Merge pull request #814 from router-for-me/aistudio

Fix model alias thinking suffix

internal/managementasset/updater.go

@@ -24,10 +24,11 @@ import (
 )
 
 const (
-	defaultManagementReleaseURL = "https://api.github.com/repos/router-for-me/Cli-Proxy-API-Management-Center/releases/latest"
-	managementAssetName         = "management.html"
-	httpUserAgent               = "CLIProxyAPI-management-updater"
-	updateCheckInterval         = 3 * time.Hour
+	defaultManagementReleaseURL  = "https://api.github.com/repos/router-for-me/Cli-Proxy-API-Management-Center/releases/latest"
+	defaultManagementFallbackURL = "https://cpamc.router-for.me/"
+	managementAssetName          = "management.html"
+	httpUserAgent                = "CLIProxyAPI-management-updater"
+	updateCheckInterval          = 3 * time.Hour
 )
 
 // ManagementFileName exposes the control panel asset filename.
@@ -198,6 +199,16 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 		return
 	}
 
+	localPath := filepath.Join(staticDir, managementAssetName)
+	localFileMissing := false
+	if _, errStat := os.Stat(localPath); errStat != nil {
+		if errors.Is(errStat, os.ErrNotExist) {
+			localFileMissing = true
+		} else {
+			log.WithError(errStat).Debug("failed to stat local management asset")
+		}
+	}
+
 	// Rate limiting: check only once every 3 hours
 	lastUpdateCheckMu.Lock()
 	now := time.Now()
@@ -210,15 +221,14 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 	lastUpdateCheckTime = now
 	lastUpdateCheckMu.Unlock()
 
-	if err := os.MkdirAll(staticDir, 0o755); err != nil {
-		log.WithError(err).Warn("failed to prepare static directory for management asset")
+	if errMkdirAll := os.MkdirAll(staticDir, 0o755); errMkdirAll != nil {
+		log.WithError(errMkdirAll).Warn("failed to prepare static directory for management asset")
 		return
 	}
 
 	releaseURL := resolveReleaseURL(panelRepository)
 	client := newHTTPClient(proxyURL)
 
-	localPath := filepath.Join(staticDir, managementAssetName)
 	localHash, err := fileSHA256(localPath)
 	if err != nil {
 		if !errors.Is(err, os.ErrNotExist) {
@@ -229,6 +239,13 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 
 	asset, remoteHash, err := fetchLatestAsset(ctx, client, releaseURL)
 	if err != nil {
+		if localFileMissing {
+			log.WithError(err).Warn("failed to fetch latest management release information, trying fallback page")
+			if ensureFallbackManagementHTML(ctx, client, localPath) {
+				return
+			}
+			return
+		}
 		log.WithError(err).Warn("failed to fetch latest management release information")
 		return
 	}
@@ -240,6 +257,13 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 
 	data, downloadedHash, err := downloadAsset(ctx, client, asset.BrowserDownloadURL)
 	if err != nil {
+		if localFileMissing {
+			log.WithError(err).Warn("failed to download management asset, trying fallback page")
+			if ensureFallbackManagementHTML(ctx, client, localPath) {
+				return
+			}
+			return
+		}
 		log.WithError(err).Warn("failed to download management asset")
 		return
 	}
@@ -256,6 +280,22 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 	log.Infof("management asset updated successfully (hash=%s)", downloadedHash)
 }
 
+func ensureFallbackManagementHTML(ctx context.Context, client *http.Client, localPath string) bool {
+	data, downloadedHash, err := downloadAsset(ctx, client, defaultManagementFallbackURL)
+	if err != nil {
+		log.WithError(err).Warn("failed to download fallback management control panel page")
+		return false
+	}
+
+	if err = atomicWriteFile(localPath, data); err != nil {
+		log.WithError(err).Warn("failed to persist fallback management control panel page")
+		return false
+	}
+
+	log.Infof("management asset updated from fallback page successfully (hash=%s)", downloadedHash)
+	return true
+}
+
 func resolveReleaseURL(repo string) string {
 	repo = strings.TrimSpace(repo)
 	if repo == "" {

