fix(groups): merge main and resolve model allowlist conflicts

2026-10-30 08:22:34 +00:00 · 2026-09-07 18:33:39 +08:00
parent 55c5eed9fc 909b8c7ef4
commit 81ff9384cb
252 changed files with 11371 additions and 1598 deletions
@@ -17,6 +17,7 @@ import (
 	"sync"
 	"time"

+	"github.com/Wei-Shaw/sub2api/internal/config"
 	"github.com/Wei-Shaw/sub2api/internal/domain"
 	"github.com/Wei-Shaw/sub2api/internal/handler/dto"
 	"github.com/Wei-Shaw/sub2api/internal/pkg/antigravity"
@@ -64,6 +65,7 @@ type AccountHandler struct {
 	grokImportProber        grokImportProber
 	upstreamBillingProbe    *service.UpstreamBillingProbeService
 	ollamaCloudUsage        *service.OllamaCloudUsageService
+	cfg                     *config.Config
 }

 // SetUpstreamBillingProbeService attaches the optional remote billing probe service.
@@ -190,6 +192,7 @@ type CheckMixedChannelRequest struct {
 // AccountWithConcurrency extends Account with real-time concurrency info
 type AccountWithConcurrency struct {
 	*dto.Account
+	simpleMode         bool                         `json:"-"`
 	CurrentConcurrency int                          `json:"current_concurrency"`
 	SchedulerScore     *AccountSchedulerScore       `json:"scheduler_score,omitempty"`
 	SchedulerScores    []AccountSchedulerGroupScore `json:"scheduler_scores,omitempty"`
@@ -199,6 +202,123 @@ type AccountWithConcurrency struct {
 	CurrentRPM        *int     `json:"current_rpm,omitempty"`         // 当前分钟 RPM 计数
 }

+// AccountListItemWithConcurrency is the compact account-list envelope used
+// for lite=1. It embeds dto.AccountListItem instead of the full dto.Account,
+// so groups/account_groups never appear in the list payload.
+type AccountListItemWithConcurrency struct {
+	*dto.AccountListItem
+	CurrentConcurrency int                          `json:"current_concurrency"`
+	SchedulerScore     *AccountSchedulerScore       `json:"scheduler_score,omitempty"`
+	SchedulerScores    []AccountSchedulerGroupScore `json:"scheduler_scores,omitempty"`
+	CurrentWindowCost  *float64                     `json:"current_window_cost,omitempty"`
+	ActiveSessions     *int                         `json:"active_sessions,omitempty"`
+	CurrentRPM         *int                         `json:"current_rpm,omitempty"`
+}
+
+type simpleModeGroupReference struct {
+	ID       int64  `json:"id"`
+	Name     string `json:"name"`
+	Platform string `json:"platform"`
+	Status   string `json:"status"`
+}
+
+type simpleModeAccountGroupReference struct {
+	AccountID int64                     `json:"account_id"`
+	GroupID   int64                     `json:"group_id"`
+	Priority  int                       `json:"priority"`
+	CreatedAt time.Time                 `json:"created_at"`
+	Group     *simpleModeGroupReference `json:"group,omitempty"`
+}
+
+func simpleModeGroupReferenceFromDTO(group *dto.Group) *simpleModeGroupReference {
+	if group == nil {
+		return nil
+	}
+	return &simpleModeGroupReference{ID: group.ID, Name: group.Name, Platform: group.Platform, Status: group.Status}
+}
+
+func simpleModeCompositeGroupIDs(account *dto.Account) map[int64]struct{} {
+	hidden := make(map[int64]struct{})
+	if account == nil {
+		return hidden
+	}
+	for _, group := range account.Groups {
+		if group != nil && group.Platform == service.PlatformComposite {
+			hidden[group.ID] = struct{}{}
+		}
+	}
+	for _, accountGroup := range account.AccountGroups {
+		if accountGroup.Group != nil && accountGroup.Group.Platform == service.PlatformComposite {
+			hidden[accountGroup.GroupID] = struct{}{}
+		}
+	}
+	return hidden
+}
+
+func filterSimpleModeGroupIDs(groupIDs []int64, hidden map[int64]struct{}) []int64 {
+	visible := make([]int64, 0, len(groupIDs))
+	for _, groupID := range groupIDs {
+		if _, ok := hidden[groupID]; !ok {
+			visible = append(visible, groupID)
+		}
+	}
+	return visible
+}
+
+func simpleModeCompositeServiceGroupIDs(account *service.Account) map[int64]struct{} {
+	hidden := make(map[int64]struct{})
+	if account == nil {
+		return hidden
+	}
+	for _, group := range account.Groups {
+		if group != nil && group.Platform == service.PlatformComposite {
+			hidden[group.ID] = struct{}{}
+		}
+	}
+	for _, accountGroup := range account.AccountGroups {
+		if accountGroup.Group != nil && accountGroup.Group.Platform == service.PlatformComposite {
+			hidden[accountGroup.GroupID] = struct{}{}
+		}
+	}
+	return hidden
+}
+
+func (a AccountWithConcurrency) MarshalJSON() ([]byte, error) {
+	type alias AccountWithConcurrency
+	if !a.simpleMode || a.Account == nil {
+		return json.Marshal(alias(a))
+	}
+	groups := make([]simpleModeGroupReference, 0, len(a.Groups))
+	compositeIDs := simpleModeCompositeGroupIDs(a.Account)
+	for _, group := range a.Groups {
+		if group != nil && group.Platform == service.PlatformComposite {
+			continue
+		}
+		if ref := simpleModeGroupReferenceFromDTO(group); ref != nil {
+			groups = append(groups, *ref)
+		}
+	}
+	accountGroups := make([]simpleModeAccountGroupReference, 0, len(a.AccountGroups))
+	for _, accountGroup := range a.AccountGroups {
+		if accountGroup.Group != nil && accountGroup.Group.Platform == service.PlatformComposite {
+			continue
+		}
+		if _, hidden := compositeIDs[accountGroup.GroupID]; hidden {
+			continue
+		}
+		accountGroups = append(accountGroups, simpleModeAccountGroupReference{
+			AccountID: accountGroup.AccountID, GroupID: accountGroup.GroupID, Priority: accountGroup.Priority,
+			CreatedAt: accountGroup.CreatedAt, Group: simpleModeGroupReferenceFromDTO(accountGroup.Group),
+		})
+	}
+	return json.Marshal(struct {
+		alias
+		GroupIDs      []int64                           `json:"group_ids,omitempty"`
+		Groups        []simpleModeGroupReference        `json:"groups"`
+		AccountGroups []simpleModeAccountGroupReference `json:"account_groups"`
+	}{alias: alias(a), GroupIDs: filterSimpleModeGroupIDs(a.GroupIDs, compositeIDs), Groups: groups, AccountGroups: accountGroups})
+}
+
 type AccountSchedulerScore struct {
 	BaseScore             float64 `json:"base_score"`
 	StickyScore           float64 `json:"sticky_score"`
@@ -223,9 +343,25 @@ func (h *AccountHandler) accountResponseFromService(account *service.Account) *d
 	return out
 }

+func (h *AccountHandler) accountListResponseFromService(account *service.Account) *dto.Account {
+	out := dto.AccountFromServiceShallow(account)
+	if out != nil && account != nil {
+		out.Proxy = dto.ProxyFromService(account.Proxy)
+	}
+	if h != nil && h.ollamaCloudUsage != nil && out != nil {
+		h.ollamaCloudUsage.EnrichState(out.OllamaCloudUsage)
+	}
+	return out
+}
+
+func (h *AccountHandler) isSimpleMode() bool {
+	return h != nil && h.cfg != nil && h.cfg.RunMode == config.RunModeSimple
+}
+
 func (h *AccountHandler) buildAccountResponseWithRuntime(ctx context.Context, account *service.Account) AccountWithConcurrency {
 	item := AccountWithConcurrency{
 		Account:            h.accountResponseFromService(account),
+		simpleMode:         h.isSimpleMode(),
 		CurrentConcurrency: 0,
 	}
 	if account == nil {
@@ -650,8 +786,16 @@ func (h *AccountHandler) List(c *gin.Context) {
 	result := make([]AccountWithConcurrency, len(accounts))
 	for i := range accounts {
 		acc := &accounts[i]
+		accountResponse := h.accountResponseFromService(acc)
+		if lite {
+			accountResponse = h.accountListResponseFromService(acc)
+			if h.isSimpleMode() {
+				accountResponse.GroupIDs = filterSimpleModeGroupIDs(accountResponse.GroupIDs, simpleModeCompositeServiceGroupIDs(acc))
+			}
+		}
 		item := AccountWithConcurrency{
-			Account:            h.accountResponseFromService(acc),
+			Account:            accountResponse,
+			simpleMode:         h.isSimpleMode(),
 			CurrentConcurrency: concurrencyCounts[acc.ID],
 			SchedulerScore:     schedulerScores[acc.ID],
 			SchedulerScores:    schedulerGroupScores[acc.ID],
@@ -683,7 +827,34 @@ func (h *AccountHandler) List(c *gin.Context) {

 	h.enrichShadowParents(c.Request.Context(), result)

-	etag := buildAccountsListETag(result, total, page, pageSize, platform, accountType, status, search, lite)
+	if lite {
+		compact := make([]AccountListItemWithConcurrency, len(result))
+		for i := range result {
+			item := result[i]
+			compact[i] = AccountListItemWithConcurrency{
+				AccountListItem:    dto.AccountListItemFromAccount(item.Account),
+				CurrentConcurrency: item.CurrentConcurrency,
+				SchedulerScore:     item.SchedulerScore,
+				SchedulerScores:    item.SchedulerScores,
+				CurrentWindowCost:  item.CurrentWindowCost,
+				ActiveSessions:     item.ActiveSessions,
+				CurrentRPM:         item.CurrentRPM,
+			}
+		}
+		etag := buildAccountsListETag(compact, total, page, pageSize, platform, accountType, status, search, true)
+		if etag != "" {
+			c.Header("ETag", etag)
+			c.Header("Vary", "If-None-Match")
+			if ifNoneMatchMatched(c.GetHeader("If-None-Match"), etag) {
+				c.Status(http.StatusNotModified)
+				return
+			}
+		}
+		response.Paginated(c, compact, total, page, pageSize)
+		return
+	}
+
+	etag := buildAccountsListETag(result, total, page, pageSize, platform, accountType, status, search, false)
 	if etag != "" {
 		c.Header("ETag", etag)
 		c.Header("Vary", "If-None-Match")
@@ -696,23 +867,23 @@ func (h *AccountHandler) List(c *gin.Context) {
 	response.Paginated(c, result, total, page, pageSize)
 }

-func buildAccountsListETag(
-	items []AccountWithConcurrency,
+func buildAccountsListETag[T any](
+	items []T,
 	total int64,
 	page, pageSize int,
 	platform, accountType, status, search string,
 	lite bool,
 ) string {
 	payload := struct {
-		Total       int64                    `json:"total"`
-		Page        int                      `json:"page"`
-		PageSize    int                      `json:"page_size"`
-		Platform    string                   `json:"platform"`
-		AccountType string                   `json:"type"`
-		Status      string                   `json:"status"`
-		Search      string                   `json:"search"`
-		Lite        bool                     `json:"lite"`
-		Items       []AccountWithConcurrency `json:"items"`
+		Total       int64  `json:"total"`
+		Page        int    `json:"page"`
+		PageSize    int    `json:"page_size"`
+		Platform    string `json:"platform"`
+		AccountType string `json:"type"`
+		Status      string `json:"status"`
+		Search      string `json:"search"`
+		Lite        bool   `json:"lite"`
+		Items       []T    `json:"items"`
 	}{
 		Total:       total,
 		Page:        page,
@@ -1887,6 +2058,14 @@ func (h *AccountHandler) BatchCreate(c *gin.Context) {
 			return
 		}
 	}
+	groupIDs := make([]int64, 0)
+	for _, item := range req.Accounts {
+		groupIDs = append(groupIDs, item.GroupIDs...)
+	}
+	if err := h.adminService.ValidateAccountGroupBindings(c.Request.Context(), groupIDs); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}

 	executeAdminIdempotentJSON(c, "admin.accounts.batch_create", req, service.DefaultWriteIdempotencyTTL(), func(ctx context.Context) (any, error) {
 		success := 0
@@ -2601,6 +2780,14 @@ func (h *AccountHandler) GetAvailableModels(c *gin.Context) {

 	// Handle OpenAI accounts
 	if account.IsOpenAI() {
+		// Prefer the shared, account-keyed upstream catalog. If discovery fails,
+		// retain the legacy local catalog below so the test dialog remains usable.
+		if h.accountTestService != nil {
+			if models, fetchErr := h.accountTestService.FetchOpenAIAccountModels(c.Request.Context(), account); fetchErr == nil {
+				response.Success(c, models)
+				return
+			}
+		}
 		// OpenAI 自动透传会绕过常规模型改写，测试/模型列表也应回落到默认模型集。
 		if account.IsOpenAIPassthroughEnabled() {
 			response.Success(c, openai.DefaultModels)
@@ -1,9 +1,12 @@
 package admin

 import (
+	"bytes"
+	"compress/gzip"
 	"encoding/json"
 	"net/http"
 	"net/http/httptest"
+	"strconv"
 	"strings"
 	"testing"
 	"time"
@@ -13,6 +16,101 @@ import (
 	"github.com/stretchr/testify/require"
 )

