {
 	return s.baseDir
 }
 
-// metadataEqualIgnoringTimestamps compares two metadata JSON blobs, ignoring volatile fields that
-// change on every refresh but don't affect authentication logic.
-func metadataEqualIgnoringTimestamps(a, b []byte) bool {
-	var objA map[string]any
-	var objB map[string]any
-	if errUnmarshalA := json.Unmarshal(a, &objA); errUnmarshalA != nil {
+// DEPRECATED: Use metadataEqualIgnoringTimestamps for comparing auth metadata.
+// This function is kept for backward compatibility but can cause refresh loops.
+func jsonEqual(a, b []byte) bool {
+	var objA any
+	var objB any
+	if err := json.Unmarshal(a, &objA); err != nil {
 		return false
 	}
-	if errUnmarshalB := json.Unmarshal(b, &objB); errUnmarshalB != nil {
+	if err := json.Unmarshal(b, &objB); err != nil {
 		return false
 	}
-	stripVolatileMetadataFields(objA)
-	stripVolatileMetadataFields(objB)
-	return reflect.DeepEqual(objA, objB)
+	return deepEqualJSON(objA, objB)
 }
 
-func stripVolatileMetadataFields(metadata map[string]any) {
-	if metadata == nil {
-		return
+// metadataEqualIgnoringTimestamps compares two metadata JSON blobs,
+// ignoring fields that change on every refresh but don't affect functionality.
+// This prevents unnecessary file writes that would trigger watcher events and
+// create refresh loops.
+// The provider parameter controls whether access_token is ignored: providers like
+// Google OAuth (gemini, gemini-cli) can re-fetch tokens when needed, while others
+// like iFlow require the refreshed token to be persisted.
+func metadataEqualIgnoringTimestamps(a, b []byte, provider string) bool {
+	var objA, objB map[string]any
+	if err := json.Unmarshal(a, &objA); err != nil {
+		return false
 	}
-	// These fields change on refresh and would otherwise trigger watcher reload loops.
-	for _, field := range []string{"timestamp", "expired", "expires_in", "last_refresh", "access_token"} {
-		delete(metadata, field)
+	if err := json.Unmarshal(b, &objB); err != nil {
+		return false
+	}
+
+	// Fields to ignore: these change on every refresh but don't affect authentication logic.
+	// - timestamp, expired, expires_in, last_refresh: time-related fields that change on refresh
+	ignoredFields := []string{"timestamp", "expired", "expires_in", "last_refresh"}
+
+	// For providers that can re-fetch tokens when needed (e.g., Google OAuth),
+	// we ignore access_token to avoid unnecessary file writes.
+	switch provider {
+	case "gemini", "gemini-cli", "antigravity":
+		ignoredFields = append(ignoredFields, "access_token")
+	}
+
+	for _, field := range ignoredFields {
+		delete(objA, field)
+		delete(objB, field)
+	}
+
+	return deepEqualJSON(objA, objB)
+}
+
+func deepEqualJSON(a, b any) bool {
+	switch valA := a.(type) {
+	case map[string]any:
+		valB, ok := b.(map[string]any)
+		if !ok || len(valA) != len(valB) {
+			return false
+		}
+		for key, subA := range valA {
+			subB, ok1 := valB[key]
+			if !ok1 || !deepEqualJSON(subA, subB) {
+				return false
+			}
+		}
+		return true
+	case []any:
+		sliceB, ok := b.([]any)
+		if !ok || len(valA) != len(sliceB) {
+			return false
+		}
+		for i := range valA {
+			if !deepEqualJSON(valA[i], sliceB[i]) {
+				return false
+			}
+		}
+		return true
+	case float64:
+		valB, ok := b.(float64)
+		if !ok {
+			return false
+		}
+		return valA == valB
+	case string:
+		valB, ok := b.(string)
+		if !ok {
+			return false
+		}
+		return valA == valB
+	case bool:
+		valB, ok := b.(bool)
+		if !ok {
+			return false
+		}
+		return valA == valB
+	case nil:
+		return b == nil
+	default:
+		return false
 	}
 }

test/thinking_conversion_test.go

@@ -20,6 +20,7 @@ import (
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
 	sdktranslator "github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
 	"github.com/tidwall/gjson"
+	"github.com/tidwall/sjson"
 )
 
 // thinkingTestCase represents a common test case structure for both suffix and body tests.
@@ -2707,8 +2708,11 @@ func runThinkingTests(t *testing.T, cases []thinkingTestCase) {
 				[]byte(tc.inputJSON),
 				true,
 			)
+			if applyTo == "claude" {
+				body, _ = sjson.SetBytes(body, "max_tokens", 200000)
+			}
 
-			body, err := thinking.ApplyThinking(body, tc.model, tc.from, applyTo)
+			body, err := thinking.ApplyThinking(body, tc.model, tc.from, applyTo, applyTo)
 
 			if tc.expectErr {
 				if err == nil {