internal/registry/model_definitions.go

@@ -773,7 +773,7 @@ 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"},
-		"gemini-2.5-computer-use-preview-10-2025": {Name: "models/gemini-2.5-computer-use-preview-10-2025"},
+		"gemini-2.5-computer-use-preview-10-2025": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}, Name: "models/gemini-2.5-computer-use-preview-10-2025"},
 		"gemini-3-pro-preview":                    {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}, Name: "models/gemini-3-pro-preview"},
 		"gemini-3-pro-image-preview":              {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}, Name: "models/gemini-3-pro-image-preview"},
 		"gemini-3-flash-preview":                  {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}, Name: "models/gemini-3-flash-preview"},

internal/runtime/executor/antigravity_executor.go

@@ -96,9 +96,9 @@ func (e *AntigravityExecutor) Execute(ctx context.Context, auth *cliproxyauth.Au
 	to := sdktranslator.FromString("antigravity")
 	translated := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
 
-	translated = applyThinkingMetadataCLI(translated, req.Metadata, req.Model)
+	translated = ApplyThinkingMetadataCLI(translated, req.Metadata, req.Model)
 	translated = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, translated)
-	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, translated)
+	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, translated)
 	translated = normalizeAntigravityThinking(req.Model, translated, isClaude)
 	translated = applyPayloadConfigWithRoot(e.cfg, req.Model, "antigravity", "request", translated)
 
@@ -191,9 +191,9 @@ func (e *AntigravityExecutor) executeClaudeNonStream(ctx context.Context, auth *
 	to := sdktranslator.FromString("antigravity")
 	translated := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
 
-	translated = applyThinkingMetadataCLI(translated, req.Metadata, req.Model)
+	translated = ApplyThinkingMetadataCLI(translated, req.Metadata, req.Model)
 	translated = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, translated)
-	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, translated)
+	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, translated)
 	translated = normalizeAntigravityThinking(req.Model, translated, true)
 	translated = applyPayloadConfigWithRoot(e.cfg, req.Model, "antigravity", "request", translated)
 
@@ -527,9 +527,9 @@ func (e *AntigravityExecutor) ExecuteStream(ctx context.Context, auth *cliproxya
 	to := sdktranslator.FromString("antigravity")
 	translated := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
 
-	translated = applyThinkingMetadataCLI(translated, req.Metadata, req.Model)
+	translated = ApplyThinkingMetadataCLI(translated, req.Metadata, req.Model)
 	translated = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, translated)
-	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, translated)
+	translated = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, translated)
 	translated = normalizeAntigravityThinking(req.Model, translated, isClaude)
 	translated = applyPayloadConfigWithRoot(e.cfg, req.Model, "antigravity", "request", translated)
 
@@ -697,8 +697,8 @@ func (e *AntigravityExecutor) CountTokens(ctx context.Context, auth *cliproxyaut
 
 	for idx, baseURL := range baseURLs {
 		payload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
-		payload = applyThinkingMetadataCLI(payload, req.Metadata, req.Model)
-		payload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, payload)
+		payload = ApplyThinkingMetadataCLI(payload, req.Metadata, req.Model)
+		payload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, payload)
 		payload = normalizeAntigravityThinking(req.Model, payload, isClaude)
 		payload = deleteJSONField(payload, "project")
 		payload = deleteJSONField(payload, "model")

internal/runtime/executor/gemini_cli_executor.go

@@ -78,9 +78,9 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini-cli")
 	basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
-	basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
+	basePayload = ApplyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
 	basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, basePayload)
-	basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, basePayload)
+	basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, basePayload)
 	basePayload = util.NormalizeGeminiCLIThinkingBudget(req.Model, basePayload)
 	basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
 	basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
@@ -217,9 +217,9 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini-cli")
 	basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
-	basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
+	basePayload = ApplyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
 	basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, basePayload)
-	basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, basePayload)
+	basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, req.Metadata, basePayload)
 	basePayload = util.NormalizeGeminiCLIThinkingBudget(req.Model, basePayload)
 	basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
 	basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
@@ -421,7 +421,7 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
 	// Gemini CLI endpoint when iterating fallback variants.
 	for _, attemptModel := range models {
 		payload := sdktranslator.TranslateRequest(from, to, attemptModel, bytes.Clone(req.Payload), false)
-		payload = applyThinkingMetadataCLI(payload, req.Metadata, req.Model)
+		payload = ApplyThinkingMetadataCLI(payload, req.Metadata, req.Model)
 		payload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, payload)
 		payload = deleteJSONField(payload, "project")
 		payload = deleteJSONField(payload, "model")