+func TestAccountHandlerListLiteUsesCompactDTOAndETag(t *testing.T) {
+	router, adminSvc := setupAccountListRouter()
+	now := time.Now().UTC()
+	groupID := int64(77)
+	adminSvc.accounts = []service.Account{{
+		ID: 501, Name: "compact-account", Platform: service.PlatformOpenAI, Type: service.AccountTypeOAuth,
+		Credentials: map[string]any{"email": "compact@example.com", "access_token": strings.Repeat("x", 4096)},
+		Extra:       map[string]any{"privacy_mode": "training_off"}, Status: service.StatusActive,
+		Schedulable: true, Concurrency: 4, GroupIDs: []int64{groupID},
+		Groups:        []*service.Group{{ID: groupID, Name: "codex", Platform: service.PlatformOpenAI}},
+		AccountGroups: []service.AccountGroup{{AccountID: 501, GroupID: groupID, Priority: 2, Group: &service.Group{ID: groupID, Name: "codex", Platform: service.PlatformOpenAI}}},
+		CreatedAt:     now, UpdatedAt: now,
+	}}
+
+	rec := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodGet, "/api/v1/admin/accounts?page=1&page_size=20&lite=1", nil)
+	router.ServeHTTP(rec, req)
+	require.Equal(t, http.StatusOK, rec.Code)
+	require.NotEmpty(t, rec.Header().Get("ETag"))
+
+	var litePayload struct {
+		Data struct {
+			Items []map[string]any `json:"items"`
+		} `json:"data"`
+	}
+	require.NoError(t, json.Unmarshal(rec.Body.Bytes(), &litePayload))
+	require.Len(t, litePayload.Data.Items, 1)
+	liteItem := litePayload.Data.Items[0]
+	require.Equal(t, float64(501), liteItem["id"])
+	require.Equal(t, []any{float64(groupID)}, liteItem["group_ids"])
+	require.Equal(t, true, liteItem["schedulable"])
+	require.NotContains(t, liteItem, "groups")
+	require.NotContains(t, liteItem, "account_groups")
+	credentials, ok := liteItem["credentials"].(map[string]any)
+	require.True(t, ok)
+	require.Equal(t, "compact@example.com", credentials["email"])
+	require.NotContains(t, credentials, "access_token")
+	credentialsStatus, ok := liteItem["credentials_status"].(map[string]any)
+	require.True(t, ok)
+	require.Equal(t, true, credentialsStatus["has_access_token"])
+
+	// The ETag must represent the same compact body and return 304 on refresh.
+	rec304 := httptest.NewRecorder()
+	req304 := httptest.NewRequest(http.MethodGet, "/api/v1/admin/accounts?page=1&page_size=20&lite=1", nil)
+	req304.Header.Set("If-None-Match", rec.Header().Get("ETag"))
+	router.ServeHTTP(rec304, req304)
+	require.Equal(t, http.StatusNotModified, rec304.Code)
+
+	// Omitting lite preserves the legacy full response shape.
+	recFull := httptest.NewRecorder()
+	reqFull := httptest.NewRequest(http.MethodGet, "/api/v1/admin/accounts?page=1&page_size=20", nil)
+	router.ServeHTTP(recFull, reqFull)
+	require.Equal(t, http.StatusOK, recFull.Code)
+	var fullPayload struct {
+		Data struct {
+			Items []map[string]any `json:"items"`
+		} `json:"data"`
+	}
+	require.NoError(t, json.Unmarshal(recFull.Body.Bytes(), &fullPayload))
+	require.Contains(t, fullPayload.Data.Items[0], "groups")
+	require.Contains(t, fullPayload.Data.Items[0], "account_groups")
+}
+
+func TestAccountHandlerListLiteStaysBelowResponseBudget(t *testing.T) {
+	router, adminSvc := setupAccountListRouter()
+	now := time.Now().UTC()
+	accounts := make([]service.Account, 20)
+	for i := range accounts {
+		id := int64(600 + i)
+		groupID := int64(800 + i)
+		accounts[i] = service.Account{
+			ID: id, Name: "account-" + strconv.Itoa(i), Platform: service.PlatformOpenAI,
+			Type: service.AccountTypeOAuth, Status: service.StatusActive, Schedulable: true,
+			Concurrency: 4, GroupIDs: []int64{groupID},
+			Groups:        []*service.Group{{ID: groupID, Name: "group-" + strconv.Itoa(i), Description: strings.Repeat("description ", 20)}},
+			AccountGroups: []service.AccountGroup{{AccountID: id, GroupID: groupID, Group: &service.Group{ID: groupID, Name: "group-" + strconv.Itoa(i), Description: strings.Repeat("description ", 20)}}},
+			CreatedAt:     now, UpdatedAt: now,
+		}
+	}
+	adminSvc.accounts = accounts
+
+	rec := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodGet, "/api/v1/admin/accounts?page=1&page_size=20&lite=1", nil)
+	router.ServeHTTP(rec, req)
+	require.Equal(t, http.StatusOK, rec.Code)
+	require.Less(t, rec.Body.Len(), 80*1024)
+
+	var compressed bytes.Buffer
+	zw := gzip.NewWriter(&compressed)
+	_, err := zw.Write(rec.Body.Bytes())
+	require.NoError(t, err)
+	require.NoError(t, zw.Close())
+	require.Less(t, compressed.Len(), 15*1024)
+}
+
 func setupAccountListRouter() (*gin.Engine, *stubAdminService) {
 	gin.SetMode(gin.TestMode)
 	router := gin.New()
@@ -0,0 +1,240 @@
+package admin
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/Wei-Shaw/sub2api/internal/config"
+	infraerrors "github.com/Wei-Shaw/sub2api/internal/pkg/errors"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+type simpleModeAccountService struct {
+	*stubAdminService
+	account     service.Account
+	createCalls int
+	updateCalls int
+	bulkCalls   int
+}
+
+func (s *simpleModeAccountService) GetAccount(context.Context, int64) (*service.Account, error) {
+	return &s.account, nil
+}
+
+func (s *simpleModeAccountService) CreateAccount(ctx context.Context, input *service.CreateAccountInput) (*service.Account, error) {
+	if err := s.ValidateAccountGroupBindings(ctx, input.GroupIDs); err != nil {
+		return nil, err
+	}
+	s.createCalls++
+	return &s.account, nil
+}
+
+func (s *simpleModeAccountService) UpdateAccount(ctx context.Context, _ int64, input *service.UpdateAccountInput) (*service.Account, error) {
+	if input.GroupIDs != nil {
+		if err := s.ValidateAccountGroupBindings(ctx, *input.GroupIDs); err != nil {
+			return nil, err
+		}
+	}
+	s.updateCalls++
+	return &s.account, nil
+}
+
+func (s *simpleModeAccountService) BulkUpdateAccounts(ctx context.Context, input *service.BulkUpdateAccountsInput) (*service.BulkUpdateAccountsResult, error) {
+	if input.GroupIDs != nil {
+		if err := s.ValidateAccountGroupBindings(ctx, *input.GroupIDs); err != nil {
+			return nil, err
+		}
+	}
+	s.bulkCalls++
+	return &service.BulkUpdateAccountsResult{}, nil
+}
+
+func (s *simpleModeAccountService) ValidateAccountGroupBindings(_ context.Context, groupIDs []int64) error {
+	for _, id := range groupIDs {
+		for i := range s.groups {
+			if s.groups[i].ID == id && s.groups[i].Platform == service.PlatformComposite {
+				return infraerrors.BadRequest("SIMPLE_MODE_GROUP_NOT_BINDABLE", "composite groups cannot be bound in simple mode")
+			}
+		}
+	}
+	return nil
+}
+
+func TestAccountHandlerSimpleModeUsesMinimalGroupReferences(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	richGroup := &service.Group{ID: 7, Name: "basic", Platform: service.PlatformAnthropic, Status: service.StatusActive, RateMultiplier: 9, IsExclusive: true, SubscriptionType: service.SubscriptionTypeSubscription, RPMLimit: 99}
+	historicalComposite := &service.Group{ID: 9, Name: "historical composite", Platform: service.PlatformComposite, Status: service.StatusActive}
+	account := service.Account{
+		ID: 3, Name: "account", Platform: service.PlatformAnthropic, Type: service.AccountTypeAPIKey, Status: service.StatusActive,
+		GroupIDs: []int64{7, 9},
+		Groups:   []*service.Group{richGroup, historicalComposite},
+		AccountGroups: []service.AccountGroup{
+			{AccountID: 3, GroupID: 7, Priority: 2, Group: richGroup, Account: &service.Account{ID: 3, Extra: map[string]any{"secret": true}}},
+			{AccountID: 3, GroupID: 9, Priority: 3, Group: historicalComposite},
+		},
+	}
+	svc := &simpleModeAccountService{stubAdminService: newStubAdminService(), account: account}
+	svc.accounts = []service.Account{account}
+	h := NewAccountHandler(svc, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+	h.cfg = &config.Config{RunMode: config.RunModeSimple}
+	r := gin.New()
+	r.GET("/accounts", h.List)
+	r.GET("/accounts/:id", h.GetByID)
+	r.POST("/accounts", h.Create)
+	r.PUT("/accounts/:id", h.Update)
+
+	tests := []struct {
+		method string
+		path   string
+		body   string
+	}{
+		{http.MethodGet, "/accounts", ""},
+		{http.MethodGet, "/accounts/3", ""},
+		{http.MethodPost, "/accounts", `{"name":"account","platform":"anthropic","type":"apikey","credentials":{"key":"x"}}`},
+		{http.MethodPut, "/accounts/3", `{"name":"account"}`},
+	}
+	for _, tt := range tests {
+		t.Run(tt.method+tt.path, func(t *testing.T) {
+			req := httptest.NewRequest(tt.method, tt.path, bytes.NewBufferString(tt.body))
+			req.Header.Set("Content-Type", "application/json")
+			res := httptest.NewRecorder()
+			r.ServeHTTP(res, req)
+			require.Equal(t, http.StatusOK, res.Code, res.Body.String())
+			var payload map[string]any
+			require.NoError(t, json.Unmarshal(res.Body.Bytes(), &payload))
+			data, ok := payload["data"].(map[string]any)
+			require.True(t, ok)
+			if items, ok := data["items"].([]any); ok {
+				require.Len(t, items, 1)
+				data, ok = items[0].(map[string]any)
+				require.True(t, ok)
+			}
+			groups, ok := data["groups"].([]any)
+			require.True(t, ok)
+			require.Len(t, groups, 1)
+			require.Equal(t, []any{float64(7)}, data["group_ids"])
+			group, ok := groups[0].(map[string]any)
+			require.True(t, ok)
+			require.ElementsMatch(t, []string{"id", "name", "platform", "status"}, mapKeys(group))
+			accountGroups, ok := data["account_groups"].([]any)
+			require.True(t, ok)
+			require.Len(t, accountGroups, 1)
+			accountGroup, ok := accountGroups[0].(map[string]any)
+			require.True(t, ok)
+			require.ElementsMatch(t, []string{"account_id", "group_id", "priority", "created_at", "group"}, mapKeys(accountGroup))
+			nestedGroup, ok := accountGroup["group"].(map[string]any)
+			require.True(t, ok)
+			require.ElementsMatch(t, []string{"id", "name", "platform", "status"}, mapKeys(nestedGroup))
+		})
+	}
+}
+
+func TestAccountHandlerSimpleModeLitePreservesCompactShapeAndETag(t *testing.T) {
+	richGroup := &service.Group{ID: 7, Name: "basic", Platform: service.PlatformAnthropic, Status: service.StatusActive, RateMultiplier: 9}
+	historicalComposite := &service.Group{ID: 9, Name: "historical composite", Platform: service.PlatformComposite, Status: service.StatusActive}
+	account := service.Account{
+		ID: 3, Name: "account", Platform: service.PlatformAnthropic, Type: service.AccountTypeAPIKey, Status: service.StatusActive,
+		GroupIDs: []int64{7, 9}, Groups: []*service.Group{richGroup, historicalComposite},
+		AccountGroups: []service.AccountGroup{{AccountID: 3, GroupID: 7, Group: richGroup}, {AccountID: 3, GroupID: 9, Group: historicalComposite}},
+	}
+	svc := &simpleModeAccountService{stubAdminService: newStubAdminService(), account: account}
+	svc.accounts = []service.Account{account}
+	h := NewAccountHandler(svc, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+	h.cfg = &config.Config{RunMode: config.RunModeSimple}
+	r := gin.New()
+	r.GET("/accounts", h.List)
+
+	res := httptest.NewRecorder()
+	r.ServeHTTP(res, httptest.NewRequest(http.MethodGet, "/accounts?lite=1", nil))
+	require.Equal(t, http.StatusOK, res.Code, res.Body.String())
+	require.NotEmpty(t, res.Header().Get("ETag"))
+	var payload struct {
+		Data struct {
+			Items []map[string]any `json:"items"`
+		} `json:"data"`
+	}
+	require.NoError(t, json.Unmarshal(res.Body.Bytes(), &payload))
+	require.Len(t, payload.Data.Items, 1)
+	require.Equal(t, []any{float64(7)}, payload.Data.Items[0]["group_ids"])
+	require.NotContains(t, payload.Data.Items[0], "groups")
+	require.NotContains(t, payload.Data.Items[0], "account_groups")
+
+	notModified := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodGet, "/accounts?lite=1", nil)
+	req.Header.Set("If-None-Match", res.Header().Get("ETag"))
+	r.ServeHTTP(notModified, req)
+	require.Equal(t, http.StatusNotModified, notModified.Code)
+}
+
+func TestAccountHandlerSimpleModeRejectsCompositeGroupBindingsBeforeWrites(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	for _, tt := range []struct{ name, method, path, body string }{
+		{"create", http.MethodPost, "/accounts", `{"name":"account","platform":"anthropic","type":"apikey","credentials":{"key":"x"},"group_ids":[9]}`},
+		{"update", http.MethodPut, "/accounts/3", `{"group_ids":[9]}`},
+		{"bulk", http.MethodPost, "/accounts/bulk-update", `{"account_ids":[3],"group_ids":[9]}`},
+	} {
+		t.Run(tt.name, func(t *testing.T) {
+			svc := &simpleModeAccountService{stubAdminService: newStubAdminService()}
+			svc.groups = []service.Group{{ID: 7, Platform: service.PlatformAnthropic}, {ID: 9, Platform: service.PlatformComposite}}
+			h := NewAccountHandler(svc, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+			h.cfg = &config.Config{RunMode: config.RunModeSimple}
+			r := gin.New()
+			r.POST("/accounts", h.Create)
+			r.PUT("/accounts/:id", h.Update)
+			r.POST("/accounts/bulk-update", h.BulkUpdate)
+			res := httptest.NewRecorder()
+			req := httptest.NewRequest(tt.method, tt.path, bytes.NewBufferString(tt.body))
+			req.Header.Set("Content-Type", "application/json")
+			r.ServeHTTP(res, req)
+			require.Equal(t, http.StatusBadRequest, res.Code, res.Body.String())
+			require.Zero(t, svc.createCalls)
+			require.Zero(t, svc.updateCalls)
+			require.Zero(t, svc.bulkCalls)
+		})
+	}
+}
+
+func TestAccountHandlerSimpleModePreservesBasicGroupBinding(t *testing.T) {
+	svc := &simpleModeAccountService{stubAdminService: newStubAdminService(), account: service.Account{ID: 3}}
+	svc.groups = []service.Group{{ID: 7, Platform: service.PlatformAnthropic}}
+	h := NewAccountHandler(svc, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+	h.cfg = &config.Config{RunMode: config.RunModeSimple}
+	r := gin.New()
+	r.POST("/accounts", h.Create)
+	res := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodPost, "/accounts", bytes.NewBufferString(`{"name":"account","platform":"anthropic","type":"apikey","credentials":{"key":"x"},"group_ids":[7]}`))
+	req.Header.Set("Content-Type", "application/json")
+	r.ServeHTTP(res, req)
+	require.Equal(t, http.StatusOK, res.Code, res.Body.String())
+	require.Equal(t, 1, svc.createCalls)
+}
+
+func TestAccountHandlerSimpleModeBatchPrevalidatesAllGroupsAtomically(t *testing.T) {
+	svc := &simpleModeAccountService{stubAdminService: newStubAdminService()}
+	svc.groups = []service.Group{{ID: 7, Platform: service.PlatformAnthropic}, {ID: 9, Platform: service.PlatformComposite}}
+	h := NewAccountHandler(svc, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+	h.cfg = &config.Config{RunMode: config.RunModeSimple}
+	r := gin.New()
+	r.POST("/accounts/batch", h.BatchCreate)
+	res := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodPost, "/accounts/batch", bytes.NewBufferString(`{"accounts":[{"name":"first","platform":"anthropic","type":"apikey","credentials":{"key":"x"},"group_ids":[7]},{"name":"second","platform":"anthropic","type":"apikey","credentials":{"key":"y"},"group_ids":[9]}]}`))
+	req.Header.Set("Content-Type", "application/json")
+	r.ServeHTTP(res, req)
+	require.Equal(t, http.StatusBadRequest, res.Code, res.Body.String())
+	require.Zero(t, svc.createCalls)
+}
+
+func TestAccountHandlerAdvancedModeKeepsFullGroupReferences(t *testing.T) {
+	group := &service.Group{ID: 7, Name: "advanced", RateMultiplier: 9, SubscriptionType: service.SubscriptionTypeSubscription}
+	h := NewAccountHandler(nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil)
+	raw, err := json.Marshal(h.buildAccountResponseWithRuntime(context.Background(), &service.Account{Groups: []*service.Group{group}}))
+	require.NoError(t, err)
+	require.Contains(t, string(raw), `"rate_multiplier":9`)
+	require.Contains(t, string(raw), `"subscription_type":"subscription"`)
+}
@@ -278,13 +278,15 @@ func TestOpenAIFastPolicySettingsFromDTO_NormalizesServiceTier(t *testing.T) {
 			Rules: []dto.OpenAIFastPolicyRule{
 				{ServiceTier: "priority", Action: "filter", Scope: "all"},
 				{ServiceTier: "flex", Action: "block", Scope: "oauth"},
+				{ServiceTier: "ultrafast", Action: "pass", Scope: "all"},
 				{ServiceTier: "all", Action: "pass", Scope: "apikey"},
 			},
 		}
 		out := openaiFastPolicySettingsFromDTO(in)
-		require.Len(t, out.Rules, 3)
+		require.Len(t, out.Rules, 4)
 		require.Equal(t, service.OpenAIFastTierPriority, out.Rules[0].ServiceTier)
 		require.Equal(t, service.OpenAIFastTierFlex, out.Rules[1].ServiceTier)
-		require.Equal(t, service.OpenAIFastTierAny, out.Rules[2].ServiceTier)
+		require.Equal(t, service.OpenAIFastTierUltrafast, out.Rules[2].ServiceTier)
+		require.Equal(t, service.OpenAIFastTierAny, out.Rules[3].ServiceTier)
 	})
 }
@@ -24,6 +24,13 @@ type stubAdminService struct {
 	boundAuthIdentity                   *service.AdminBindAuthIdentityInput
 	boundAuthIdentityFor                int64
 	createdAccounts                     []*service.CreateAccountInput
+	createdGroups                       []*service.CreateGroupInput
+	updatedGroups                       []*service.UpdateGroupInput
+	deletedGroupIDs                     []int64
+	guardedDeletedGroupIDs              []int64
+	deleteGroupIfEmptyErr               error
+	advancedGroupOperationCalls         int
+	lastListGroupsIsExclusive           *bool
 	createdProxies                      []*service.CreateProxyInput
 	updatedProxyIDs                     []int64
 	updatedProxies                      []*service.UpdateProxyInput
@@ -267,6 +274,7 @@ func (s *stubAdminService) BindUserAuthIdentity(ctx context.Context, userID int6
 }

 func (s *stubAdminService) ListGroups(ctx context.Context, page, pageSize int, platform, status, search string, isExclusive *bool, sortBy, sortOrder string) ([]service.Group, int64, error) {
+	s.lastListGroupsIsExclusive = isExclusive
 	return s.groups, int64(len(s.groups)), nil
 }

@@ -295,6 +303,7 @@ func (s *stubAdminService) GetGroupModelsListCandidates(ctx context.Context, id
 }

 func (s *stubAdminService) ListCompositeRoutes(ctx context.Context, groupID int64) ([]service.CompositeModelRoute, error) {
+	s.advancedGroupOperationCalls++
 	return []service.CompositeModelRoute{
 		{
 			ID:             1,
@@ -311,6 +320,7 @@ func (s *stubAdminService) ListCompositeRoutes(ctx context.Context, groupID int6
 }

 func (s *stubAdminService) CreateCompositeRoute(ctx context.Context, groupID int64, input service.CompositeRouteInput) (*service.CompositeModelRoute, error) {
+	s.advancedGroupOperationCalls++
 	return &service.CompositeModelRoute{
 		ID:             1,
 		GroupID:        groupID,
@@ -326,6 +336,7 @@ func (s *stubAdminService) CreateCompositeRoute(ctx context.Context, groupID int
 }

 func (s *stubAdminService) UpdateCompositeRoute(ctx context.Context, groupID, routeID int64, input service.CompositeRouteInput) (*service.CompositeModelRoute, error) {
+	s.advancedGroupOperationCalls++
 	return &service.CompositeModelRoute{
 		ID:             routeID,
 		GroupID:        groupID,
@@ -341,10 +352,12 @@ func (s *stubAdminService) UpdateCompositeRoute(ctx context.Context, groupID, ro
 }

 func (s *stubAdminService) DeleteCompositeRoute(ctx context.Context, groupID, routeID int64) error {
+	s.advancedGroupOperationCalls++
 	return nil
 }

 func (s *stubAdminService) PreviewCompositeRoute(ctx context.Context, groupID int64, input service.CompositeRoutePreviewRequest) (*service.CompositeRouteDecision, error) {
+	s.advancedGroupOperationCalls++
 	decision, err := service.NewCompositeRouteResolver(nil).Resolve(ctx, groupID, input.Model, input.Endpoint)
 	if err != nil {
 		return nil, err
@@ -353,11 +366,13 @@ func (s *stubAdminService) PreviewCompositeRoute(ctx context.Context, groupID in
 }

 func (s *stubAdminService) CreateGroup(ctx context.Context, input *service.CreateGroupInput) (*service.Group, error) {
+	s.createdGroups = append(s.createdGroups, input)
 	group := service.Group{ID: 200, Name: input.Name, Status: service.StatusActive}
 	return &group, nil
 }

 func (s *stubAdminService) DuplicateGroup(ctx context.Context, id int64, actorScope, operationKey string) (*service.Group, error) {
+	s.advancedGroupOperationCalls++
 	group := service.Group{ID: 201, Name: "group (Copy)", Status: "inactive"}
 	return &group, nil
 }
@@ -367,14 +382,21 @@ func (s *stubAdminService) RecoverDuplicateGroup(ctx context.Context, id int64,
 }

 func (s *stubAdminService) UpdateGroup(ctx context.Context, id int64, input *service.UpdateGroupInput) (*service.Group, error) {
+	s.updatedGroups = append(s.updatedGroups, input)
 	group := service.Group{ID: id, Name: input.Name, Status: service.StatusActive}
 	return &group, nil
 }

 func (s *stubAdminService) DeleteGroup(ctx context.Context, id int64) error {
+	s.deletedGroupIDs = append(s.deletedGroupIDs, id)
 	return nil
 }

+func (s *stubAdminService) DeleteGroupIfEmpty(ctx context.Context, id int64) error {
+	s.guardedDeletedGroupIDs = append(s.guardedDeletedGroupIDs, id)
+	return s.deleteGroupIfEmptyErr
+}
+
 func (s *stubAdminService) GetGroupAPIKeys(ctx context.Context, groupID int64, page, pageSize int) ([]service.APIKey, int64, error) {
 	return s.apiKeys, int64(len(s.apiKeys)), nil
 }
@@ -384,10 +406,12 @@ func (s *stubAdminService) GetGroupRateMultipliers(_ context.Context, _ int64) (
 }

 func (s *stubAdminService) ClearGroupRateMultipliers(_ context.Context, _ int64) error {
+	s.advancedGroupOperationCalls++
 	return nil
 }

 func (s *stubAdminService) BatchSetGroupRateMultipliers(_ context.Context, _ int64, _ []service.GroupRateMultiplierInput) error {
+	s.advancedGroupOperationCalls++
 	return nil
 }

@@ -396,6 +420,7 @@ func (s *stubAdminService) ClearGroupRPMOverrides(_ context.Context, _ int64) er
 }

 func (s *stubAdminService) BatchSetGroupRPMOverrides(_ context.Context, _ int64, _ []service.GroupRPMOverrideInput) error {
+	s.advancedGroupOperationCalls++
 	return nil
 }

@@ -491,6 +516,8 @@ func (s *stubAdminService) CreateAccount(ctx context.Context, input *service.Cre
 	return &account, nil
 }