internal/runtime/executor/payload_helpers.go

@@ -14,32 +14,54 @@ import (
 // ApplyThinkingMetadata applies thinking config from model suffix metadata (e.g., (high), (8192))
 // for standard Gemini format payloads. It normalizes the budget when the model supports thinking.
 func ApplyThinkingMetadata(payload []byte, metadata map[string]any, model string) []byte {
-	budgetOverride, includeOverride, ok := util.ResolveThinkingConfigFromMetadata(model, metadata)
+	// Use the alias from metadata if available, as it's registered in the global registry
+	// with thinking metadata; the upstream model name may not be registered.
+	lookupModel := util.ResolveOriginalModel(model, metadata)
+
+	// Determine which model to use for thinking support check.
+	// If the alias (lookupModel) is not in the registry, fall back to the upstream model.
+	thinkingModel := lookupModel
+	if !util.ModelSupportsThinking(lookupModel) && util.ModelSupportsThinking(model) {
+		thinkingModel = model
+	}
+
+	budgetOverride, includeOverride, ok := util.ResolveThinkingConfigFromMetadata(thinkingModel, metadata)
 	if !ok || (budgetOverride == nil && includeOverride == nil) {
 		return payload
 	}
-	if !util.ModelSupportsThinking(model) {
+	if !util.ModelSupportsThinking(thinkingModel) {
 		return payload
 	}
 	if budgetOverride != nil {
-		norm := util.NormalizeThinkingBudget(model, *budgetOverride)
+		norm := util.NormalizeThinkingBudget(thinkingModel, *budgetOverride)
 		budgetOverride = &norm
 	}
 	return util.ApplyGeminiThinkingConfig(payload, budgetOverride, includeOverride)
 }
 
-// applyThinkingMetadataCLI applies thinking config from model suffix metadata (e.g., (high), (8192))
+// ApplyThinkingMetadataCLI applies thinking config from model suffix metadata (e.g., (high), (8192))
 // for Gemini CLI format payloads (nested under "request"). It normalizes the budget when the model supports thinking.
-func applyThinkingMetadataCLI(payload []byte, metadata map[string]any, model string) []byte {
-	budgetOverride, includeOverride, ok := util.ResolveThinkingConfigFromMetadata(model, metadata)
+func ApplyThinkingMetadataCLI(payload []byte, metadata map[string]any, model string) []byte {
+	// Use the alias from metadata if available, as it's registered in the global registry
+	// with thinking metadata; the upstream model name may not be registered.
+	lookupModel := util.ResolveOriginalModel(model, metadata)
+
+	// Determine which model to use for thinking support check.
+	// If the alias (lookupModel) is not in the registry, fall back to the upstream model.
+	thinkingModel := lookupModel
+	if !util.ModelSupportsThinking(lookupModel) && util.ModelSupportsThinking(model) {
+		thinkingModel = model
+	}
+
+	budgetOverride, includeOverride, ok := util.ResolveThinkingConfigFromMetadata(thinkingModel, metadata)
 	if !ok || (budgetOverride == nil && includeOverride == nil) {
 		return payload
 	}
-	if !util.ModelSupportsThinking(model) {
+	if !util.ModelSupportsThinking(thinkingModel) {
 		return payload
 	}
 	if budgetOverride != nil {
-		norm := util.NormalizeThinkingBudget(model, *budgetOverride)
+		norm := util.NormalizeThinkingBudget(thinkingModel, *budgetOverride)
 		budgetOverride = &norm
 	}
 	return util.ApplyGeminiCLIThinkingConfig(payload, budgetOverride, includeOverride)

internal/util/gemini_thinking.go

@@ -288,37 +288,73 @@ func ApplyDefaultThinkingIfNeeded(model string, body []byte) []byte {
 
 // ApplyGemini3ThinkingLevelFromMetadata applies thinkingLevel from metadata for Gemini 3 models.
 // For standard Gemini API format (generationConfig.thinkingConfig path).
-// This handles the case where reasoning_effort is specified via model name suffix (e.g., model(minimal)).
+// This handles the case where reasoning_effort is specified via model name suffix (e.g., model(minimal))
+// or numeric budget suffix (e.g., model(1000)) which gets converted to a thinkingLevel.
 func ApplyGemini3ThinkingLevelFromMetadata(model string, metadata map[string]any, body []byte) []byte {
-	if !IsGemini3Model(model) {
+	// Use the alias from metadata if available for model type detection
+	lookupModel := ResolveOriginalModel(model, metadata)
+	if !IsGemini3Model(lookupModel) && !IsGemini3Model(model) {
 		return body
 	}
+
+	// Determine which model to use for validation
+	checkModel := model
+	if IsGemini3Model(lookupModel) {
+		checkModel = lookupModel
+	}
+
+	// First try to get effort string from metadata
 	effort, ok := ReasoningEffortFromMetadata(metadata)
-	if !ok || effort == "" {
-		return body
+	if ok && effort != "" {
+		if level, valid := ValidateGemini3ThinkingLevel(checkModel, effort); valid {
+			return ApplyGeminiThinkingLevel(body, level, nil)
+		}
 	}
-	// Validate and apply the thinkingLevel
-	if level, valid := ValidateGemini3ThinkingLevel(model, effort); valid {
-		return ApplyGeminiThinkingLevel(body, level, nil)
+
+	// Fallback: check for numeric budget and convert to thinkingLevel
+	budget, _, _, matched := ThinkingFromMetadata(metadata)
+	if matched && budget != nil {
+		if level, valid := ThinkingBudgetToGemini3Level(checkModel, *budget); valid {
+			return ApplyGeminiThinkingLevel(body, level, nil)
+		}
 	}
+
 	return body
 }
 
 // ApplyGemini3ThinkingLevelFromMetadataCLI applies thinkingLevel from metadata for Gemini 3 models.
 // For Gemini CLI API format (request.generationConfig.thinkingConfig path).
-// This handles the case where reasoning_effort is specified via model name suffix (e.g., model(minimal)).
+// This handles the case where reasoning_effort is specified via model name suffix (e.g., model(minimal))
+// or numeric budget suffix (e.g., model(1000)) which gets converted to a thinkingLevel.
 func ApplyGemini3ThinkingLevelFromMetadataCLI(model string, metadata map[string]any, body []byte) []byte {
-	if !IsGemini3Model(model) {
+	// Use the alias from metadata if available for model type detection
+	lookupModel := ResolveOriginalModel(model, metadata)
+	if !IsGemini3Model(lookupModel) && !IsGemini3Model(model) {
 		return body
 	}
+
+	// Determine which model to use for validation
+	checkModel := model
+	if IsGemini3Model(lookupModel) {
+		checkModel = lookupModel
+	}
+
+	// First try to get effort string from metadata
 	effort, ok := ReasoningEffortFromMetadata(metadata)
-	if !ok || effort == "" {
-		return body
+	if ok && effort != "" {
+		if level, valid := ValidateGemini3ThinkingLevel(checkModel, effort); valid {
+			return ApplyGeminiCLIThinkingLevel(body, level, nil)
+		}
 	}
-	// Validate and apply the thinkingLevel
-	if level, valid := ValidateGemini3ThinkingLevel(model, effort); valid {
-		return ApplyGeminiCLIThinkingLevel(body, level, nil)
+
+	// Fallback: check for numeric budget and convert to thinkingLevel
+	budget, _, _, matched := ThinkingFromMetadata(metadata)
+	if matched && budget != nil {
+		if level, valid := ThinkingBudgetToGemini3Level(checkModel, *budget); valid {
+			return ApplyGeminiCLIThinkingLevel(body, level, nil)
+		}
 	}
+
 	return body
 }
 