+func (s *stubAdminService) ValidateAccountGroupBindings(context.Context, []int64) error { return nil }
+
 func (s *stubAdminService) DuplicateAccount(ctx context.Context, id int64, actorScope, operationKey string) (*service.Account, error) {
 	account := service.Account{ID: 301, Name: "account (Copy)", Status: service.StatusActive, Schedulable: false}
 	return &account, nil
@@ -57,21 +57,22 @@ type updateChannelRequest struct {
 }

 type channelModelPricingRequest struct {
-	Platform          string                     `json:"platform" binding:"omitempty,max=50"`
-	Models            []string                   `json:"models" binding:"required,min=1,max=100"`
-	BillingMode       string                     `json:"billing_mode" binding:"omitempty,oneof=token per_request image"`
-	InputPrice        *float64                   `json:"input_price" binding:"omitempty,min=0"`
-	OutputPrice       *float64                   `json:"output_price" binding:"omitempty,min=0"`
-	CacheWritePrice   *float64                   `json:"cache_write_price" binding:"omitempty,min=0"`
-	CacheWrite1hPrice *float64                   `json:"cache_write_1h_price" binding:"omitempty,min=0"`
-	CacheReadPrice    *float64                   `json:"cache_read_price" binding:"omitempty,min=0"`
-	FastMultiplier    *float64                   `json:"fast_multiplier" binding:"omitempty,gt=0"`
-	FlexMultiplier    *float64                   `json:"flex_multiplier" binding:"omitempty,gt=0"`
-	ImageInputPrice   *float64                   `json:"image_input_price" binding:"omitempty,min=0"`
-	ImageOutputPrice  *float64                   `json:"image_output_price" binding:"omitempty,min=0"`
-	PerRequestPrice   *float64                   `json:"per_request_price" binding:"omitempty,min=0"`
-	Intervals         []pricingIntervalRequest   `json:"intervals"`
-	TimePricing       *channelTimePricingRequest `json:"time_pricing"`
+	Platform                     string                     `json:"platform" binding:"omitempty,max=50"`
+	Models                       []string                   `json:"models" binding:"required,min=1,max=100"`
+	BillingMode                  string                     `json:"billing_mode" binding:"omitempty,oneof=token per_request image"`
+	InputPrice                   *float64                   `json:"input_price" binding:"omitempty,min=0"`
+	OutputPrice                  *float64                   `json:"output_price" binding:"omitempty,min=0"`
+	CacheWritePrice              *float64                   `json:"cache_write_price" binding:"omitempty,min=0"`
+	CacheWrite1hPrice            *float64                   `json:"cache_write_1h_price" binding:"omitempty,min=0"`
+	CacheReadPrice               *float64                   `json:"cache_read_price" binding:"omitempty,min=0"`
+	FastMultiplier               *float64                   `json:"fast_multiplier" binding:"omitempty,gt=0"`
+	FlexMultiplier               *float64                   `json:"flex_multiplier" binding:"omitempty,gt=0"`
+	MaxReasoningEffortMultiplier *float64                   `json:"max_reasoning_effort_multiplier" binding:"omitempty,gt=0"`
+	ImageInputPrice              *float64                   `json:"image_input_price" binding:"omitempty,min=0"`
+	ImageOutputPrice             *float64                   `json:"image_output_price" binding:"omitempty,min=0"`
+	PerRequestPrice              *float64                   `json:"per_request_price" binding:"omitempty,min=0"`
+	Intervals                    []pricingIntervalRequest   `json:"intervals"`
+	TimePricing                  *channelTimePricingRequest `json:"time_pricing"`
 }

 type channelTimePricingRequest struct {
@@ -129,22 +130,23 @@ type channelResponse struct {
 }

 type channelModelPricingResponse struct {
-	ID                int64                       `json:"id"`
-	Platform          string                      `json:"platform"`
-	Models            []string                    `json:"models"`
-	BillingMode       string                      `json:"billing_mode"`
-	InputPrice        *float64                    `json:"input_price"`
-	OutputPrice       *float64                    `json:"output_price"`
-	CacheWritePrice   *float64                    `json:"cache_write_price"`
-	CacheWrite1hPrice *float64                    `json:"cache_write_1h_price"`
-	CacheReadPrice    *float64                    `json:"cache_read_price"`
-	FastMultiplier    *float64                    `json:"fast_multiplier"`
-	FlexMultiplier    *float64                    `json:"flex_multiplier"`
-	ImageInputPrice   *float64                    `json:"image_input_price"`
-	ImageOutputPrice  *float64                    `json:"image_output_price"`
-	PerRequestPrice   *float64                    `json:"per_request_price"`
-	Intervals         []pricingIntervalResponse   `json:"intervals"`
-	TimePricing       *channelTimePricingResponse `json:"time_pricing"`
+	ID                           int64                       `json:"id"`
+	Platform                     string                      `json:"platform"`
+	Models                       []string                    `json:"models"`
+	BillingMode                  string                      `json:"billing_mode"`
+	InputPrice                   *float64                    `json:"input_price"`
+	OutputPrice                  *float64                    `json:"output_price"`
+	CacheWritePrice              *float64                    `json:"cache_write_price"`
+	CacheWrite1hPrice            *float64                    `json:"cache_write_1h_price"`
+	CacheReadPrice               *float64                    `json:"cache_read_price"`
+	FastMultiplier               *float64                    `json:"fast_multiplier"`
+	FlexMultiplier               *float64                    `json:"flex_multiplier"`
+	MaxReasoningEffortMultiplier *float64                    `json:"max_reasoning_effort_multiplier"`
+	ImageInputPrice              *float64                    `json:"image_input_price"`
+	ImageOutputPrice             *float64                    `json:"image_output_price"`
+	PerRequestPrice              *float64                    `json:"per_request_price"`
+	Intervals                    []pricingIntervalResponse   `json:"intervals"`
+	TimePricing                  *channelTimePricingResponse `json:"time_pricing"`
 }

 type channelTimePricingResponse struct {
@@ -258,22 +260,23 @@ func pricingToResponse(p *service.ChannelModelPricing) channelModelPricingRespon
 		intervals = append(intervals, intervalToResponse(iv))
 	}
 	return channelModelPricingResponse{
-		ID:                p.ID,
-		Platform:          platform,
-		Models:            models,
-		BillingMode:       billingMode,
-		InputPrice:        p.InputPrice,
-		OutputPrice:       p.OutputPrice,
-		CacheWritePrice:   p.CacheWritePrice,
-		CacheWrite1hPrice: p.CacheWrite1hPrice,
-		CacheReadPrice:    p.CacheReadPrice,
-		FastMultiplier:    p.FastMultiplier,
-		FlexMultiplier:    p.FlexMultiplier,
-		ImageInputPrice:   p.ImageInputPrice,
-		ImageOutputPrice:  p.ImageOutputPrice,
-		PerRequestPrice:   p.PerRequestPrice,
-		Intervals:         intervals,
-		TimePricing:       timePricingToResponse(p.TimePricing),
+		ID:                           p.ID,
+		Platform:                     platform,
+		Models:                       models,
+		BillingMode:                  billingMode,
+		InputPrice:                   p.InputPrice,
+		OutputPrice:                  p.OutputPrice,
+		CacheWritePrice:              p.CacheWritePrice,
+		CacheWrite1hPrice:            p.CacheWrite1hPrice,
+		CacheReadPrice:               p.CacheReadPrice,
+		FastMultiplier:               p.FastMultiplier,
+		FlexMultiplier:               p.FlexMultiplier,
+		MaxReasoningEffortMultiplier: p.MaxReasoningEffortMultiplier,
+		ImageInputPrice:              p.ImageInputPrice,
+		ImageOutputPrice:             p.ImageOutputPrice,
+		PerRequestPrice:              p.PerRequestPrice,
+		Intervals:                    intervals,
+		TimePricing:                  timePricingToResponse(p.TimePricing),
 	}
 }

@@ -350,27 +353,29 @@ func pricingRequestToService(reqs []channelModelPricingRequest, allowChannelMult
 				SortOrder:            iv.SortOrder,
 			})
 		}
-		var fastMultiplier, flexMultiplier *float64
+		var fastMultiplier, flexMultiplier, maxReasoningEffortMultiplier *float64
 		if allowChannelMultipliers {
 			fastMultiplier = r.FastMultiplier
 			flexMultiplier = r.FlexMultiplier
+			maxReasoningEffortMultiplier = r.MaxReasoningEffortMultiplier
 		}
 		result = append(result, service.ChannelModelPricing{
-			Platform:          platform,
-			Models:            r.Models,
-			BillingMode:       billingMode,
-			InputPrice:        r.InputPrice,
-			OutputPrice:       r.OutputPrice,
-			CacheWritePrice:   r.CacheWritePrice,
-			CacheWrite1hPrice: r.CacheWrite1hPrice,
-			CacheReadPrice:    r.CacheReadPrice,
-			FastMultiplier:    fastMultiplier,
-			FlexMultiplier:    flexMultiplier,
-			ImageInputPrice:   r.ImageInputPrice,
-			ImageOutputPrice:  r.ImageOutputPrice,
-			PerRequestPrice:   r.PerRequestPrice,
-			Intervals:         intervals,
-			TimePricing:       timePricingRequestToService(r.TimePricing),
+			Platform:                     platform,
+			Models:                       r.Models,
+			BillingMode:                  billingMode,
+			InputPrice:                   r.InputPrice,
+			OutputPrice:                  r.OutputPrice,
+			CacheWritePrice:              r.CacheWritePrice,
+			CacheWrite1hPrice:            r.CacheWrite1hPrice,
+			CacheReadPrice:               r.CacheReadPrice,
+			FastMultiplier:               fastMultiplier,
+			FlexMultiplier:               flexMultiplier,
+			MaxReasoningEffortMultiplier: maxReasoningEffortMultiplier,
+			ImageInputPrice:              r.ImageInputPrice,
+			ImageOutputPrice:             r.ImageOutputPrice,
+			PerRequestPrice:              r.PerRequestPrice,
+			Intervals:                    intervals,
+			TimePricing:                  timePricingRequestToService(r.TimePricing),
 		})
 	}
 	return result
@@ -615,14 +620,15 @@ func (h *ChannelHandler) GetModelDefaultPricing(c *gin.Context) {
 	}

 	response.Success(c, gin.H{
-		"found":                true,
-		"input_price":          pricing.InputPricePerToken,
-		"output_price":         pricing.OutputPricePerToken,
-		"cache_write_price":    cacheWritePrice,
-		"cache_write_1h_price": cacheWrite1hPrice,
-		"cache_read_price":     pricing.CacheReadPricePerToken,
-		"image_input_price":    pricing.ImageInputPricePerToken,
-		"image_output_price":   pricing.ImageOutputPricePerToken,
+		"found":                           true,
+		"input_price":                     pricing.InputPricePerToken,
+		"output_price":                    pricing.OutputPricePerToken,
+		"cache_write_price":               cacheWritePrice,
+		"cache_write_1h_price":            cacheWrite1hPrice,
+		"cache_read_price":                pricing.CacheReadPricePerToken,
+		"max_reasoning_effort_multiplier": pricing.MaxReasoningEffortMultiplier,
+		"image_input_price":               pricing.ImageInputPricePerToken,
+		"image_output_price":              pricing.ImageOutputPricePerToken,
 	})
 }

@@ -453,10 +453,11 @@ func TestPricingRequestToService_TimePricingNil(t *testing.T) {
 // 避免渠道倍率意外污染账号成本口径。
 func TestPricingRequestToService_MultipliersGatedByFlag(t *testing.T) {
 	req := channelModelPricingRequest{
-		Models:         []string{"gpt-5"},
-		BillingMode:    "token",
-		FastMultiplier: float64Ptr(2.5),
-		FlexMultiplier: float64Ptr(0.5),
+		Models:                       []string{"gpt-5"},
+		BillingMode:                  "token",
+		FastMultiplier:               float64Ptr(2.5),
+		FlexMultiplier:               float64Ptr(0.5),
+		MaxReasoningEffortMultiplier: float64Ptr(3),
 		Intervals: []pricingIntervalRequest{{
 			MinTokens:            272000,
 			InputMultiplier:      float64Ptr(2),
@@ -469,6 +470,7 @@ func TestPricingRequestToService_MultipliersGatedByFlag(t *testing.T) {
 	allowed := pricingRequestToService([]channelModelPricingRequest{req}, true)
 	require.Equal(t, float64Ptr(2.5), allowed[0].FastMultiplier)
 	require.Equal(t, float64Ptr(0.5), allowed[0].FlexMultiplier)
+	require.Equal(t, float64Ptr(3), allowed[0].MaxReasoningEffortMultiplier)
 	require.Equal(t, float64Ptr(2), allowed[0].Intervals[0].InputMultiplier)
 	require.Equal(t, float64Ptr(1.5), allowed[0].Intervals[0].OutputMultiplier)
 	require.Equal(t, float64Ptr(2), allowed[0].Intervals[0].CacheWriteMultiplier)
@@ -477,6 +479,7 @@ func TestPricingRequestToService_MultipliersGatedByFlag(t *testing.T) {
 	dropped := pricingRequestToService([]channelModelPricingRequest{req}, false)
 	require.Nil(t, dropped[0].FastMultiplier)
 	require.Nil(t, dropped[0].FlexMultiplier)
+	require.Nil(t, dropped[0].MaxReasoningEffortMultiplier)
 	require.Nil(t, dropped[0].Intervals[0].InputMultiplier)
 	require.Nil(t, dropped[0].Intervals[0].OutputMultiplier)
 	require.Nil(t, dropped[0].Intervals[0].CacheWriteMultiplier)
@@ -580,9 +583,10 @@ func TestGetModelDefaultPricing_ReturnsFable51CacheTTLs(t *testing.T) {
 	require.Equal(t, http.StatusOK, w.Code)
 	var body struct {
 		Data struct {
-			Found             bool     `json:"found"`
-			CacheWritePrice   float64  `json:"cache_write_price"`
-			CacheWrite1hPrice *float64 `json:"cache_write_1h_price"`
+			Found                        bool     `json:"found"`
+			CacheWritePrice              float64  `json:"cache_write_price"`
+			CacheWrite1hPrice            *float64 `json:"cache_write_1h_price"`
+			MaxReasoningEffortMultiplier *float64 `json:"max_reasoning_effort_multiplier"`
 		} `json:"data"`
 	}
 	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &body))
@@ -590,6 +594,8 @@ func TestGetModelDefaultPricing_ReturnsFable51CacheTTLs(t *testing.T) {
 	require.InDelta(t, 12.5e-6, body.Data.CacheWritePrice, 1e-12)
 	require.NotNil(t, body.Data.CacheWrite1hPrice)
 	require.InDelta(t, 20e-6, *body.Data.CacheWrite1hPrice, 1e-12)
+	require.NotNil(t, body.Data.MaxReasoningEffortMultiplier)
+	require.Equal(t, 3.0, *body.Data.MaxReasoningEffortMultiplier)
 }

 func TestGetModelDefaultPricing_OmitsUnsupportedCache1hPrice(t *testing.T) {
@@ -6,6 +6,7 @@ import (
 	"sync"
 	"time"

+	"github.com/Wei-Shaw/sub2api/internal/config"
 	infraerrors "github.com/Wei-Shaw/sub2api/internal/pkg/errors"
 	"github.com/Wei-Shaw/sub2api/internal/service"
 )
@@ -194,6 +195,7 @@ func (h *GrokOAuthHandler) scheduleGrokImportProbe(account *service.Account) {
 // ProvideAccountHandler injects the Grok active prober for production while
 // keeping NewAccountHandler convenient for focused unit tests.
 func ProvideAccountHandler(
+	cfg *config.Config,
 	adminService service.AdminService,
 	oauthService *service.OAuthService,
 	openaiOAuthService *service.OpenAIOAuthService,
@@ -227,5 +229,6 @@ func ProvideAccountHandler(
 		tokenCacheInvalidator,
 	)
 	handler.grokImportProber = grokQuotaService
+	handler.cfg = cfg
 	return handler
 }
@@ -10,6 +10,7 @@ import (
 	"strings"
 	"time"

+	"github.com/Wei-Shaw/sub2api/internal/config"
 	"github.com/Wei-Shaw/sub2api/internal/handler/dto"
 	infraerrors "github.com/Wei-Shaw/sub2api/internal/pkg/errors"
 	"github.com/Wei-Shaw/sub2api/internal/pkg/response"
@@ -24,10 +25,14 @@ type GroupHandler struct {
 	adminService         service.AdminService
 	dashboardService     *service.DashboardService
 	groupCapacityService *service.GroupCapacityService
+	cfg                  *config.Config
 }

 // GetLiveCapability 返回当前服务端是否具备生成 Live attestation 的运行环境。
 func (h *GroupHandler) GetLiveCapability(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "live_capability") {
+		return
+	}
 	err := liveattestation.NewProvider().Check(c.Request.Context())
 	result := gin.H{"supported": err == nil}
 	if err != nil {
@@ -87,13 +92,94 @@ func (f optionalLimitField) ToServiceInput() *float64 {

 // NewGroupHandler creates a new admin group handler
 func NewGroupHandler(adminService service.AdminService, dashboardService *service.DashboardService, groupCapacityService *service.GroupCapacityService) *GroupHandler {
+	return NewGroupHandlerWithConfig(adminService, dashboardService, groupCapacityService, nil)
+}
+
+func NewGroupHandlerWithConfig(adminService service.AdminService, dashboardService *service.DashboardService, groupCapacityService *service.GroupCapacityService, cfg *config.Config) *GroupHandler {
 	return &GroupHandler{
 		adminService:         adminService,
 		dashboardService:     dashboardService,
 		groupCapacityService: groupCapacityService,
+		cfg:                  cfg,
 	}
 }