@@ -326,15 +362,17 @@ func ApplyGemini3ThinkingLevelFromMetadataCLI(model string, metadata map[string]
 // For Gemini CLI API format (request.generationConfig.thinkingConfig path).
 // Returns the modified body if thinkingConfig was added, otherwise returns the original.
 // For Gemini 3 models, uses thinkingLevel instead of thinkingBudget per Google's documentation.
-func ApplyDefaultThinkingIfNeededCLI(model string, body []byte) []byte {
-	if !ModelHasDefaultThinking(model) {
+func ApplyDefaultThinkingIfNeededCLI(model string, metadata map[string]any, body []byte) []byte {
+	// Use the alias from metadata if available for model property lookup
+	lookupModel := ResolveOriginalModel(model, metadata)
+	if !ModelHasDefaultThinking(lookupModel) && !ModelHasDefaultThinking(model) {
 		return body
 	}
 	if gjson.GetBytes(body, "request.generationConfig.thinkingConfig").Exists() {
 		return body
 	}
 	// Gemini 3 models use thinkingLevel instead of thinkingBudget
-	if IsGemini3Model(model) {
+	if IsGemini3Model(lookupModel) || IsGemini3Model(model) {
 		// Don't set a default - let the API use its dynamic default ("high")
 		// Only set includeThoughts
 		updated, _ := sjson.SetBytes(body, "request.generationConfig.thinkingConfig.includeThoughts", true)

internal/util/thinking.go

@@ -12,9 +12,18 @@ func ModelSupportsThinking(model string) bool {
 	if model == "" {
 		return false
 	}
+	// First check the global dynamic registry
 	if info := registry.GetGlobalRegistry().GetModelInfo(model); info != nil {
 		return info.Thinking != nil
 	}
+	// Fallback: check static model definitions
+	if info := registry.LookupStaticModelInfo(model); info != nil {
+		return info.Thinking != nil
+	}
+	// Fallback: check Antigravity static config
+	if cfg := registry.GetAntigravityModelConfig()[model]; cfg != nil {
+		return cfg.Thinking != nil
+	}
 	return false
 }
 
@@ -63,11 +72,19 @@ func thinkingRangeFromRegistry(model string) (found bool, min int, max int, zero
 	if model == "" {
 		return false, 0, 0, false, false
 	}
-	info := registry.GetGlobalRegistry().GetModelInfo(model)
-	if info == nil || info.Thinking == nil {
-		return false, 0, 0, false, false
+	// First check global dynamic registry
+	if info := registry.GetGlobalRegistry().GetModelInfo(model); info != nil && info.Thinking != nil {
+		return true, info.Thinking.Min, info.Thinking.Max, info.Thinking.ZeroAllowed, info.Thinking.DynamicAllowed
 	}
-	return true, info.Thinking.Min, info.Thinking.Max, info.Thinking.ZeroAllowed, info.Thinking.DynamicAllowed
+	// Fallback: check static model definitions
+	if info := registry.LookupStaticModelInfo(model); info != nil && info.Thinking != nil {
+		return true, info.Thinking.Min, info.Thinking.Max, info.Thinking.ZeroAllowed, info.Thinking.DynamicAllowed
+	}
+	// Fallback: check Antigravity static config
+	if cfg := registry.GetAntigravityModelConfig()[model]; cfg != nil && cfg.Thinking != nil {
+		return true, cfg.Thinking.Min, cfg.Thinking.Max, cfg.Thinking.ZeroAllowed, cfg.Thinking.DynamicAllowed
+	}
+	return false, 0, 0, false, false
 }
 
 // GetModelThinkingLevels returns the discrete reasoning effort levels for the model.

sdk/api/management.go

@@ -20,6 +20,7 @@ type ManagementTokenRequester interface {
 	RequestQwenToken(*gin.Context)
 	RequestIFlowToken(*gin.Context)
 	RequestIFlowCookieToken(*gin.Context)
+	GetAuthStatus(c *gin.Context)
 }
 
 type managementTokenRequester struct {
@@ -60,3 +61,7 @@ func (m *managementTokenRequester) RequestIFlowToken(c *gin.Context) {
 func (m *managementTokenRequester) RequestIFlowCookieToken(c *gin.Context) {
 	m.handler.RequestIFlowCookieToken(c)
 }
+
+func (m *managementTokenRequester) GetAuthStatus(c *gin.Context) {
+	m.handler.GetAuthStatus(c)
+}

sdk/cliproxy/auth/model_name_mappings.go

@@ -81,7 +81,9 @@ func (m *Manager) applyOAuthModelMapping(auth *Auth, requestedModel string, meta
 			out[k] = v
 		}
 	}
-	out[util.ModelMappingOriginalModelMetadataKey] = upstreamModel
+	// Store the requested alias (e.g., "gp") so downstream can use it to look up
+	// model metadata from the global registry where it was registered under this alias.
+	out[util.ModelMappingOriginalModelMetadataKey] = requestedModel
 	return upstreamModel, out
 }
 

test/model_alias_thinking_suffix_test.go

@@ -0,0 +1,211 @@
+package test
+
+import (
+	"testing"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/runtime/executor"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
+	"github.com/tidwall/gjson"
+)
+
+// TestModelAliasThinkingSuffix tests the 32 test cases defined in docs/thinking_suffix_test_cases.md
+// These tests verify the thinking suffix parsing and application logic across different providers.
+func TestModelAliasThinkingSuffix(t *testing.T) {
+	tests := []struct {
+		id            int
+		name          string
+		provider      string
+		requestModel  string
+		suffixType    string
+		expectedField string // "thinkingBudget", "thinkingLevel", "budget_tokens", "reasoning_effort", "enable_thinking"
+		expectedValue any
+		upstreamModel string // The upstream model after alias resolution
+		isAlias       bool
+	}{
+		// === 1. Antigravity Provider ===
+		// 1.1 Budget-only models (Gemini 2.5)
+		{1, "antigravity_original_numeric", "antigravity", "gemini-2.5-computer-use-preview-10-2025(1000)", "numeric", "thinkingBudget", 1000, "gemini-2.5-computer-use-preview-10-2025", false},
+		{2, "antigravity_alias_numeric", "antigravity", "gp(1000)", "numeric", "thinkingBudget", 1000, "gemini-2.5-computer-use-preview-10-2025", true},
+		// 1.2 Budget+Levels models (Gemini 3)
+		{3, "antigravity_original_numeric_to_level", "antigravity", "gemini-3-flash-preview(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{4, "antigravity_original_level", "antigravity", "gemini-3-flash-preview(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{5, "antigravity_alias_numeric_to_level", "antigravity", "gf(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+		{6, "antigravity_alias_level", "antigravity", "gf(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+
+		// === 2. Gemini CLI Provider ===
+		// 2.1 Budget-only models
+		{7, "gemini_cli_original_numeric", "gemini-cli", "gemini-2.5-pro(8192)", "numeric", "thinkingBudget", 8192, "gemini-2.5-pro", false},
+		{8, "gemini_cli_alias_numeric", "gemini-cli", "g25p(8192)", "numeric", "thinkingBudget", 8192, "gemini-2.5-pro", true},
+		// 2.2 Budget+Levels models
+		{9, "gemini_cli_original_numeric_to_level", "gemini-cli", "gemini-3-flash-preview(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{10, "gemini_cli_original_level", "gemini-cli", "gemini-3-flash-preview(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{11, "gemini_cli_alias_numeric_to_level", "gemini-cli", "gf(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+		{12, "gemini_cli_alias_level", "gemini-cli", "gf(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+
+		// === 3. Vertex Provider ===
+		// 3.1 Budget-only models
+		{13, "vertex_original_numeric", "vertex", "gemini-2.5-pro(16384)", "numeric", "thinkingBudget", 16384, "gemini-2.5-pro", false},
+		{14, "vertex_alias_numeric", "vertex", "vg25p(16384)", "numeric", "thinkingBudget", 16384, "gemini-2.5-pro", true},
+		// 3.2 Budget+Levels models