+func (h *GroupHandler) isSimpleMode() bool {
+	return h != nil && h.cfg != nil && h.cfg.RunMode == config.RunModeSimple
+}
+
+type simpleModeGroupOperation string
+
+const (
+	simpleModeGroupList   simpleModeGroupOperation = "list"
+	simpleModeGroupGetAll simpleModeGroupOperation = "get_all"
+	simpleModeGroupGet    simpleModeGroupOperation = "get"
+	simpleModeGroupCreate simpleModeGroupOperation = "create"
+	simpleModeGroupUpdate simpleModeGroupOperation = "update"
+	simpleModeGroupDelete simpleModeGroupOperation = "delete"
+)
+
+var simpleModeGroupOperations = map[simpleModeGroupOperation]struct{}{
+	simpleModeGroupList: {}, simpleModeGroupGetAll: {}, simpleModeGroupGet: {},
+	simpleModeGroupCreate: {}, simpleModeGroupUpdate: {}, simpleModeGroupDelete: {},
+}
+
+func (h *GroupHandler) rejectUnsupportedSimpleModeOperation(c *gin.Context, operation simpleModeGroupOperation) bool {
+	if _, allowed := simpleModeGroupOperations[operation]; allowed {
+		return false
+	}
+	if err := service.ValidateSimpleModeGroupOperation(h.cfg, service.AdminGroupOperation(operation)); err != nil {
+		response.ErrorFrom(c, err)
+		return true
+	}
+	return false
+}
+
+type simpleModeGroupResponse struct {
+	ID          int64  `json:"id"`
+	Name        string `json:"name"`
+	Description string `json:"description"`
+	Platform    string `json:"platform"`
+	Status      string `json:"status"`
+
+	AccountCount            int64     `json:"account_count,omitempty"`
+	ActiveAccountCount      int64     `json:"active_account_count,omitempty"`
+	RateLimitedAccountCount int64     `json:"rate_limited_account_count,omitempty"`
+	SortOrder               int       `json:"sort_order"`
+	CreatedAt               time.Time `json:"created_at"`
+	UpdatedAt               time.Time `json:"updated_at"`
+}
+
+func groupForSimpleMode(group *service.Group) *simpleModeGroupResponse {
+	if group == nil {
+		return nil
+	}
+	return &simpleModeGroupResponse{
+		ID: group.ID, Name: group.Name, Description: group.Description, Platform: group.Platform,
+		Status:             group.Status,
+		AccountCount:       group.AccountCount,
+		ActiveAccountCount: group.ActiveAccountCount, RateLimitedAccountCount: group.RateLimitedAccountCount,
+		SortOrder: group.SortOrder, CreatedAt: group.CreatedAt, UpdatedAt: group.UpdatedAt,
+	}
+}
+
+func sanitizeCreateGroupRequestForSimpleMode(req *CreateGroupRequest) {
+	if req == nil {
+		return
+	}
+	allowed := CreateGroupRequest{Name: req.Name, Description: req.Description, Platform: req.Platform}
+	allowed.RateMultiplier = 1
+	allowed.SubscriptionType = service.SubscriptionTypeStandard
+	*req = allowed
+}
+
+func sanitizeUpdateGroupRequestForSimpleMode(req *UpdateGroupRequest) {
+	if req == nil {
+		return
+	}
+	*req = UpdateGroupRequest{Name: req.Name, Description: req.Description}
+}
+
 // CreateGroupRequest represents create group request
 type CreateGroupRequest struct {
 	Name                      string                        `json:"name" binding:"required"`
@@ -158,11 +244,11 @@ type CreateGroupRequest struct {
 	CodexModelsManifestConfig service.GroupCodexModelsManifestConfig `json:"codex_models_manifest_config"`
 	// 分组 RPM 上限（0 = 不限制）
 	RPMLimit int `json:"rpm_limit"`
-	// OpenAI/Codex 请求推理强度上限，空字符串表示不限制。
+	// Anthropic/OpenAI 请求推理强度上限，空字符串表示不限制。
 	MaxReasoningEffort string `json:"max_reasoning_effort"`
 	// 超过上限时的访问控制：downgrade（默认）或 deny。
 	MaxReasoningEffortOverLimit string `json:"max_reasoning_effort_over_limit"`
-	// OpenAI/Codex 推理强度映射，可按模型精确名、前缀或后缀限定。
+	// Anthropic/OpenAI 推理强度映射，可按模型精确名、前缀或后缀限定。
 	ReasoningEffortMappings []service.ReasoningEffortMapping `json:"reasoning_effort_mappings"`
 	// 从指定分组复制账号（创建后自动绑定）
 	CopyAccountsFromGroupIDs []int64 `json:"copy_accounts_from_group_ids"`
@@ -233,7 +319,7 @@ type UpdateGroupRequest struct {
 	CodexModelsManifestConfig *service.GroupCodexModelsManifestConfig `json:"codex_models_manifest_config"`
 	// 分组 RPM 上限（0 = 不限制）；nil 表示未提供不改动
 	RPMLimit *int `json:"rpm_limit"`
-	// OpenAI/Codex 请求推理强度上限；空字符串清除，nil 不修改。
+	// Anthropic/OpenAI 请求推理强度上限；空字符串清除，nil 不修改。
 	MaxReasoningEffort *string `json:"max_reasoning_effort"`
 	// 超过上限时的访问控制；空字符串视为 downgrade，nil 不修改。
 	MaxReasoningEffortOverLimit *string `json:"max_reasoning_effort_over_limit"`
@@ -262,6 +348,9 @@ type CompositeRoutePreviewRequest struct {
 // List handles listing all groups with pagination
 // GET /api/v1/admin/groups
 func (h *GroupHandler) List(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupList) {
+		return
+	}
 	page, pageSize := response.ParsePagination(c)
 	platform := c.Query("platform")
 	status := c.Query("status")
@@ -276,7 +365,7 @@ func (h *GroupHandler) List(c *gin.Context) {
 	sortOrder := c.DefaultQuery("sort_order", "asc")

 	var isExclusive *bool
-	if isExclusiveStr != "" {
+	if !h.isSimpleMode() && isExclusiveStr != "" {
 		val := isExclusiveStr == "true"
 		isExclusive = &val
 	}
@@ -287,6 +376,16 @@ func (h *GroupHandler) List(c *gin.Context) {
 		return
 	}

+	if h.isSimpleMode() {
+		simpleGroups := make([]simpleModeGroupResponse, 0, len(groups))
+		for i := range groups {
+			if service.IsGroupBindableInSimpleMode(&groups[i]) {
+				simpleGroups = append(simpleGroups, *groupForSimpleMode(&groups[i]))
+			}
+		}
+		response.Paginated(c, simpleGroups, total, page, pageSize)
+		return
+	}
 	outGroups := make([]dto.AdminGroup, 0, len(groups))
 	for i := range groups {
 		outGroups = append(outGroups, *dto.GroupFromServiceAdmin(&groups[i]))
@@ -297,6 +396,9 @@ func (h *GroupHandler) List(c *gin.Context) {
 // ListCompositeRoutes handles listing composite model routes for one group.
 // GET /api/v1/admin/groups/:id/composite-routes
 func (h *GroupHandler) ListCompositeRoutes(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, ok := parsePositiveIDParam(c, "id")
 	if !ok {
 		return
@@ -312,6 +414,9 @@ func (h *GroupHandler) ListCompositeRoutes(c *gin.Context) {
 // CreateCompositeRoute handles creating one composite model route.
 // POST /api/v1/admin/groups/:id/composite-routes
 func (h *GroupHandler) CreateCompositeRoute(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, ok := parsePositiveIDParam(c, "id")
 	if !ok {
 		return
@@ -332,6 +437,9 @@ func (h *GroupHandler) CreateCompositeRoute(c *gin.Context) {
 // UpdateCompositeRoute handles replacing one composite model route.
 // PUT /api/v1/admin/groups/:id/composite-routes/:route_id
 func (h *GroupHandler) UpdateCompositeRoute(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, ok := parsePositiveIDParam(c, "id")
 	if !ok {
 		return
@@ -356,6 +464,9 @@ func (h *GroupHandler) UpdateCompositeRoute(c *gin.Context) {
 // DeleteCompositeRoute handles deleting one composite model route.
 // DELETE /api/v1/admin/groups/:id/composite-routes/:route_id
 func (h *GroupHandler) DeleteCompositeRoute(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, ok := parsePositiveIDParam(c, "id")
 	if !ok {
 		return
@@ -374,6 +485,9 @@ func (h *GroupHandler) DeleteCompositeRoute(c *gin.Context) {
 // PreviewCompositeRoute resolves a model without mutating routes.
 // POST /api/v1/admin/groups/:id/composite-routes/preview
 func (h *GroupHandler) PreviewCompositeRoute(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, ok := parsePositiveIDParam(c, "id")
 	if !ok {
 		return
@@ -427,6 +541,9 @@ func parsePositiveIDParam(c *gin.Context, name string) (int64, bool) {
 // bound to them even after the group is disabled).
 // GET /api/v1/admin/groups/all
 func (h *GroupHandler) GetAll(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupGetAll) {
+		return
+	}
 	platform := c.Query("platform")
 	includeInactive := c.Query("include_inactive") == "true"

@@ -446,6 +563,16 @@ func (h *GroupHandler) GetAll(c *gin.Context) {
 		return
 	}

+	if h.isSimpleMode() {
+		simpleGroups := make([]simpleModeGroupResponse, 0, len(groups))
+		for i := range groups {
+			if service.IsGroupBindableInSimpleMode(&groups[i]) {
+				simpleGroups = append(simpleGroups, *groupForSimpleMode(&groups[i]))
+			}
+		}
+		response.Success(c, simpleGroups)
+		return
+	}
 	outGroups := make([]dto.AdminGroup, 0, len(groups))
 	for i := range groups {
 		outGroups = append(outGroups, *dto.GroupFromServiceAdmin(&groups[i]))
@@ -456,6 +583,9 @@ func (h *GroupHandler) GetAll(c *gin.Context) {
 // GetByID handles getting a group by ID
 // GET /api/v1/admin/groups/:id
 func (h *GroupHandler) GetByID(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupGet) {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -468,12 +598,19 @@ func (h *GroupHandler) GetByID(c *gin.Context) {
 		return
 	}

+	if h.isSimpleMode() {
+		response.Success(c, groupForSimpleMode(group))
+		return
+	}
 	response.Success(c, dto.GroupFromServiceAdmin(group))
 }

 // GetGroupModelAllowlistCandidates handles getting candidate model IDs for the group model allowlist.
 // GET /api/v1/admin/groups/:id/model-allowlist-candidates
 func (h *GroupHandler) GetGroupModelAllowlistCandidates(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil || groupID < 0 {
 		response.BadRequest(c, "Invalid group ID")
@@ -496,11 +633,21 @@ func (h *GroupHandler) GetGroupModelAllowlistCandidates(c *gin.Context) {
 // Create handles creating a new group
 // POST /api/v1/admin/groups
 func (h *GroupHandler) Create(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupCreate) {
+		return
+	}
 	var req CreateGroupRequest
 	if err := c.ShouldBindJSON(&req); err != nil {
 		response.BadRequest(c, "Invalid request: "+err.Error())
 		return
 	}
+	if h.isSimpleMode() && req.Platform == service.PlatformComposite {
+		response.BadRequest(c, "Platform is not supported in simple mode")
+		return
+	}
+	if h.isSimpleMode() {
+		sanitizeCreateGroupRequestForSimpleMode(&req)
+	}

 	if err := service.ValidatePeakRateConfig(req.SubscriptionType, req.PeakRateEnabled, req.PeakStart, req.PeakEnd, float64ValueOrDefault(req.PeakRateMultiplier, 1.0)); err != nil {
 		response.BadRequest(c, err.Error())
@@ -581,12 +728,19 @@ func (h *GroupHandler) Create(c *gin.Context) {
 		return
 	}

+	if h.isSimpleMode() {
+		response.Success(c, groupForSimpleMode(group))
+		return
+	}
 	response.Success(c, dto.GroupFromServiceAdmin(group))
 }

 // Duplicate handles creating an inactive group copy with the source account bindings.
 // POST /api/v1/admin/groups/:id/duplicate
 func (h *GroupHandler) Duplicate(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil || groupID <= 0 {
 		response.BadRequest(c, "Invalid group ID")
@@ -634,6 +788,9 @@ func (h *GroupHandler) Duplicate(c *gin.Context) {
 // Update handles updating a group
 // PUT /api/v1/admin/groups/:id
 func (h *GroupHandler) Update(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupUpdate) {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -645,6 +802,9 @@ func (h *GroupHandler) Update(c *gin.Context) {
 		response.BadRequest(c, "Invalid request: "+err.Error())
 		return
 	}
+	if h.isSimpleMode() {
+		sanitizeUpdateGroupRequestForSimpleMode(&req)
+	}

 	group, err := h.adminService.UpdateGroup(c.Request.Context(), groupID, &service.UpdateGroupInput{
 		Name:                            req.Name,
@@ -714,19 +874,30 @@ func (h *GroupHandler) Update(c *gin.Context) {
 		return
 	}

+	if h.isSimpleMode() {
+		response.Success(c, groupForSimpleMode(group))
+		return
+	}
 	response.Success(c, dto.GroupFromServiceAdmin(group))
 }

 // Delete handles deleting a group
 // DELETE /api/v1/admin/groups/:id
 func (h *GroupHandler) Delete(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, simpleModeGroupDelete) {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
 		return
 	}

-	err = h.adminService.DeleteGroup(c.Request.Context(), groupID)
+	if h.isSimpleMode() {
+		err = h.adminService.DeleteGroupIfEmpty(c.Request.Context(), groupID)
+	} else {
+		err = h.adminService.DeleteGroup(c.Request.Context(), groupID)
+	}
 	if err != nil {
 		response.ErrorFrom(c, err)
 		return
@@ -738,6 +909,9 @@ func (h *GroupHandler) Delete(c *gin.Context) {
 // GetStats handles getting group statistics
 // GET /api/v1/admin/groups/:id/stats
 func (h *GroupHandler) GetStats(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "stats") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -757,6 +931,9 @@ func (h *GroupHandler) GetStats(c *gin.Context) {
 // GetUsageSummary returns today's, yesterday's, and cumulative cost for all groups.
 // GET /api/v1/admin/groups/usage-summary
 func (h *GroupHandler) GetUsageSummary(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	todayStart := service.GroupUsageTodayStart(time.Now())