+		{15, "vertex_original_numeric_to_level", "vertex", "gemini-3-flash-preview(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{16, "vertex_original_level", "vertex", "gemini-3-flash-preview(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{17, "vertex_alias_numeric_to_level", "vertex", "vgf(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+		{18, "vertex_alias_level", "vertex", "vgf(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+
+		// === 4. AI Studio Provider ===
+		// 4.1 Budget-only models
+		{19, "aistudio_original_numeric", "aistudio", "gemini-2.5-pro(12000)", "numeric", "thinkingBudget", 12000, "gemini-2.5-pro", false},
+		{20, "aistudio_alias_numeric", "aistudio", "ag25p(12000)", "numeric", "thinkingBudget", 12000, "gemini-2.5-pro", true},
+		// 4.2 Budget+Levels models
+		{21, "aistudio_original_numeric_to_level", "aistudio", "gemini-3-flash-preview(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{22, "aistudio_original_level", "aistudio", "gemini-3-flash-preview(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", false},
+		{23, "aistudio_alias_numeric_to_level", "aistudio", "agf(1000)", "numeric", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+		{24, "aistudio_alias_level", "aistudio", "agf(low)", "level", "thinkingLevel", "low", "gemini-3-flash-preview", true},
+
+		// === 5. Claude Provider ===
+		{25, "claude_original_numeric", "claude", "claude-sonnet-4-5-20250929(16384)", "numeric", "budget_tokens", 16384, "claude-sonnet-4-5-20250929", false},
+		{26, "claude_alias_numeric", "claude", "cs45(16384)", "numeric", "budget_tokens", 16384, "claude-sonnet-4-5-20250929", true},
+
+		// === 6. Codex Provider ===
+		{27, "codex_original_level", "codex", "gpt-5(high)", "level", "reasoning_effort", "high", "gpt-5", false},
+		{28, "codex_alias_level", "codex", "g5(high)", "level", "reasoning_effort", "high", "gpt-5", true},
+
+		// === 7. Qwen Provider ===
+		{29, "qwen_original_level", "qwen", "qwen3-coder-plus(high)", "level", "enable_thinking", true, "qwen3-coder-plus", false},
+		{30, "qwen_alias_level", "qwen", "qcp(high)", "level", "enable_thinking", true, "qwen3-coder-plus", true},
+
+		// === 8. iFlow Provider ===
+		{31, "iflow_original_level", "iflow", "glm-4.7(high)", "level", "reasoning_effort", "high", "glm-4.7", false},
+		{32, "iflow_alias_level", "iflow", "glm(high)", "level", "reasoning_effort", "high", "glm-4.7", true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// Step 1: Parse model suffix (simulates SDK layer normalization)
+			// For "gp(1000)" -> requestedModel="gp", metadata={thinking_budget: 1000}
+			requestedModel, metadata := util.NormalizeThinkingModel(tt.requestModel)
+
+			// Verify suffix was parsed
+			if metadata == nil && (tt.suffixType == "numeric" || tt.suffixType == "level") {
+				t.Errorf("Case #%d: NormalizeThinkingModel(%q) metadata is nil", tt.id, tt.requestModel)
+				return
+			}
+
+			// Step 2: Simulate OAuth model mapping
+			// Real flow: applyOAuthModelMapping stores requestedModel (the alias) in metadata
+			if tt.isAlias {
+				if metadata == nil {
+					metadata = make(map[string]any)
+				}
+				metadata[util.ModelMappingOriginalModelMetadataKey] = requestedModel
+			}
+
+			// Step 3: Verify metadata extraction
+			switch tt.suffixType {
+			case "numeric":
+				budget, _, _, matched := util.ThinkingFromMetadata(metadata)
+				if !matched {
+					t.Errorf("Case #%d: ThinkingFromMetadata did not match", tt.id)
+					return
+				}
+				if budget == nil {
+					t.Errorf("Case #%d: expected budget in metadata", tt.id)
+					return
+				}
+				// For thinkingBudget/budget_tokens, verify the parsed budget value
+				if tt.expectedField == "thinkingBudget" || tt.expectedField == "budget_tokens" {
+					expectedBudget := tt.expectedValue.(int)
+					if *budget != expectedBudget {
+						t.Errorf("Case #%d: budget = %d, want %d", tt.id, *budget, expectedBudget)
+					}
+				}
+				// For thinkingLevel (Gemini 3), verify conversion from budget to level
+				if tt.expectedField == "thinkingLevel" {
+					level, ok := util.ThinkingBudgetToGemini3Level(tt.upstreamModel, *budget)
+					if !ok {
+						t.Errorf("Case #%d: ThinkingBudgetToGemini3Level failed", tt.id)
+						return
+					}
+					expectedLevel := tt.expectedValue.(string)
+					if level != expectedLevel {
+						t.Errorf("Case #%d: converted level = %q, want %q", tt.id, level, expectedLevel)
+					}
+				}
+
+			case "level":
+				_, _, effort, matched := util.ThinkingFromMetadata(metadata)
+				if !matched {
+					t.Errorf("Case #%d: ThinkingFromMetadata did not match", tt.id)
+					return
+				}
+				if effort == nil {
+					t.Errorf("Case #%d: expected effort in metadata", tt.id)
+					return
+				}
+				if tt.expectedField == "thinkingLevel" || tt.expectedField == "reasoning_effort" {
+					expectedEffort := tt.expectedValue.(string)
+					if *effort != expectedEffort {
+						t.Errorf("Case #%d: effort = %q, want %q", tt.id, *effort, expectedEffort)
+					}
+				}
+			}
+
+			// Step 4: Test Gemini-specific thinkingLevel conversion for Gemini 3 models
+			if tt.expectedField == "thinkingLevel" && util.IsGemini3Model(tt.upstreamModel) {
+				body := []byte(`{"request":{"contents":[]}}`)
+
+				// Build metadata simulating real OAuth flow:
+				// - requestedModel (alias like "gf") is stored in model_mapping_original_model
+				// - upstreamModel is passed as the model parameter
+				testMetadata := make(map[string]any)
+				if tt.isAlias {
+					// Real flow: applyOAuthModelMapping stores requestedModel (the alias)
+					testMetadata[util.ModelMappingOriginalModelMetadataKey] = requestedModel
+				}
+				// Copy parsed metadata (thinking_budget, reasoning_effort, etc.)
+				for k, v := range metadata {
+					testMetadata[k] = v
+				}
+
+				result := util.ApplyGemini3ThinkingLevelFromMetadataCLI(tt.upstreamModel, testMetadata, body)
+				levelVal := gjson.GetBytes(result, "request.generationConfig.thinkingConfig.thinkingLevel")
+
+				expectedLevel := tt.expectedValue.(string)
+				if !levelVal.Exists() {
+					t.Errorf("Case #%d: expected thinkingLevel in result", tt.id)
+				} else if levelVal.String() != expectedLevel {
+					t.Errorf("Case #%d: thinkingLevel = %q, want %q", tt.id, levelVal.String(), expectedLevel)
+				}
+			}
+
+			// Step 5: Test Gemini 2.5 thinkingBudget application using real ApplyThinkingMetadataCLI flow
+			if tt.expectedField == "thinkingBudget" && util.IsGemini25Model(tt.upstreamModel) {
+				body := []byte(`{"request":{"contents":[]}}`)
+
+				// Build metadata simulating real OAuth flow:
+				// - requestedModel (alias like "gp") is stored in model_mapping_original_model
+				// - upstreamModel is passed as the model parameter
+				testMetadata := make(map[string]any)
+				if tt.isAlias {
+					// Real flow: applyOAuthModelMapping stores requestedModel (the alias)
+					testMetadata[util.ModelMappingOriginalModelMetadataKey] = requestedModel
+				}
+				// Copy parsed metadata (thinking_budget, reasoning_effort, etc.)
+				for k, v := range metadata {
+					testMetadata[k] = v
+				}
+
+				// Use the exported ApplyThinkingMetadataCLI which includes the fallback logic
+				result := executor.ApplyThinkingMetadataCLI(body, testMetadata, tt.upstreamModel)
+				budgetVal := gjson.GetBytes(result, "request.generationConfig.thinkingConfig.thinkingBudget")
+
+				expectedBudget := tt.expectedValue.(int)
+				if !budgetVal.Exists() {
+					t.Errorf("Case #%d: expected thinkingBudget in result", tt.id)
+				} else if int(budgetVal.Int()) != expectedBudget {
+					t.Errorf("Case #%d: thinkingBudget = %d, want %d", tt.id, int(budgetVal.Int()), expectedBudget)
+				}
+			}
+		})
+	}
+}