 	results, err := h.dashboardService.GetGroupUsageSummary(c.Request.Context(), todayStart)
@@ -771,6 +948,9 @@ func (h *GroupHandler) GetUsageSummary(c *gin.Context) {
 // GetCapacitySummary returns aggregated capacity (concurrency/sessions/RPM) for all active groups.
 // GET /api/v1/admin/groups/capacity-summary
 func (h *GroupHandler) GetCapacitySummary(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	results, err := h.groupCapacityService.GetAllGroupCapacity(c.Request.Context())
 	if err != nil {
 		response.Error(c, 500, "Failed to get group capacity summary")
@@ -782,6 +962,9 @@ func (h *GroupHandler) GetCapacitySummary(c *gin.Context) {
 // GetGroupAPIKeys handles getting API keys in a group
 // GET /api/v1/admin/groups/:id/api-keys
 func (h *GroupHandler) GetGroupAPIKeys(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "api_keys") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -806,6 +989,9 @@ func (h *GroupHandler) GetGroupAPIKeys(c *gin.Context) {
 // GetGroupRateMultipliers handles getting rate multipliers for users in a group
 // GET /api/v1/admin/groups/:id/rate-multipliers
 func (h *GroupHandler) GetGroupRateMultipliers(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -827,6 +1013,9 @@ func (h *GroupHandler) GetGroupRateMultipliers(c *gin.Context) {
 // ClearGroupRateMultipliers handles clearing all rate multipliers for a group
 // DELETE /api/v1/admin/groups/:id/rate-multipliers
 func (h *GroupHandler) ClearGroupRateMultipliers(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -849,6 +1038,9 @@ type BatchSetGroupRateMultipliersRequest struct {
 // BatchSetGroupRateMultipliers handles batch setting rate multipliers for a group
 // PUT /api/v1/admin/groups/:id/rate-multipliers
 func (h *GroupHandler) BatchSetGroupRateMultipliers(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -877,6 +1069,9 @@ type BatchSetGroupRPMOverridesRequest struct {
 // BatchSetGroupRPMOverrides handles batch setting rpm_override for users in a group
 // PUT /api/v1/admin/groups/:id/rpm-overrides
 func (h *GroupHandler) BatchSetGroupRPMOverrides(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -900,6 +1095,9 @@ func (h *GroupHandler) BatchSetGroupRPMOverrides(c *gin.Context) {
 // ClearGroupRPMOverrides handles clearing all rpm_override for a group
 // DELETE /api/v1/admin/groups/:id/rpm-overrides
 func (h *GroupHandler) ClearGroupRPMOverrides(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	groupID, err := strconv.ParseInt(c.Param("id"), 10, 64)
 	if err != nil {
 		response.BadRequest(c, "Invalid group ID")
@@ -925,6 +1123,9 @@ type UpdateSortOrderRequest struct {
 // UpdateSortOrder handles updating group sort orders
 // PUT /api/v1/admin/groups/sort-order
 func (h *GroupHandler) UpdateSortOrder(c *gin.Context) {
+	if h.rejectUnsupportedSimpleModeOperation(c, "advanced") {
+		return
+	}
 	var req UpdateSortOrderRequest
 	if err := c.ShouldBindJSON(&req); err != nil {
 		response.BadRequest(c, "Invalid request: "+err.Error())
@@ -0,0 +1,309 @@
+package admin
+
+import (
+	"bytes"
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/Wei-Shaw/sub2api/internal/config"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+func newSimpleModeGroupRouter(svc *stubAdminService) *gin.Engine {
+	h := NewGroupHandlerWithConfig(svc, nil, nil, &config.Config{RunMode: config.RunModeSimple})
+	r := gin.New()
+	r.GET("/groups", h.List)
+	r.GET("/groups/all", h.GetAll)
+	r.GET("/groups/live-capability", h.GetLiveCapability)
+	r.GET("/groups/usage-summary", h.GetUsageSummary)
+	r.GET("/groups/capacity-summary", h.GetCapacitySummary)
+	r.PUT("/groups/sort-order", h.UpdateSortOrder)
+	r.GET("/groups/:id", h.GetByID)
+	r.GET("/groups/:id/stats", h.GetStats)
+	r.GET("/groups/:id/api-keys", h.GetGroupAPIKeys)
+	r.GET("/groups/:id/model-allowlist-candidates", h.GetGroupModelAllowlistCandidates)
+	r.POST("/groups", h.Create)
+	r.PUT("/groups/:id", h.Update)
+	r.POST("/groups/:id/duplicate", h.Duplicate)
+	r.GET("/groups/:id/composite-routes", h.ListCompositeRoutes)
+	r.POST("/groups/:id/composite-routes", h.CreateCompositeRoute)
+	r.POST("/groups/:id/composite-routes/preview", h.PreviewCompositeRoute)
+	r.PUT("/groups/:id/composite-routes/:route_id", h.UpdateCompositeRoute)
+	r.DELETE("/groups/:id/composite-routes/:route_id", h.DeleteCompositeRoute)
+	r.GET("/groups/:id/rate-multipliers", h.GetGroupRateMultipliers)
+	r.PUT("/groups/:id/rate-multipliers", h.BatchSetGroupRateMultipliers)
+	r.DELETE("/groups/:id/rate-multipliers", h.ClearGroupRateMultipliers)
+	r.PUT("/groups/:id/rpm-overrides", h.BatchSetGroupRPMOverrides)
+	r.DELETE("/groups/:id/rpm-overrides", h.ClearGroupRPMOverrides)
+	r.DELETE("/groups/:id", h.Delete)
+	return r
+}
+
+func TestGroupHandlerSimpleModeSanitizesCommercialFields(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	svc := newStubAdminService()
+	r := newSimpleModeGroupRouter(svc)
+
+	create := `{"name":"simple","description":"basic grouping","platform":"anthropic","rate_multiplier":7,"is_exclusive":true,"subscription_type":"subscription","daily_limit_usd":10,"long_context_pricing_enabled":true,"model_pricing":[{"model":"claude","input_price":1}],"allow_image_generation":true,"allow_batch_image_generation":true,"video_price_720p":2,"web_search_price_per_call":3,"audio_realtime_price_per_min":4,"rpm_limit":99}`
+	req := httptest.NewRequest(http.MethodPost, "/groups", bytes.NewBufferString(create))
+	req.Header.Set("Content-Type", "application/json")
+	res := httptest.NewRecorder()
+	r.ServeHTTP(res, req)
+	require.Equal(t, http.StatusOK, res.Code)
+	require.Len(t, svc.createdGroups, 1)
+	created := svc.createdGroups[0]
+	require.Equal(t, "basic grouping", created.Description)
+	require.Equal(t, 1.0, created.RateMultiplier)
+	require.False(t, created.IsExclusive)
+	require.Equal(t, service.SubscriptionTypeStandard, created.SubscriptionType)
+	require.Nil(t, created.DailyLimitUSD)
+	require.False(t, created.AllowImageGeneration)
+	require.False(t, created.LongContextPricingEnabled)
+	require.Empty(t, created.ModelPricing)
+	require.False(t, created.AllowBatchImageGeneration)
+	require.Nil(t, created.VideoPrice720P)
+	require.Nil(t, created.WebSearchPricePerCall)
+	require.Nil(t, created.AudioRealtimePricePerMin)
+	require.Zero(t, created.RPMLimit)
+
+	update := `{"name":"renamed","description":"still basic","rate_multiplier":9,"is_exclusive":true,"subscription_type":"subscription","daily_limit_usd":12,"long_context_pricing_enabled":true,"model_pricing":[{"model":"claude","input_price":1}],"allow_image_generation":true,"allow_batch_image_generation":true,"video_price_720p":2,"web_search_price_per_call":3,"audio_realtime_price_per_min":4,"status":"inactive","rpm_limit":123}`
+	req = httptest.NewRequest(http.MethodPut, "/groups/2", bytes.NewBufferString(update))
+	req.Header.Set("Content-Type", "application/json")
+	res = httptest.NewRecorder()
+	r.ServeHTTP(res, req)
+	require.Equal(t, http.StatusOK, res.Code)
+	require.Len(t, svc.updatedGroups, 1)
+	updated := svc.updatedGroups[0]
+	require.NotNil(t, updated.Description)
+	require.Equal(t, "still basic", *updated.Description)
+	require.Nil(t, updated.RateMultiplier)
+	require.Nil(t, updated.IsExclusive)
+	require.Empty(t, updated.SubscriptionType)
+	require.Nil(t, updated.DailyLimitUSD)
+	require.Nil(t, updated.AllowImageGeneration)
+	require.Nil(t, updated.LongContextPricingEnabled)
+	require.Nil(t, updated.ModelPricing)
+	require.Nil(t, updated.AllowBatchImageGeneration)
+	require.Nil(t, updated.VideoPrice720P)
+	require.Nil(t, updated.WebSearchPricePerCall)
+	require.Nil(t, updated.AudioRealtimePricePerMin)
+	require.Empty(t, updated.Status)
+	require.Nil(t, updated.RPMLimit)
+}
+
+func TestGroupHandlerSimpleModeRejectsCompositeBeforeService(t *testing.T) {
+	svc := newStubAdminService()
+	r := newSimpleModeGroupRouter(svc)
+	req := httptest.NewRequest(http.MethodPost, "/groups", bytes.NewBufferString(`{"name":"composite","platform":"composite"}`))
+	req.Header.Set("Content-Type", "application/json")
+	res := httptest.NewRecorder()
+	r.ServeHTTP(res, req)
+	require.Equal(t, http.StatusBadRequest, res.Code)
+	require.Empty(t, svc.createdGroups)
+}
+
+func TestGroupHandlerSimpleModeDeleteRejectsNonEmptyGroupBeforeCascade(t *testing.T) {
+	svc := newStubAdminService()
+	svc.deleteGroupIfEmptyErr = service.ErrGroupNotEmpty
+	r := newSimpleModeGroupRouter(svc)
+	res := httptest.NewRecorder()
+
+	r.ServeHTTP(res, httptest.NewRequest(http.MethodDelete, "/groups/7", nil))
+
+	require.Equal(t, http.StatusConflict, res.Code)
+	require.Equal(t, []int64{7}, svc.guardedDeletedGroupIDs)
+	require.Empty(t, svc.deletedGroupIDs)
+}
+
+func TestGroupHandlerSimpleModeDeletePermitsEmptyGroup(t *testing.T) {
+	svc := newStubAdminService()
+	r := newSimpleModeGroupRouter(svc)
+	res := httptest.NewRecorder()
+
+	r.ServeHTTP(res, httptest.NewRequest(http.MethodDelete, "/groups/8", nil))
+
+	require.Equal(t, http.StatusOK, res.Code)
+	require.Equal(t, []int64{8}, svc.guardedDeletedGroupIDs)
+	require.Empty(t, svc.deletedGroupIDs)
+}
+
+func TestGroupHandlerSimpleModeIgnoresExclusiveFilter(t *testing.T) {
+	svc := newStubAdminService()
+	r := newSimpleModeGroupRouter(svc)
+	res := httptest.NewRecorder()
+	r.ServeHTTP(res, httptest.NewRequest(http.MethodGet, "/groups?is_exclusive=true", nil))
+	require.Equal(t, http.StatusOK, res.Code)
+	require.Nil(t, svc.lastListGroupsIsExclusive)
+}
+
+func float64PtrForSimpleModeTest(value float64) *float64 { return &value }
+
+func TestGroupHandlerSimpleModeResponseUsesFieldAllowlist(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	svc := newStubAdminService()
+	svc.groups = []service.Group{{
+		ID: 1, Name: "basic", Description: "allowed", Platform: service.PlatformAnthropic,
+		AccountCount: 3, ActiveAccountCount: 2, RateLimitedAccountCount: 1,
+		Status: service.StatusActive, RateMultiplier: 9, RPMLimit: 42,
+		LongContextPricingEnabled: true,
+		ModelPricing:              []service.ChannelModelPricing{{Models: []string{"claude"}}},
+		AllowBatchImageGeneration: true, VideoPrice720P: float64PtrForSimpleModeTest(2),
+		WebSearchPricePerCall: float64PtrForSimpleModeTest(3), AudioRealtimePricePerMin: float64PtrForSimpleModeTest(4),
+		ModelRouting: map[string][]int64{"claude": {2}},
+	}}
+	r := newSimpleModeGroupRouter(svc)
+	res := httptest.NewRecorder()
+	r.ServeHTTP(res, httptest.NewRequest(http.MethodGet, "/groups", nil))
+	require.Equal(t, http.StatusOK, res.Code)
+
+	var payload struct {
+		Data struct {
+			Items []map[string]any `json:"items"`
+		} `json:"data"`
+	}
+	require.NoError(t, json.Unmarshal(res.Body.Bytes(), &payload))
+	require.Len(t, payload.Data.Items, 1)
+	item := payload.Data.Items[0]
+	require.Equal(t, "basic", item["name"])
+	require.Equal(t, "allowed", item["description"])
+	require.Equal(t, "anthropic", item["platform"])
+	require.Equal(t, float64(3), item["account_count"])
+	require.ElementsMatch(t, []string{
+		"id", "name", "description", "platform", "status", "account_count",
+		"active_account_count", "rate_limited_account_count", "sort_order", "created_at", "updated_at",
+	}, mapKeys(item))
+	for _, forbidden := range []string{
+		"rate_multiplier", "rpm_limit", "long_context_pricing_enabled", "model_pricing",
+		"allow_batch_image_generation", "video_price_720p", "web_search_price_per_call",
+		"audio_realtime_price_per_min", "model_routing",
+	} {
+		_, exposed := item[forbidden]
+		require.Falsef(t, exposed, "simple mode exposed %s", forbidden)
+	}
+}
+
+func TestGroupHandlerSimpleModeListsOnlyBindableGroups(t *testing.T) {
+	svc := newStubAdminService()
+	svc.groups = []service.Group{{ID: 1, Name: "basic", Platform: service.PlatformAnthropic}, {ID: 2, Name: "composite", Platform: service.PlatformComposite}}
+	r := newSimpleModeGroupRouter(svc)
+	for _, path := range []string{"/groups", "/groups/all"} {
+		res := httptest.NewRecorder()
+		r.ServeHTTP(res, httptest.NewRequest(http.MethodGet, path, nil))
+		require.Equal(t, http.StatusOK, res.Code)
+		require.Contains(t, res.Body.String(), `"name":"basic"`)
+		require.NotContains(t, res.Body.String(), `"name":"composite"`)
+	}
+}
+
+func mapKeys(values map[string]any) []string {
+	keys := make([]string, 0, len(values))
+	for key := range values {
+		keys = append(keys, key)
+	}
+	return keys
+}
+
+func TestGroupHandlerSimpleModeAllReadAndWriteResponsesUseFieldAllowlist(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	svc := newStubAdminService()
+	svc.groups = []service.Group{{
+		ID: 1, Name: "basic", Description: "allowed", Platform: service.PlatformAnthropic, AccountCount: 3,
+		Status: service.StatusActive, RateMultiplier: 9, RPMLimit: 42,
+		LongContextPricingEnabled: true,
+		ModelPricing:              []service.ChannelModelPricing{{Models: []string{"claude"}}},
+	}}
+	r := newSimpleModeGroupRouter(svc)
+
+	assertNoAdvancedFields := func(t *testing.T, item map[string]any) {
+		t.Helper()
+		for _, forbidden := range []string{
+			"rate_multiplier", "rpm_limit", "long_context_pricing_enabled", "model_pricing",
+			"allow_batch_image_generation", "video_price_720p", "web_search_price_per_call",
+			"audio_realtime_price_per_min", "model_routing",
+		} {
+			_, exposed := item[forbidden]
+			require.Falsef(t, exposed, "simple mode exposed %s", forbidden)
+		}
+	}
+
+	for _, tt := range []struct {
+		name   string
+		method string
+		path   string
+		body   string
+		list   bool
+	}{
+		{name: "get all", method: http.MethodGet, path: "/groups/all", list: true},
+		{name: "get by id", method: http.MethodGet, path: "/groups/1"},
+		{name: "create", method: http.MethodPost, path: "/groups", body: `{"name":"created","platform":"anthropic","model_pricing":[{"models":["claude"]}]}`},
+		{name: "update", method: http.MethodPut, path: "/groups/1", body: `{"name":"updated","model_pricing":[{"models":["claude"]}]}`},
+	} {
+		t.Run(tt.name, func(t *testing.T) {
+			res := httptest.NewRecorder()
+			req := httptest.NewRequest(tt.method, tt.path, bytes.NewBufferString(tt.body))
+			req.Header.Set("Content-Type", "application/json")
+			r.ServeHTTP(res, req)
+			require.Equal(t, http.StatusOK, res.Code)
+
+			var payload struct {
+				Data json.RawMessage `json:"data"`
+			}
+			require.NoError(t, json.Unmarshal(res.Body.Bytes(), &payload))
+			if tt.list {
+				var items []map[string]any
+				require.NoError(t, json.Unmarshal(payload.Data, &items))
+				require.Len(t, items, 1)
+				assertNoAdvancedFields(t, items[0])
+				return
+			}
+			var item map[string]any
+			require.NoError(t, json.Unmarshal(payload.Data, &item))
+			assertNoAdvancedFields(t, item)
+		})
+	}
+}
+
+func TestGroupHandlerSimpleModeBlocksAdvancedOperations(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	tests := []struct {
+		method string
+		path   string
+		body   string
+	}{
+		{http.MethodPost, "/groups/1/duplicate", ""},
+		{http.MethodGet, "/groups/1/model-allowlist-candidates", ""},
+		{http.MethodGet, "/groups/1/composite-routes", ""},
+		{http.MethodPost, "/groups/1/composite-routes", `{"public_model":"x","target_platform":"openai"}`},
+		{http.MethodPost, "/groups/1/composite-routes/preview", `{"model":"x"}`},
+		{http.MethodPut, "/groups/1/composite-routes/2", `{"public_model":"x","target_platform":"openai"}`},
+		{http.MethodDelete, "/groups/1/composite-routes/2", ""},
+		{http.MethodGet, "/groups/1/rate-multipliers", ""},
+		{http.MethodPut, "/groups/1/rate-multipliers", `{"entries":[]}`},
+		{http.MethodDelete, "/groups/1/rate-multipliers", ""},
+		{http.MethodPut, "/groups/1/rpm-overrides", `{"entries":[]}`},
+		{http.MethodDelete, "/groups/1/rpm-overrides", ""},
+		{http.MethodGet, "/groups/1/stats", ""},
+		{http.MethodGet, "/groups/1/api-keys", ""},
+		{http.MethodGet, "/groups/live-capability", ""},
+		{http.MethodGet, "/groups/usage-summary", ""},
+		{http.MethodGet, "/groups/capacity-summary", ""},
+		{http.MethodPut, "/groups/sort-order", `{"updates":[{"id":1,"sort_order":1}]}`},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.method+" "+tt.path, func(t *testing.T) {
+			svc := newStubAdminService()
+			r := newSimpleModeGroupRouter(svc)
+			res := httptest.NewRecorder()
+			req := httptest.NewRequest(tt.method, tt.path, bytes.NewBufferString(tt.body))
+			req.Header.Set("Content-Type", "application/json")
+			r.ServeHTTP(res, req)
+			require.Equal(t, http.StatusForbidden, res.Code)
+			require.Zero(t, svc.advancedGroupOperationCalls)
+		})
+	}
+}
@@ -228,7 +228,7 @@ func (h *RedeemHandler) CreateAndRedeem(c *gin.Context) {
 			return nil, createErr
 		}

-		redeemed, redeemErr := h.redeemService.Redeem(ctx, req.UserID, req.Code)
+		redeemed, redeemErr := h.redeemService.RedeemForAdminFulfillment(ctx, req.UserID, req.Code)
 		if redeemErr != nil {
 			return nil, redeemErr
 		}
@@ -246,7 +246,7 @@ func (h *RedeemHandler) resolveCreateAndRedeemExisting(ctx context.Context, exis
 		return nil, service.ErrRedeemCodeExpired
 	}
 	if existing.CanUse() {
-		redeemed, err := h.redeemService.Redeem(ctx, userID, existing.Code)
+		redeemed, err := h.redeemService.RedeemForAdminFulfillment(ctx, userID, existing.Code)
 		if err == nil {
 			return gin.H{"redeem_code": dto.RedeemCodeFromServiceAdmin(redeemed)}, nil
 		}
@@ -66,16 +66,17 @@ type userAvailableGroup struct {

 // userSupportedModelPricing 用户可见的定价字段白名单。
 type userSupportedModelPricing struct {
-	BillingMode       string                   `json:"billing_mode"`
-	InputPrice        *float64                 `json:"input_price"`
-	OutputPrice       *float64                 `json:"output_price"`
-	CacheWritePrice   *float64                 `json:"cache_write_price"`
-	CacheWrite1hPrice *float64                 `json:"cache_write_1h_price"`
-	CacheReadPrice    *float64                 `json:"cache_read_price"`
-	ImageInputPrice   *float64                 `json:"image_input_price"`
-	ImageOutputPrice  *float64                 `json:"image_output_price"`
-	PerRequestPrice   *float64                 `json:"per_request_price"`
-	Intervals         []userPricingIntervalDTO `json:"intervals"`
+	BillingMode                  string                   `json:"billing_mode"`
+	InputPrice                   *float64                 `json:"input_price"`
+	OutputPrice                  *float64                 `json:"output_price"`
+	CacheWritePrice              *float64                 `json:"cache_write_price"`
+	CacheWrite1hPrice            *float64                 `json:"cache_write_1h_price"`
+	CacheReadPrice               *float64                 `json:"cache_read_price"`
+	MaxReasoningEffortMultiplier *float64                 `json:"max_reasoning_effort_multiplier,omitempty"`
+	ImageInputPrice              *float64                 `json:"image_input_price"`
+	ImageOutputPrice             *float64                 `json:"image_output_price"`
+	PerRequestPrice              *float64                 `json:"per_request_price"`
+	Intervals                    []userPricingIntervalDTO `json:"intervals"`
 }

 // userPricingIntervalDTO 定价区间白名单（去掉内部 ID、SortOrder 等前端不渲染的字段）。
@@ -323,15 +324,16 @@ func toUserPricing(p *service.ChannelModelPricing) *userSupportedModelPricing {
 		billingMode = string(service.BillingModeToken)
 	}
 	return &userSupportedModelPricing{
-		BillingMode:       billingMode,
-		InputPrice:        p.InputPrice,
-		OutputPrice:       p.OutputPrice,
-		CacheWritePrice:   p.CacheWritePrice,
-		CacheWrite1hPrice: p.CacheWrite1hPrice,
-		CacheReadPrice:    p.CacheReadPrice,
-		ImageInputPrice:   p.ImageInputPrice,
-		ImageOutputPrice:  p.ImageOutputPrice,
-		PerRequestPrice:   p.PerRequestPrice,
-		Intervals:         intervals,
+		BillingMode:                  billingMode,
+		InputPrice:                   p.InputPrice,
+		OutputPrice:                  p.OutputPrice,
+		CacheWritePrice:              p.CacheWritePrice,
+		CacheWrite1hPrice:            p.CacheWrite1hPrice,
+		CacheReadPrice:               p.CacheReadPrice,
+		MaxReasoningEffortMultiplier: p.MaxReasoningEffortMultiplier,
+		ImageInputPrice:              p.ImageInputPrice,
+		ImageOutputPrice:             p.ImageOutputPrice,
+		PerRequestPrice:              p.PerRequestPrice,
+		Intervals:                    intervals,
 	}
 }
@@ -64,13 +64,45 @@ func openAIReasoningEffortPolicyForRequest(c *gin.Context, apiKey *service.APIKe
 	if apiKey == nil || apiKey.Group == nil {
 		return "", nil, "", false
 	}
-	if apiKey.Group.Platform != service.PlatformOpenAI && apiKey.Group.Platform != service.PlatformComposite {
+	if apiKey.Group.Platform != service.PlatformAnthropic && apiKey.Group.Platform != service.PlatformOpenAI && apiKey.Group.Platform != service.PlatformComposite {
 		return "", nil, "", false
 	}
-	if effectiveAPIKeyPlatform(c, apiKey) != service.PlatformOpenAI {
+	effectivePlatform := effectiveAPIKeyPlatform(c, apiKey)
+	if effectivePlatform != service.PlatformAnthropic && effectivePlatform != service.PlatformOpenAI {
 		return "", nil, "", false
 	}
-	return apiKey.Group.MaxReasoningEffort, apiKey.Group.ReasoningEffortMappings, apiKey.Group.MaxReasoningEffortOverLimit, true
+	maxEffort, mappings := apiKey.Group.MaxReasoningEffort, apiKey.Group.ReasoningEffortMappings
+	if effectivePlatform == service.PlatformAnthropic {
+		maxEffort, mappings = anthropicCompatibleReasoningEffortPolicy(maxEffort, mappings)
+	}
+	return maxEffort, mappings, apiKey.Group.MaxReasoningEffortOverLimit, true
+}
+
+func anthropicReasoningEffortPolicyForRequest(c *gin.Context, apiKey *service.APIKey) (string, []service.ReasoningEffortMapping, string, bool) {
+	if apiKey == nil || apiKey.Group == nil {
+		return "", nil, "", false
+	}
+	if apiKey.Group.Platform != service.PlatformAnthropic && apiKey.Group.Platform != service.PlatformComposite {
+		return "", nil, "", false
+	}
+	if effectiveAPIKeyPlatform(c, apiKey) != service.PlatformAnthropic {
+		return "", nil, "", false
+	}
+	maxEffort, mappings := anthropicCompatibleReasoningEffortPolicy(apiKey.Group.MaxReasoningEffort, apiKey.Group.ReasoningEffortMappings)
+	return maxEffort, mappings, apiKey.Group.MaxReasoningEffortOverLimit, true
+}
+
+func anthropicCompatibleReasoningEffortPolicy(maxEffort string, mappings []service.ReasoningEffortMapping) (string, []service.ReasoningEffortMapping) {
+	if service.NormalizeMaxReasoningEffort(maxEffort) == "minimal" {
+		maxEffort = "low"
+	}
+	normalizedMappings := append([]service.ReasoningEffortMapping(nil), mappings...)
+	for i := range normalizedMappings {
+		if service.NormalizeMaxReasoningEffort(normalizedMappings[i].To) == "minimal" {
+			normalizedMappings[i].To = "low"
+		}
+	}
+	return maxEffort, normalizedMappings
 }

 func bindRequestedReasoningEffort(c *gin.Context, body []byte, model string) {
@@ -118,6 +150,14 @@ func respondOpenAIReasoningEffortPolicyError(c *gin.Context, err error, write fu
 	write(c, http.StatusForbidden, "permission_error", err.Error())
 }

+func applyAnthropicReasoningEffortPolicyForRequest(c *gin.Context, apiKey *service.APIKey, body []byte) ([]byte, bool, error) {
+	maxEffort, mappings, overLimit, ok := anthropicReasoningEffortPolicyForRequest(c, apiKey)
+	if !ok {
+		return body, false, nil
+	}
+	return service.ApplyReasoningEffortPolicy(body, maxEffort, mappings, overLimit)
+}
+
 func bindOpenAIReasoningEffortPolicyForMessagesRequest(c *gin.Context, apiKey *service.APIKey, body []byte) {
 	if c == nil || c.Request == nil {
 		return
@@ -8,6 +8,7 @@ import (
 	"github.com/Wei-Shaw/sub2api/internal/service"
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/require"
+	"github.com/tidwall/gjson"
 )

 func TestCompositeTargetPlatformAllowedResolvesKnownAllowedModel(t *testing.T) {
@@ -160,6 +161,75 @@ func TestOpenAIReasoningEffortPolicyForCompositeTarget(t *testing.T) {
 	require.Equal(t, body, got)
 }

+func TestOpenAIShapedReasoningEffortPolicyForAnthropicGroup(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	c, _ := gin.CreateTestContext(httptest.NewRecorder())
+	c.Request = httptest.NewRequest("POST", "/v1/responses", nil)
+	apiKey := &service.APIKey{Group: &service.Group{
+		Platform:           service.PlatformAnthropic,
+		MaxReasoningEffort: "xhigh",
+	}}
+
+	got, changed, err := applyOpenAIReasoningEffortPolicyForRequest(c, apiKey, []byte(`{"reasoning":{"effort":"max"}}`))
+	require.NoError(t, err)
+	require.True(t, changed)
+	require.Equal(t, "xhigh", gjson.GetBytes(got, "reasoning.effort").String())
+}
+
+func TestAnthropicReasoningEffortPolicyForConcreteAndCompositeTargets(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	body := []byte(`{"model":"claude-fable-5-1","output_config":{"effort":"max"}}`)
+	policy := func(platform string) *service.APIKey {
+		return &service.APIKey{Group: &service.Group{
+			Platform:           platform,
+			MaxReasoningEffort: "xhigh",
+			ReasoningEffortMappings: []service.ReasoningEffortMapping{
+				{From: "max", To: "xhigh"},
+			},
+		}}
+	}
+
+	concrete, _ := gin.CreateTestContext(httptest.NewRecorder())
+	concrete.Request = httptest.NewRequest("POST", "/v1/messages", nil)
+	got, changed, err := applyAnthropicReasoningEffortPolicyForRequest(concrete, policy(service.PlatformAnthropic), body)
+	require.NoError(t, err)
+	require.True(t, changed)
+	require.JSONEq(t, `{"model":"claude-fable-5-1","output_config":{"effort":"xhigh"}}`, string(got))
+
+	denyMax := policy(service.PlatformAnthropic)
+	denyMax.Group.ReasoningEffortMappings = nil
+	denyMax.Group.MaxReasoningEffortOverLimit = service.ReasoningEffortOverLimitDeny
+	_, changed, err = applyAnthropicReasoningEffortPolicyForRequest(concrete, denyMax, body)
+	require.Error(t, err)
+	require.False(t, changed)
+	var overLimit *service.ReasoningEffortOverLimitError
+	require.ErrorAs(t, err, &overLimit)
+	require.Equal(t, "max", overLimit.Requested)
+	require.Equal(t, "xhigh", overLimit.Max)
+
+	composite, _ := gin.CreateTestContext(httptest.NewRecorder())
+	composite.Request = httptest.NewRequest("POST", "/v1/messages", nil)
+	composite.Request = composite.Request.WithContext(service.WithResolvedTargetPlatform(composite.Request.Context(), service.PlatformAnthropic))
+	got, changed, err = applyAnthropicReasoningEffortPolicyForRequest(composite, policy(service.PlatformComposite), body)
+	require.NoError(t, err)
+	require.True(t, changed)
+	require.Equal(t, "xhigh", gjson.GetBytes(got, "output_config.effort").String())
+
+	openAI, _ := gin.CreateTestContext(httptest.NewRecorder())
+	openAI.Request = httptest.NewRequest("POST", "/v1/messages", nil)
+	openAI.Request = openAI.Request.WithContext(service.WithResolvedTargetPlatform(openAI.Request.Context(), service.PlatformOpenAI))
+	got, changed, err = applyAnthropicReasoningEffortPolicyForRequest(openAI, policy(service.PlatformComposite), body)
+	require.NoError(t, err)
+	require.False(t, changed)
+	require.Equal(t, body, got)
+}
+
+func TestAnthropicCompositePolicyClampsUnsupportedMinimalToLow(t *testing.T) {
+	maxEffort, mappings := anthropicCompatibleReasoningEffortPolicy("minimal", []service.ReasoningEffortMapping{{From: "max", To: "minimal"}})
+	require.Equal(t, "low", maxEffort)
+	require.Equal(t, []service.ReasoningEffortMapping{{From: "max", To: "low"}}, mappings)
+}
+
 func TestClientRequestedModelUsesCompositePublicModel(t *testing.T) {
 	gin.SetMode(gin.TestMode)
 	c, _ := gin.CreateTestContext(httptest.NewRecorder())
@@ -9,10 +9,15 @@ import (

 const statusClientClosedRequest = 499

-func concurrencyErrorResponse(err error, slotType string) (int, string, string) {
+const (
+	gatewayQueueFullCode        = "gateway_queue_full"
+	gatewayConcurrencyLimitCode = "gateway_concurrency_limit"
+)
+
+func concurrencyErrorResponse(err error, slotType string) (int, string, string, string) {
 	var waitQueueFullErr *WaitQueueFullError
 	if errors.As(err, &waitQueueFullErr) {
-		return http.StatusTooManyRequests, "rate_limit_error",
+		return http.StatusTooManyRequests, "rate_limit_error", gatewayQueueFullCode,
 			"Too many pending requests, please retry later"
 	}

@@ -21,13 +26,13 @@ func concurrencyErrorResponse(err error, slotType string) (int, string, string)
 		if concurrencyErr.SlotType != "" {
 			slotType = concurrencyErr.SlotType
 		}
-		return http.StatusTooManyRequests, "rate_limit_error",
+		return http.StatusTooManyRequests, "rate_limit_error", gatewayConcurrencyLimitCode,
 			fmt.Sprintf("Concurrency limit exceeded for %s, please retry later", slotType)
 	}

 	if errors.Is(err, context.Canceled) {
-		return statusClientClosedRequest, "api_error", "context canceled"
+		return statusClientClosedRequest, "api_error", "", "context canceled"
 	}

-	return http.StatusServiceUnavailable, "api_error", "Service temporarily unavailable, please retry later"
+	return http.StatusServiceUnavailable, "api_error", "", "Service temporarily unavailable, please retry later"
 }
@@ -16,6 +16,7 @@ func TestConcurrencyErrorResponse(t *testing.T) {
 		slotType    string
 		wantStatus  int
 		wantType    string
+		wantCode    string
 		wantMessage string
 	}{
 		{
@@ -24,8 +25,18 @@ func TestConcurrencyErrorResponse(t *testing.T) {
 			slotType:    "user",
 			wantStatus:  http.StatusTooManyRequests,
 			wantType:    "rate_limit_error",
+			wantCode:    gatewayConcurrencyLimitCode,
 			wantMessage: "Concurrency limit exceeded for account, please retry later",
 		},
+		{
+			name:        "full local wait queue has gateway code",
+			err:         &WaitQueueFullError{SlotType: "account"},
+			slotType:    "account",
+			wantStatus:  http.StatusTooManyRequests,
+			wantType:    "rate_limit_error",
+			wantCode:    gatewayQueueFullCode,
+			wantMessage: "Too many pending requests, please retry later",
+		},
 		{
 			name:        "client cancellation is not classified as concurrency limit",
 			err:         context.Canceled,
@@ -54,9 +65,10 @@ func TestConcurrencyErrorResponse(t *testing.T) {

 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			status, errType, message := concurrencyErrorResponse(tt.err, tt.slotType)
+			status, errType, code, message := concurrencyErrorResponse(tt.err, tt.slotType)
 			require.Equal(t, tt.wantStatus, status)
 			require.Equal(t, tt.wantType, errType)
+			require.Equal(t, tt.wantCode, code)
 			require.Equal(t, tt.wantMessage, message)
 		})
 	}
@@ -448,6 +448,47 @@ func AccountFromService(a *service.Account) *Account {
 	return out
 }

+// AccountListItemFromAccount projects a full account response into the
+// compact shape used by the paginated admin account list. Keeping this
+// projection separate from Account preserves the existing detail API.
+func AccountListItemFromAccount(a *Account) *AccountListItem {
+	if a == nil {
+		return nil
+	}
+	return &AccountListItem{
+		ID: a.ID, Name: a.Name, Notes: a.Notes, Platform: a.Platform, Type: a.Type,
+		Credentials: a.Credentials, CredentialsStatus: a.CredentialsStatus, Extra: a.Extra,
+		OllamaCloudUsage: a.OllamaCloudUsage,
+		ProxyID:          a.ProxyID, ProxyFallbackOriginID: a.ProxyFallbackOriginID, ProxyFallbackOriginName: a.ProxyFallbackOriginName,
+		Concurrency: a.Concurrency, LoadFactor: a.LoadFactor, Priority: a.Priority, RateMultiplier: a.RateMultiplier,
+		Status: a.Status, ErrorMessage: a.ErrorMessage, LastUsedAt: a.LastUsedAt, ExpiresAt: a.ExpiresAt,
+		AutoPauseOnExpired: a.AutoPauseOnExpired, CreatedAt: a.CreatedAt, UpdatedAt: a.UpdatedAt,
+		Schedulable: a.Schedulable, RateLimitedAt: a.RateLimitedAt, RateLimitResetAt: a.RateLimitResetAt,
+		OverloadUntil: a.OverloadUntil, TempUnschedulableUntil: a.TempUnschedulableUntil,
+		TempUnschedulableReason: a.TempUnschedulableReason, SessionWindowStart: a.SessionWindowStart,
+		SessionWindowEnd: a.SessionWindowEnd, SessionWindowStatus: a.SessionWindowStatus,
+		WindowCostLimit: a.WindowCostLimit, WindowCostStickyReserve: a.WindowCostStickyReserve,
+		MaxSessions: a.MaxSessions, SessionIdleTimeoutMin: a.SessionIdleTimeoutMin, BaseRPM: a.BaseRPM,
+		RPMStrategy: a.RPMStrategy, RPMStickyBuffer: a.RPMStickyBuffer, UserMsgQueueMode: a.UserMsgQueueMode,
+		EnableTLSFingerprint: a.EnableTLSFingerprint, TLSFingerprintProfileID: a.TLSFingerprintProfileID,
+		EnableSessionIDMasking: a.EnableSessionIDMasking, CacheTTLOverrideEnabled: a.CacheTTLOverrideEnabled,
+		CacheTTLOverrideTarget: a.CacheTTLOverrideTarget, CustomBaseURLEnabled: a.CustomBaseURLEnabled,
+		CustomBaseURL: a.CustomBaseURL, QuotaLimit: a.QuotaLimit, QuotaUsed: a.QuotaUsed,
+		QuotaDailyLimit: a.QuotaDailyLimit, QuotaDailyUsed: a.QuotaDailyUsed, QuotaWeeklyLimit: a.QuotaWeeklyLimit,
+		QuotaWeeklyUsed: a.QuotaWeeklyUsed, QuotaDailyResetMode: a.QuotaDailyResetMode,
+		QuotaDailyResetHour: a.QuotaDailyResetHour, QuotaWeeklyResetMode: a.QuotaWeeklyResetMode,
+		QuotaWeeklyResetDay: a.QuotaWeeklyResetDay, QuotaWeeklyResetHour: a.QuotaWeeklyResetHour,
+		QuotaResetTimezone: a.QuotaResetTimezone, QuotaDailyResetAt: a.QuotaDailyResetAt,
+		QuotaWeeklyResetAt: a.QuotaWeeklyResetAt, QuotaNotifyDailyEnabled: a.QuotaNotifyDailyEnabled,
+		QuotaNotifyDailyThreshold: a.QuotaNotifyDailyThreshold, QuotaNotifyWeeklyEnabled: a.QuotaNotifyWeeklyEnabled,
+		QuotaNotifyWeeklyThreshold: a.QuotaNotifyWeeklyThreshold, QuotaNotifyTotalEnabled: a.QuotaNotifyTotalEnabled,
+		QuotaNotifyTotalThreshold: a.QuotaNotifyTotalThreshold, ParentAccountID: a.ParentAccountID,
+		QuotaDimension: a.QuotaDimension, ParentEmail: a.ParentEmail, ParentPlanType: a.ParentPlanType,
+		ParentPrivacyMode: a.ParentPrivacyMode, ParentSubscriptionExpiresAt: a.ParentSubscriptionExpiresAt,
+		ParentChatGPTAccountID: a.ParentChatGPTAccountID, Proxy: a.Proxy, GroupIDs: a.GroupIDs,
+	}
+}
+
 func timeToUnixSeconds(value *time.Time) *int64 {
 	if value == nil {
 		return nil
@@ -151,11 +151,11 @@ type Group struct {

 	// RPMLimit 分组级每分钟请求数上限（0 = 不限制），设置后覆盖用户级 rpm_limit。
 	RPMLimit int `json:"rpm_limit"`
-	// MaxReasoningEffort OpenAI/Codex 请求的推理强度上限，空字符串表示不限制。
+	// MaxReasoningEffort Anthropic/OpenAI 请求的推理强度上限，空字符串表示不限制。
 	MaxReasoningEffort string `json:"max_reasoning_effort"`
 	// MaxReasoningEffortOverLimit 超过上限时的访问控制：downgrade（默认）或 deny。
 	MaxReasoningEffortOverLimit string `json:"max_reasoning_effort_over_limit"`
-	// ReasoningEffortMappings OpenAI/Codex 推理强度映射，可按模型精确名、前缀或后缀限定。
+	// ReasoningEffortMappings Anthropic/OpenAI 推理强度映射，可按模型精确名、前缀或后缀限定。
 	ReasoningEffortMappings []domain.ReasoningEffortMapping `json:"reasoning_effort_mappings"`

 	CreatedAt time.Time `json:"created_at"`
@@ -324,6 +324,102 @@ type Account struct {
 	Groups   []*Group `json:"groups,omitempty"`
 }

+// AccountListItem is the compact representation returned by the admin account
+// list when the lite query parameter is enabled. It contains the fields used
+// by the account table and runtime indicators, but intentionally omits the
+// repeated account_groups and groups object graphs. Fetch /admin/accounts/:id
+// for the complete Account DTO when editing or inspecting an account.
+type AccountListItem struct {
+	ID       int64   `json:"id"`
+	Name     string  `json:"name"`
+	Notes    *string `json:"notes"`
+	Platform string  `json:"platform"`
+	Type     string  `json:"type"`
+
+	Credentials       map[string]any                 `json:"credentials,omitempty"`
+	CredentialsStatus map[string]bool                `json:"credentials_status,omitempty"`
+	Extra             map[string]any                 `json:"extra,omitempty"`
+	OllamaCloudUsage  *service.OllamaCloudUsageState `json:"ollama_cloud_usage,omitempty"`
+
+	ProxyID                 *int64     `json:"proxy_id"`
+	ProxyFallbackOriginID   *int64     `json:"proxy_fallback_origin_id"`
+	ProxyFallbackOriginName *string    `json:"proxy_fallback_origin_name,omitempty"`
+	Concurrency             int        `json:"concurrency"`
+	LoadFactor              *int       `json:"load_factor,omitempty"`
+	Priority                int        `json:"priority"`
+	RateMultiplier          float64    `json:"rate_multiplier"`
+	Status                  string     `json:"status"`
+	ErrorMessage            string     `json:"error_message"`
+	LastUsedAt              *time.Time `json:"last_used_at"`
+	ExpiresAt               *int64     `json:"expires_at"`
+	AutoPauseOnExpired      bool       `json:"auto_pause_on_expired"`
+	CreatedAt               time.Time  `json:"created_at"`
+	UpdatedAt               time.Time  `json:"updated_at"`
+
+	Schedulable bool `json:"schedulable"`
+
+	RateLimitedAt    *time.Time `json:"rate_limited_at"`
+	RateLimitResetAt *time.Time `json:"rate_limit_reset_at"`
+	OverloadUntil    *time.Time `json:"overload_until"`
+
+	TempUnschedulableUntil  *time.Time `json:"temp_unschedulable_until"`
+	TempUnschedulableReason string     `json:"temp_unschedulable_reason"`
+
+	SessionWindowStart  *time.Time `json:"session_window_start"`
+	SessionWindowEnd    *time.Time `json:"session_window_end"`
+	SessionWindowStatus string     `json:"session_window_status"`
+
+	WindowCostLimit         *float64 `json:"window_cost_limit,omitempty"`
+	WindowCostStickyReserve *float64 `json:"window_cost_sticky_reserve,omitempty"`
+	MaxSessions             *int     `json:"max_sessions,omitempty"`
+	SessionIdleTimeoutMin   *int     `json:"session_idle_timeout_minutes,omitempty"`
+	BaseRPM                 *int     `json:"base_rpm,omitempty"`
+	RPMStrategy             *string  `json:"rpm_strategy,omitempty"`
+	RPMStickyBuffer         *int     `json:"rpm_sticky_buffer,omitempty"`
+	UserMsgQueueMode        *string  `json:"user_msg_queue_mode,omitempty"`
+	EnableTLSFingerprint    *bool    `json:"enable_tls_fingerprint,omitempty"`
+	TLSFingerprintProfileID *int64   `json:"tls_fingerprint_profile_id,omitempty"`
+	EnableSessionIDMasking  *bool    `json:"session_id_masking_enabled,omitempty"`
+	CacheTTLOverrideEnabled *bool    `json:"cache_ttl_override_enabled,omitempty"`
+	CacheTTLOverrideTarget  *string  `json:"cache_ttl_override_target,omitempty"`
+	CustomBaseURLEnabled    *bool    `json:"custom_base_url_enabled,omitempty"`
+	CustomBaseURL           *string  `json:"custom_base_url,omitempty"`
+
+	QuotaLimit       *float64 `json:"quota_limit,omitempty"`
+	QuotaUsed        *float64 `json:"quota_used,omitempty"`
+	QuotaDailyLimit  *float64 `json:"quota_daily_limit,omitempty"`
+	QuotaDailyUsed   *float64 `json:"quota_daily_used,omitempty"`
+	QuotaWeeklyLimit *float64 `json:"quota_weekly_limit,omitempty"`
+	QuotaWeeklyUsed  *float64 `json:"quota_weekly_used,omitempty"`
+
+	QuotaDailyResetMode  *string `json:"quota_daily_reset_mode,omitempty"`
+	QuotaDailyResetHour  *int    `json:"quota_daily_reset_hour,omitempty"`
+	QuotaWeeklyResetMode *string `json:"quota_weekly_reset_mode,omitempty"`
+	QuotaWeeklyResetDay  *int    `json:"quota_weekly_reset_day,omitempty"`
+	QuotaWeeklyResetHour *int    `json:"quota_weekly_reset_hour,omitempty"`
+	QuotaResetTimezone   *string `json:"quota_reset_timezone,omitempty"`
+	QuotaDailyResetAt    *string `json:"quota_daily_reset_at,omitempty"`
+	QuotaWeeklyResetAt   *string `json:"quota_weekly_reset_at,omitempty"`
+
+	QuotaNotifyDailyEnabled    *bool    `json:"quota_notify_daily_enabled,omitempty"`
+	QuotaNotifyDailyThreshold  *float64 `json:"quota_notify_daily_threshold,omitempty"`
+	QuotaNotifyWeeklyEnabled   *bool    `json:"quota_notify_weekly_enabled,omitempty"`
+	QuotaNotifyWeeklyThreshold *float64 `json:"quota_notify_weekly_threshold,omitempty"`
+	QuotaNotifyTotalEnabled    *bool    `json:"quota_notify_total_enabled,omitempty"`
+	QuotaNotifyTotalThreshold  *float64 `json:"quota_notify_total_threshold,omitempty"`
+
+	ParentAccountID             *int64 `json:"parent_account_id,omitempty"`
+	QuotaDimension              string `json:"quota_dimension,omitempty"`
+	ParentEmail                 string `json:"parent_email,omitempty"`
+	ParentPlanType              string `json:"parent_plan_type,omitempty"`
+	ParentPrivacyMode           string `json:"parent_privacy_mode,omitempty"`
+	ParentSubscriptionExpiresAt string `json:"parent_subscription_expires_at,omitempty"`
+	ParentChatGPTAccountID      string `json:"parent_chatgpt_account_id,omitempty"`
+
+	Proxy    *Proxy  `json:"proxy,omitempty"`
+	GroupIDs []int64 `json:"group_ids,omitempty"`
+}
+
 type AccountGroup struct {
 	AccountID int64     `json:"account_id"`
 	GroupID   int64     `json:"group_id"`
@@ -171,6 +171,19 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 	reqStream := parsedReq.Stream
 	bindRequestedReasoningEffort(c, body, reqModel)
 	ensureCompositeTargetPlatform(c, apiKey, reqModel)
+	if policyBody, changed, err := applyAnthropicReasoningEffortPolicyForRequest(c, apiKey, body); err != nil {
+		respondOpenAIReasoningEffortPolicyError(c, err, h.errorResponse)
+		return
+	} else if changed {
+		if err := parsedReq.ReplaceBody(policyBody); err != nil {
+			reqLog.Warn("gateway.reasoning_effort_policy_parse_failed", zap.Error(err))
+			h.errorResponse(c, http.StatusBadRequest, "invalid_request_error", "Failed to apply reasoning effort policy")
+			return
+		}
+		body = parsedReq.Body.Bytes()
+		reqModel = parsedReq.Model
+		reqStream = parsedReq.Stream
+	}
 	reqLog = reqLog.With(zap.String("model", reqModel), zap.Bool("stream", reqStream))

 	// 解析渠道级模型映射
@@ -395,7 +408,7 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 						zap.Int64("account_id", account.ID),
 						zap.Int("max_waiting", selection.WaitPlan.MaxWaiting),
 					)
-					h.handleStreamingAwareError(c, http.StatusTooManyRequests, "rate_limit_error", "Too many pending requests, please retry later", streamStarted)
+					h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error", gatewayQueueFullCode, "Too many pending requests, please retry later", streamStarted)
 					return
 				}
 				if err == nil && canWait {
@@ -610,6 +623,23 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 		c.Request = c.Request.WithContext(ctx)
 	}

+	// 会话槽失败释放：failover 链上每个选中的 Anthropic OAuth 账号都注册过同一会话
+	// （checkAndRegisterSession）。若请求最终失败（转发失败/客户端中断/换号耗尽），
+	// 须立即释放本次会话注册——上游从未真正服务该会话，继续占槽会让 max_sessions
+	// 受限的账号被失败请求的 session hash 卡满整个空闲窗口，后续新会话全部被拒。
+	// 成功请求保持既有空闲超时语义（会话按最后活动时间过期）。
+	sessionSlotAccounts := make(map[int64]*service.Account)
+	upstreamServedSession := false
+	defer func() {
+		if upstreamServedSession {
+			return
+		}
+		// 客户端可能已断开、请求 ctx 已取消，用独立 ctx 执行释放
+		for _, acc := range sessionSlotAccounts {
+			h.gatewayService.ReleaseAccountSession(context.Background(), acc, sessionKey)
+		}
+	}()
+
 	for {
 		fs := NewFailoverState(h.maxAccountSwitches, hasBoundSession)
 		retryWithFallback := false
@@ -719,7 +749,7 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 						zap.Int64("account_id", account.ID),
 						zap.Int("max_waiting", selection.WaitPlan.MaxWaiting),
 					)
-					h.handleStreamingAwareError(c, http.StatusTooManyRequests, "rate_limit_error", "Too many pending requests, please retry later", streamStarted)
+					h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error", gatewayQueueFullCode, "Too many pending requests, please retry later", streamStarted)
 					return
 				}
 				if err == nil && canWait {
@@ -763,10 +793,15 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 					h.handleStreamingAwareError(c, http.StatusServiceUnavailable, "api_error", profitVetoExhaustedMessage, streamStarted)
 					return
 				}
+				// 尝试被否决（从未转发），立即释放该账号的会话注册
+				h.gatewayService.ReleaseAccountSession(context.Background(), account, sessionKey)
+				delete(sessionSlotAccounts, account.ID)
 				continue
 			}
 			account = latest
 			selection.Account = latest
+			// 记录本请求注册过会话槽的账号（profit 准入后账号已定）
+			sessionSlotAccounts[account.ID] = account
 			// 等待路径保持既有 eager 绑定（无门时 helper 直接绑定）；调度器已
 			// 抢槽的直达路径无门时由选号内部绑定，这里只在门下补准入后绑定。
 			if selection.ProfitGateActive() || !selection.Acquired {
@@ -983,6 +1018,11 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 						currentSubscription = nil
 						fallbackUsed = true
 						retryWithFallback = true
+						// 原分组账号已确定性失败（prompt too long），先释放其会话注册再走兜底分组
+						for _, acc := range sessionSlotAccounts {
+							h.gatewayService.ReleaseAccountSession(context.Background(), acc, sessionKey)
+						}
+						sessionSlotAccounts = make(map[int64]*service.Account)
 						break
 					}
 					_ = h.antigravityGatewayService.WriteMappedClaudeError(c, account, promptTooLongErr.StatusCode, promptTooLongErr.RequestID, promptTooLongErr.Body)
@@ -998,6 +1038,9 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 					action := fs.HandleFailoverError(c.Request.Context(), h.gatewayService, account.ID, account.Platform, account.GetPoolModeRetryCount(), failoverErr)
 					switch action {
 					case FailoverContinue:
+						// 本次尝试已确定性失败，立即释放该账号的会话注册
+						h.gatewayService.ReleaseAccountSession(context.Background(), account, sessionKey)
+						delete(sessionSlotAccounts, account.ID)
 						continue
 					case FailoverExhausted:
 						h.handleFailoverExhausted(c, fs.LastFailoverErr, account.Platform, streamStarted)
@@ -1036,6 +1079,8 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 				// 不会走到这里重复计费。
 				if result != nil {
 					submitForwardUsage(result)
+					// 上游已接受并计量本次会话（流中断），会话槽保持既有语义
+					upstreamServedSession = true
 				}
 				return
 			}
@@ -1061,6 +1106,8 @@ func (h *GatewayHandler) Messages(c *gin.Context) {
 			}

 			submitForwardUsage(result)
+			// 转发成功，会话槽保持既有空闲超时语义
+			upstreamServedSession = true
 			return
 		}
 		if !retryWithFallback {
@@ -1087,6 +1134,12 @@ func (h *GatewayHandler) Models(c *gin.Context) {
 		platform = forcedPlatform
 	}

+	if platform == service.PlatformOpenAI && apiKey != nil && apiKey.Group != nil &&
+		apiKey.Group.Platform == service.PlatformOpenAI && apiKey.Group.CodexModelsManifestConfig.Enabled {
+		h.pinnedOpenAIModels(c, apiKey.Group)
+		return
+	}
+
 	if platform == service.PlatformComposite {
 		availableModels := h.compositeAvailableModels(c.Request.Context(), groupID)
 		if apiKey != nil && apiKey.Group != nil && apiKey.Group.ModelAllowlistEnabled() {
@@ -1813,8 +1866,8 @@ func (h *GatewayHandler) calculateSubscriptionRemaining(group *service.Group, su

 // handleConcurrencyError handles concurrency-related acquire errors.
 func (h *GatewayHandler) handleConcurrencyError(c *gin.Context, err error, slotType string, streamStarted bool) {
-	status, errType, message := concurrencyErrorResponse(err, slotType)
-	h.handleStreamingAwareError(c, status, errType, message, streamStarted)
+	status, errType, code, message := concurrencyErrorResponse(err, slotType)
+	h.handleStreamingAwareErrorWithCode(c, status, errType, code, message, streamStarted)
 }

 func (h *GatewayHandler) handleFailoverExhausted(c *gin.Context, failoverErr *service.UpstreamFailoverError, platform string, streamStarted bool) {
@@ -1885,6 +1938,10 @@ func (h *GatewayHandler) mapUpstreamError(statusCode int) (int, string, string)

 // handleStreamingAwareError handles errors that may occur after streaming has started
 func (h *GatewayHandler) handleStreamingAwareError(c *gin.Context, status int, errType, message string, streamStarted bool) {
+	h.handleStreamingAwareErrorWithCode(c, status, errType, "", message, streamStarted)
+}
+
+func (h *GatewayHandler) handleStreamingAwareErrorWithCode(c *gin.Context, status int, errType, code, message string, streamStarted bool) {
 	if streamStarted {
 		// 响应状态码已固化为 200（ping/部分数据已 flush），错误只能就地以 SSE 帧回传。
 		// 标记本次流内错误，供 ops_error_logger 补记——否则该中间件按 status>=400 采集，
@@ -1895,7 +1952,7 @@ func (h *GatewayHandler) handleStreamingAwareError(c *gin.Context, status int, e
 		// response.completed/failed/incomplete/cancelled 集合。
 		// Anthropic-backed Responses 路径同样会因为通用 error 帧被拒。
 		if inboundIsResponses(c) {
-			if writeResponsesFailedSSE(c, errType, message) {
+			if writeResponsesFailedSSE(c, errType, code, message) {
 				return
 			}
 		}
@@ -1903,7 +1960,11 @@ func (h *GatewayHandler) handleStreamingAwareError(c *gin.Context, status int, e
 		flusher, ok := c.Writer.(http.Flusher)
 		if ok {
 			// SSE 错误事件固定 schema，使用 Quote 直拼可避免额外 Marshal 分配。
-			errorEvent := `data: {"type":"error","error":{"type":` + strconv.Quote(errType) + `,"message":` + strconv.Quote(message) + `}}` + "\n\n"
+			errorCode := ""
+			if code != "" {
+				errorCode = `,"code":` + strconv.Quote(code)
+			}
+			errorEvent := `data: {"type":"error","error":{"type":` + strconv.Quote(errType) + errorCode + `,"message":` + strconv.Quote(message) + `}}` + "\n\n"
 			if _, err := fmt.Fprint(c.Writer, errorEvent); err != nil {
 				_ = c.Error(err)
 			}
@@ -1913,7 +1974,7 @@ func (h *GatewayHandler) handleStreamingAwareError(c *gin.Context, status int, e
 	}

 	// Normal case: return JSON response with proper status code
-	h.errorResponse(c, status, errType, message)
+	h.errorResponseWithCode(c, status, errType, code, message)
 }

 // ensureForwardErrorResponse 在 Forward 返回错误但尚未写响应时补写统一错误响应。
@@ -2005,12 +2066,17 @@ func (h *GatewayHandler) checkClaudeCodeVersion(c *gin.Context) bool {

 // errorResponse 返回Claude API格式的错误响应
 func (h *GatewayHandler) errorResponse(c *gin.Context, status int, errType, message string) {
+	h.errorResponseWithCode(c, status, errType, "", message)
+}
+
+func (h *GatewayHandler) errorResponseWithCode(c *gin.Context, status int, errType, code, message string) {
+	errorObject := gin.H{"type": errType, "message": message}
+	if code != "" {
+		errorObject["code"] = code
+	}
 	c.JSON(status, gin.H{
-		"type": "error",
-		"error": gin.H{
-			"type":    errType,
-			"message": message,
-		},
+		"type":  "error",
+		"error": errorObject,
 	})
 }

@@ -2123,6 +2189,8 @@ func (h *GatewayHandler) CountTokens(c *gin.Context) {
 	if err := h.gatewayService.ForwardCountTokens(c.Request.Context(), c, account, parsedReq); err != nil {
 		reqLog.Error("gateway.count_tokens_forward_failed", zap.Int64("account_id", account.ID), zap.Error(err))
 		// 错误响应已在 ForwardCountTokens 中处理
+		// 上游未服务该会话，立即释放选号时注册的会话槽（客户端可能已断开，用独立 ctx）
+		h.gatewayService.ReleaseAccountSession(context.Background(), account, sessionHash)
 		return
 	}
 }
@@ -394,7 +394,7 @@ func (h *GatewayHandler) handleResponsesFailoverExhausted(c *gin.Context, lastEr
 		// generic response.failed.
 		service.MarkOpsStreamError(c, code, message, status)
 		if c != nil && c.Writer != nil && (c.Writer.Size() <= 0 || gatewayStreamHasOnlyHeartbeats(c)) {
-			writeResponsesFailedSSE(c, code, message)
+			writeResponsesFailedSSE(c, code, "", message)
 		}
 		return
 	}
@@ -49,6 +49,20 @@ func TestNormalizeCodexAutomationBootstrapSupportedLastRunValues(t *testing.T) {
 	}
 }

+func TestNormalizeCodexAutomationBootstrapHeartbeat(t *testing.T) {
+	output := `<heartbeat><automation_id>wiki</automation_id></heartbeat>`
+	got, changed := normalizeCodexAutomationBootstrap(codexAutomationBootstrapBody(t, output, ""))
+	require.True(t, changed)
+	require.Equal(t, "message", gjson.GetBytes(got, "input.0.type").String())
+	require.Equal(t, "user", gjson.GetBytes(got, "input.0.role").String())
+	require.Equal(t, output, gjson.GetBytes(got, "input.0.content.0.text").String())
+	require.False(t, gjson.GetBytes(got, "input.0.call_id").Exists())
+
+	again, changedAgain := normalizeCodexAutomationBootstrap(got)
+	require.False(t, changedAgain)
+	require.Equal(t, got, again)
+}
+
 func TestNormalizeCodexAutomationBootstrapRejectsUnsafeShapes(t *testing.T) {
 	validOutput := codexAutomationBootstrap("wiki", "never", automationBootstrapPrompt)
 	tests := []struct {
@@ -75,10 +89,6 @@ func TestNormalizeCodexAutomationBootstrapRejectsUnsafeShapes(t *testing.T) {
 			name: "real call context",
 			body: []byte(`{"model":"gpt-5","input":[{"type":"function_call_output","namespace":"codex_app","name":"automation_update","output":` + mustJSON(t, validOutput) + `},{"type":"function_call","call_id":"call-1"}]}`),
 		},
-		{
-			name: "heartbeat output",
-			body: codexAutomationBootstrapBody(t, `<heartbeat><automation_id>wiki</automation_id></heartbeat>`, ""),
-		},
 		{
 			name: "mismatched memory id",
 			body: codexAutomationBootstrapBody(t, strings.Replace(validOutput, "/wiki/memory.md", "/other/memory.md", 1), ""),
@@ -117,6 +127,31 @@ func TestNormalizeCodexAutomationBootstrapRejectsUnsafeShapes(t *testing.T) {
 	}
 }

+func TestNormalizeCodexAutomationBootstrapRejectsUnsafeHeartbeatShapes(t *testing.T) {
+	tests := []struct {
+		name   string
+		output string
+	}{
+		{name: "arbitrary tool output", output: `<result><automation_id>wiki</automation_id></result>`},
+		{name: "root attribute", output: `<heartbeat status="ok"><automation_id>wiki</automation_id></heartbeat>`},
+		{name: "namespaced root", output: `<heartbeat xmlns="urn:codex"><automation_id>wiki</automation_id></heartbeat>`},
+		{name: "extra child", output: `<heartbeat><automation_id>wiki</automation_id><status>ok</status></heartbeat>`},
+		{name: "nested id content", output: `<heartbeat><automation_id><value>wiki</value></automation_id></heartbeat>`},
+		{name: "padded id", output: `<heartbeat><automation_id> wiki </automation_id></heartbeat>`},
+		{name: "unsafe id", output: `<heartbeat><automation_id>../wiki</automation_id></heartbeat>`},
+		{name: "comment", output: `<heartbeat><!-- ok --><automation_id>wiki</automation_id></heartbeat>`},
+		{name: "trailing content", output: `<heartbeat><automation_id>wiki</automation_id></heartbeat>extra`},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			body := codexAutomationBootstrapBody(t, tt.output, "")
+			got, changed := normalizeCodexAutomationBootstrap(body)
+			require.False(t, changed)
+			require.Equal(t, body, got)
+		})
+	}
+}
+
 func TestNormalizeCodexAutomationBootstrapPreservesOrderAndIsIdempotent(t *testing.T) {
 	output := codexAutomationBootstrap("wiki", "never", automationBootstrapPrompt)
 	body := []byte(`{"model":"gpt-5","input":[{"type":"message","role":"user","content":"before"},{"type":"function_call_output","namespace":"codex_app","name":"automation_update","output":` + mustJSON(t, output) + `},{"type":"message","role":"user","content":"after"}]}`)
@@ -116,6 +116,7 @@ func (h *OpenAIGatewayHandler) ChatCompletions(c *gin.Context) {

 	// 解析渠道级模型映射
 	channelMapping, _ := h.gatewayService.ResolveChannelMappingAndRestrict(c.Request.Context(), apiKey.GroupID, reqModel)
+	forwardModel := openAIChannelForwardModel(channelMapping, reqModel)

 	if h.errorPassthroughService != nil {
 		service.BindErrorPassthroughService(c, h.errorPassthroughService)
@@ -170,7 +171,7 @@ func (h *OpenAIGatewayHandler) ChatCompletions(c *gin.Context) {
 			apiKey.GroupID,
 			"",
 			sessionHash,
-			reqModel,
+			forwardModel,
 			failedAccountIDs,
 			service.OpenAIUpstreamTransportAny,
 			service.OpenAIEndpointCapabilityChatCompletions,
@@ -16,7 +16,8 @@ import (
 // Codex CLI and the Codex desktop app refresh their model picker from
 // GET {base_url}/models?client_version=... (custom provider mode) or
 // GET /backend-api/codex/models (chatgpt_base_url mode). Both routes land
-// here. Groups with explicit account model mappings are generated locally;
+// here. Pinned discovery takes precedence over local account model mappings;
+// when disabled, groups with explicit mappings are generated locally;
 // otherwise ChatGPT manifests are proxied verbatim and custom API key manifests
 // receive provider-compatibility normalization plus short-lived caching.
 func (h *OpenAIGatewayHandler) CodexModels(c *gin.Context) {
@@ -34,28 +35,10 @@ func (h *OpenAIGatewayHandler) CodexModels(c *gin.Context) {
 	}

 	ifNoneMatch := c.GetHeader("If-None-Match")
-	configuredManifest, configured, err := h.gatewayService.BuildGroupConfiguredCodexModelsManifest(
-		c.Request.Context(),
-		apiKey.Group,
-		ifNoneMatch,
-	)
-	if err != nil {
-		if c.Request.Context().Err() != nil {
-			return
-		}
-		h.errorResponse(c, http.StatusInternalServerError, "api_error", "Failed to build Codex models manifest")
-		return
-	}
-	if configured {
-		writeCodexModelsManifestResponse(c, configuredManifest)
-		return
-	}
-
 	// 固定账号分支：开启后只用选定账号拉取 manifest，不经过调度器；
 	// 全部不可用/全部失败时按 FallbackToScheduler 决定回退调度器或返回错误。
 	if apiKey.Group.Platform == service.PlatformOpenAI &&
-		apiKey.Group.CodexModelsManifestConfig.Enabled &&
-		len(apiKey.Group.CodexModelsManifestConfig.AccountIDs) > 0 {
+		apiKey.Group.CodexModelsManifestConfig.Enabled {
 		pinnedManifest, pinnedAccount, pinnedErr := h.gatewayService.FetchPinnedCodexModelsManifest(
 			c.Request.Context(),
 			apiKey.Group,
@@ -84,7 +67,26 @@ func (h *OpenAIGatewayHandler) CodexModels(c *gin.Context) {
 			if c.Request.Context().Err() != nil {
 				return
 			}
-			writeCodexModelsManifestResponse(c, pinnedManifest)
+			writeOpenAIModelsResponse(c, pinnedManifest)
+			return
+		}
+	}
+
+	if !apiKey.Group.CodexModelsManifestConfig.Enabled {
+		configuredManifest, configured, err := h.gatewayService.BuildGroupConfiguredCodexModelsManifest(
+			c.Request.Context(),
+			apiKey.Group,
+			ifNoneMatch,
+		)
+		if err != nil {
+			if c.Request.Context().Err() != nil {
+				return
+			}
+			h.errorResponse(c, http.StatusInternalServerError, "api_error", "Failed to build Codex models manifest")
+			return
+		}
+		if configured {
+			writeOpenAIModelsResponse(c, configuredManifest)
 			return
 		}
 	}
@@ -133,6 +135,10 @@ func (h *OpenAIGatewayHandler) CodexModels(c *gin.Context) {
 			h.errorResponse(c, http.StatusInternalServerError, "api_error", "Failed to complete Codex models manifest")
 			return
 		}
+		if err := service.ApplyPinnedCodexModelsMapping(manifest, account, apiKey.Group); err != nil {
+			h.errorResponse(c, http.StatusInternalServerError, "api_error", "Failed to apply model mappings")
+			return
+		}
 		if err := h.gatewayService.MergeGroupConfiguredCodexModels(c.Request.Context(), apiKey.Group, manifest, ifNoneMatch); err != nil {
 			h.errorResponse(c, http.StatusInternalServerError, "api_error", "Failed to build Codex models manifest")
 			return
@@ -141,19 +147,7 @@ func (h *OpenAIGatewayHandler) CodexModels(c *gin.Context) {
 			return
 		}

-		writeCodexModelsManifestResponse(c, manifest)
+		writeOpenAIModelsResponse(c, manifest)
 		return
 	}
 }
-
-func writeCodexModelsManifestResponse(c *gin.Context, manifest *service.CodexModelsManifest) {
-	if manifest.ETag != "" {
-		c.Header("ETag", manifest.ETag)
-	}
-	if manifest.NotModified {
-		c.Status(http.StatusNotModified)
-		c.Writer.WriteHeaderNow()
-		return
-	}
-	c.Data(http.StatusOK, "application/json", manifest.Body)
-}
@@ -43,11 +43,7 @@ func TestNormalizeCodexDelegationBootstrapRejectsUnsafeShapes(t *testing.T) {
 	}{
 		{name: "ordinary missing call id", item: `{"type":"function_call_output","name":"other","namespace":"codex_app","output":"` + delegationEnvelope + `"}`},
 		{name: "valid call id", item: `{"type":"function_call_output","call_id":"call-1","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"}`},
-		{name: "previous response", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"}`, extra: `,"previous_response_id":"resp-1"`},
-		{name: "real call context", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"function_call","call_id":"call-1"}`},
 		{name: "ambiguous call context", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"function_call"}`},
-		{name: "item reference context", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"item_reference","id":"call-1"}`},
-		{name: "built-in call context", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"computer_call","call_id":"call-1"}`},
 		{name: "ambiguous built-in call context", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"computer_call"}`},
 		{name: "mixed missing call id output", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"computer_call_output","output":"done"}`},
 		{name: "mixed tool search output", item: `{"type":"function_call_output","name":"create_thread","namespace":"codex_app","output":"` + delegationEnvelope + `"},{"type":"tool_search_output","output":"done"}`},
@@ -67,6 +63,18 @@ func TestNormalizeCodexDelegationBootstrapRejectsUnsafeShapes(t *testing.T) {
 	}
 }

+func TestNormalizeCodexDelegationBootstrapWithHistoricalContext(t *testing.T) {
+	body := []byte(`{"model":"gpt-5","previous_response_id":"resp-1","input":[{"type":"function_call_output","namespace":"codex_app","name":"send_message_to_thread","output":"` + delegationEnvelope + `"},{"type":"function_call","call_id":"call-1"},{"type":"function_call_output","call_id":"call-1","output":"done"},{"type":"item_reference","id":"item-1"},{"type":"computer_call","call_id":"call-2"}]}`)
+
+	got, changed := normalizeCodexDelegationBootstrap(body)
+	require.True(t, changed)
+	require.Equal(t, "message", gjson.GetBytes(got, "input.0.type").String())
+	require.Equal(t, "user", gjson.GetBytes(got, "input.0.role").String())
+	require.Equal(t, "resp-1", gjson.GetBytes(got, "previous_response_id").String())
+	require.Equal(t, "call-1", gjson.GetBytes(got, "input.2.call_id").String())
+	require.Equal(t, "item-1", gjson.GetBytes(got, "input.3.id").String())
+}
+
 func TestNormalizeCodexDelegationBootstrapPreviousResponseID(t *testing.T) {
 	tests := []struct {
 		name    string
@@ -75,7 +83,7 @@ func TestNormalizeCodexDelegationBootstrapPreviousResponseID(t *testing.T) {
 	}{
 		{name: "absent", changed: true},
 		{name: "blank string", value: `,"previous_response_id":"  "`, changed: true},
-		{name: "non-empty string", value: `,"previous_response_id":"resp-1"`},
+		{name: "non-empty string", value: `,"previous_response_id":"resp-1"`, changed: true},
 		{name: "null", value: `,"previous_response_id":null`},
 		{name: "number", value: `,"previous_response_id":42`},
 		{name: "boolean", value: `,"previous_response_id":false`},
@@ -85,6 +85,7 @@ func (h *OpenAIGatewayHandler) Embeddings(c *gin.Context) {
 	}

 	channelMapping, _ := h.gatewayService.ResolveChannelMappingAndRestrict(c.Request.Context(), apiKey.GroupID, reqModel)
+	forwardModel := openAIChannelForwardModel(channelMapping, reqModel)

 	subscription, _ := middleware2.GetSubscriptionFromContext(c)
 	service.SetOpsLatencyMs(c, service.OpsAuthLatencyMsKey, time.Since(requestStart).Milliseconds())
@@ -127,7 +128,7 @@ func (h *OpenAIGatewayHandler) Embeddings(c *gin.Context) {
 			apiKey.GroupID,
 			"",
 			"",
-			reqModel,
+			forwardModel,
 			failedAccountIDs,
 			service.OpenAIUpstreamTransportHTTPSSE,
 			service.OpenAIEndpointCapabilityEmbeddings,
@@ -128,6 +128,44 @@ func TestOpenAICapacityFailoverExhaustionPreservesMessageAsServerError(t *testin
 	})
 }

+func TestOpenAIManagedSingleAccountModelNotFoundExhaustionPreservesStructured400(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+	body := []byte(`{"error":{"type":"invalid_request_error","code":"model_not_found","param":"model","message":"The requested model is unavailable on this account"}}`)
+
+	(&OpenAIGatewayHandler{}).handleFailoverExhausted(c, &service.UpstreamFailoverError{
+		StatusCode:   http.StatusBadRequest,
+		ResponseBody: body,
+	}, false)
+
+	require.Equal(t, http.StatusBadRequest, recorder.Code)
+	require.Equal(t, "invalid_request_error", gjson.Get(recorder.Body.String(), "error.type").String())
+	require.Equal(t, "model_not_found", gjson.Get(recorder.Body.String(), "error.code").String())
+	require.Equal(t, "model", gjson.Get(recorder.Body.String(), "error.param").String())
+	require.Equal(t, "The requested model is unavailable on this account", gjson.Get(recorder.Body.String(), "error.message").String())
+}
+
+func TestOpenAIManagedModelNotFoundExhaustionSanitizesMessage(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+	body := []byte(`{"error":{"code":"model_not_found","message":"Model not found; inspect https://upstream.example/debug?access_token=super-secret-value&model=x"}}`)
+
+	(&OpenAIGatewayHandler{}).handleFailoverExhausted(c, &service.UpstreamFailoverError{
+		StatusCode:   http.StatusBadRequest,
+		ResponseBody: body,
+	}, false)
+
+	require.Equal(t, http.StatusBadRequest, recorder.Code)
+	require.NotContains(t, recorder.Body.String(), "super-secret-value")
+	require.Contains(t, gjson.Get(recorder.Body.String(), "error.message").String(), "access_token=***")
+	recorded, ok := c.Get(service.OpsUpstreamErrorMessageKey)
+	require.True(t, ok)
+	require.NotContains(t, recorded, "super-secret-value")
+	require.Contains(t, recorded, "access_token=***")
+}
+
 func TestResponsesFailoverExhaustedAfterForwardedTerminalMarksOpsWithoutDuplicateFrame(t *testing.T) {
 	gin.SetMode(gin.TestMode)
 	recorder := httptest.NewRecorder()
@@ -1,6 +1,7 @@
 package handler

 import (
+	"net/http"
 	"net/http/httptest"
 	"strings"
 	"testing"
@@ -102,6 +103,44 @@ func TestClearCyberPolicyTurnState(t *testing.T) {
 	require.Equal(t, "turn2", service.GetOpsCyberPolicy(c).Message)
 }

+func TestAdvanceOpenAIWSCyberBlockStateDefersBlockAcrossFailover(t *testing.T) {
+	failoverErr := &service.UpstreamFailoverError{StatusCode: http.StatusTooManyRequests}
+
+	blocked, pending := advanceOpenAIWSCyberBlockState(false, false, true, failoverErr)
+	require.False(t, blocked, "the replacement account must receive the current turn")
+	require.True(t, pending, "the cyber hit must still block later turns")
+
+	blocked, pending = advanceOpenAIWSCyberBlockState(blocked, pending, false, failoverErr)
+	require.False(t, blocked, "additional failover attempts must remain eligible")
+	require.True(t, pending)
+
+	blocked, pending = advanceOpenAIWSCyberBlockState(blocked, pending, false, nil)
+	require.True(t, blocked, "the next client turn must be blocked after failover finishes")
+	require.False(t, pending)
+}
+
+func TestClearCyberPolicyAttemptStatePreservesRecordedGuardDuringFailover(t *testing.T) {
+	c := newTestGinContext()
+	h := &OpenAIGatewayHandler{}
+
+	service.MarkOpsCyberPolicy(c, service.CyberPolicyMark{Message: "account-a", UpstreamStatus: http.StatusOK})
+	h.recordCyberPolicyIfMarked(c, nil, nil, nil, "gpt-5", true, nil, service.ChannelUsageFields{}, "")
+	require.True(t, c.GetBool(cyberPolicyRecordedKey))
+
+	clearCyberPolicyAttemptState(c, false)
+	require.Nil(t, service.GetOpsCyberPolicy(c))
+	require.True(t, c.GetBool(cyberPolicyRecordedKey), "the same logical turn must not record again after failover")
+
+	service.MarkOpsCyberPolicy(c, service.CyberPolicyMark{Message: "account-b", UpstreamStatus: http.StatusOK})
+	h.recordCyberPolicyIfMarked(c, nil, nil, nil, "gpt-5", true, nil, service.ChannelUsageFields{}, "")
+	require.Equal(t, "account-b", service.GetOpsCyberPolicy(c).Message)
+	require.True(t, c.GetBool(cyberPolicyRecordedKey))
+
+	clearCyberPolicyAttemptState(c, true)
+	require.Nil(t, service.GetOpsCyberPolicy(c))
+	require.False(t, c.GetBool(cyberPolicyRecordedKey), "a completed logical turn must reset the guard")
+}
+
 // TestBuildCyberSessionBlockedOpsEntry verifies the locally-rejected request is
 // auditable: 403 / phase=request / type=cyber_policy_session_blocked — distinct
 // from upstream cyber_policy hits, and it must NOT touch moderation/violation.
@@ -54,6 +54,21 @@ type openAIWSTurnChannelMappingSnapshot struct {
 	mapping service.ChannelMappingResult
 }

+func advanceOpenAIWSCyberBlockState(blocked, pending, marked bool, turnErr error) (bool, bool) {
+	var failoverErr *service.UpstreamFailoverError
+	isFailover := errors.As(turnErr, &failoverErr)
+	if marked {
+		if isFailover {
+			return false, true
+		}
+		return true, false
+	}
+	if pending && !isFailover {
+		return true, false
+	}
+	return blocked, pending
+}
+
 var errOpenAIWSUnsupportedModelSwitch = errors.New("selected account does not support websocket model switch")

 func newOpenAIWSUnsupportedModelSwitchError(model string) error {
@@ -141,6 +156,18 @@ func openAIWSTurnBillingModel(result *service.OpenAIForwardResult, mapping servi
 	return billingModel
 }

+// openAIChannelForwardModel returns the model used for upstream capability
+// routing. The client-requested model remains separate for logs, billing, and
+// responses.
+func openAIChannelForwardModel(mapping service.ChannelMappingResult, requestedModel string) string {
+	if mapping.Mapped {
+		if mappedModel := strings.TrimSpace(mapping.MappedModel); mappedModel != "" {
+			return mappedModel
+		}
+	}
+	return requestedModel
+}
+
 type grokMediaEligibilityProber interface {
 	ProbeMediaEligibility(ctx context.Context, accountID int64) (bool, string, error)
 }
@@ -534,10 +561,7 @@ func (h *OpenAIGatewayHandler) Responses(c *gin.Context) {
 	channelMapping, _ := h.gatewayService.ResolveChannelMappingAndRestrict(c.Request.Context(), apiKey.GroupID, reqModel)
 	forwardBody := openAIModelMappedBody(body, channelMapping.Mapped, channelMapping.MappedModel, h.gatewayService.ReplaceModelInBody)
 	seedOpenAIForwardImageIntentHint(c, channelMapping.Mapped, imageIntent)
-	forwardModel := reqModel
-	if channelMapping.Mapped {
-		forwardModel = channelMapping.MappedModel
-	}
+	forwardModel := openAIChannelForwardModel(channelMapping, reqModel)
 	c.Request = c.Request.WithContext(service.WithOpenAIForwardModel(
 		c.Request.Context(),
 		forwardModel,
@@ -630,7 +654,7 @@ func (h *OpenAIGatewayHandler) Responses(c *gin.Context) {
 			apiKey.GroupID,
 			previousResponseID,
 			sessionHash,
-			reqModel,
+			forwardModel,
 			failedAccountIDs,
 			service.OpenAIUpstreamTransportAny,
 			requiredCapability,
@@ -1601,14 +1625,17 @@ func (h *OpenAIGatewayHandler) validateFunctionCallOutputRequest(c *gin.Context,
 }

 func normalizeCodexDelegationBootstrap(body []byte) ([]byte, bool) {
-	return normalizeCodexCallOutputBootstrap(body, isCodexDelegationCandidate)
+	// 已有任务通过 send_message_to_thread 唤醒时会携带 previous_response_id；
+	// 完整历史回放还会带有已配对的调用项。delegation 仍是客户端注入的用户输入，
+	// 不属于这些历史调用的结果，因此允许它与可明确配对的历史上下文共存。
+	return normalizeCodexCallOutputBootstrap(body, isCodexDelegationCandidate, true)
 }

 func normalizeCodexAutomationBootstrap(body []byte) ([]byte, bool) {
-	return normalizeCodexCallOutputBootstrap(body, isCodexAutomationCandidate)
+	return normalizeCodexCallOutputBootstrap(body, isCodexAutomationCandidate, false)
 }

-func normalizeCodexCallOutputBootstrap(body []byte, isCandidate func(map[string]any) bool) ([]byte, bool) {
+func normalizeCodexCallOutputBootstrap(body []byte, isCandidate func(map[string]any) bool, allowHistoricalContext bool) ([]byte, bool) {
 	if !hasUniqueJSONMembers(body) {
 		return body, false
 	}
@@ -1620,7 +1647,7 @@ func normalizeCodexCallOutputBootstrap(body []byte, isCandidate func(map[string]
 	}
 	if previousResponseID, exists := request["previous_response_id"]; exists {
 		value, ok := previousResponseID.(string)
-		if !ok || strings.TrimSpace(value) != "" {
+		if !ok || (!allowHistoricalContext && strings.TrimSpace(value) != "") {
 			return body, false
 		}
 	}
@@ -1629,27 +1656,35 @@ func normalizeCodexCallOutputBootstrap(body []byte, isCandidate func(map[string]
 		return body, false
 	}

-	// Any call/reference anchor makes a call-less output ambiguous. Responses
-	// built-ins follow the *_call / *_call_output naming convention, so classify
-	// by the wire type shape instead of maintaining an incomplete allowlist.
+	// Responses built-ins follow the *_call / *_call_output naming convention,
+	// so classify by the wire type shape instead of maintaining an incomplete
+	// allowlist. Delegation may coexist with historical anchors only when their
+	// IDs make them unambiguous; automation retains the bootstrap-only boundary.
 	for _, raw := range input {
 		item, ok := raw.(map[string]any)
 		if !ok {
 			continue
 		}
 		typ := stringField(item, "type")
-		if typ == "item_reference" || strings.HasSuffix(typ, "_call") {
-			return body, false
-		}
-		if isResponsesCallOutputType(typ) {
+		if isCandidate(item) {
 			callIDValue, exists := item["call_id"]
 			callID, isString := callIDValue.(string)
 			if exists && (!isString || strings.TrimSpace(callID) != "") {
 				return body, false
 			}
-			if !isCandidate(item) {
-				return body, false
+			continue
+		}
+		if typ == "item_reference" {
+			if allowHistoricalContext && strings.TrimSpace(stringField(item, "id")) != "" {
+				continue
 			}
+			return body, false
+		}
+		if strings.HasSuffix(typ, "_call") || isResponsesCallOutputType(typ) {
+			if allowHistoricalContext && strings.TrimSpace(stringField(item, "call_id")) != "" {
+				continue
+			}
+			return body, false
 		}
 	}

@@ -1757,7 +1792,7 @@ func isCodexAutomationCandidate(item map[string]any) bool {
 		return false
 	}
 	output, ok := item["output"].(string)
-	return ok && validCodexAutomationBootstrap(output)
+	return ok && (validCodexAutomationBootstrap(output) || validCodexAutomationHeartbeat(output))
 }

 func stringField(item map[string]any, key string) string {
@@ -1835,6 +1870,56 @@ func validCodexAutomationLastRun(value string) bool {
 	return err == nil && runAt.UnixMilli() == epochMillis
 }

+func validCodexAutomationHeartbeat(value string) bool {
+	decoder := xml.NewDecoder(strings.NewReader(value))
+	var rootSeen, automationIDSeen bool
+	var automationID bytes.Buffer
+	depth := 0
+	for {
+		token, err := decoder.Token()
+		if err == io.EOF {
+			id := automationID.String()
+			return rootSeen && automationIDSeen && depth == 0 &&
+				strings.TrimSpace(id) == id && validCodexAutomationID(id)
+		}
+		if err != nil {
+			return false
+		}
+		switch current := token.(type) {
+		case xml.StartElement:
+			depth++
+			if current.Name.Space != "" || len(current.Attr) != 0 || depth > 2 {
+				return false
+			}
+			if depth == 1 {
+				if rootSeen || current.Name.Local != "heartbeat" {
+					return false
+				}
+				rootSeen = true
+			} else if automationIDSeen || current.Name.Local != "automation_id" {
+				return false
+			}
+			automationIDSeen = depth == 2
+		case xml.EndElement:
+			if current.Name.Space != "" {
+				return false
+			}
+			depth--
+			if depth < 0 {
+				return false
+			}
+		case xml.CharData:
+			if depth == 2 {
+				_, _ = automationID.Write(current)
+			} else if len(bytes.TrimSpace(current)) != 0 {
+				return false
+			}
+		case xml.Comment, xml.ProcInst, xml.Directive:
+			return false
+		}
+	}
+}
+
 func validCodexDelegationEnvelope(value string) bool {
 	decoder := xml.NewDecoder(strings.NewReader(value))
 	var rootSeen, sourceSeen, inputSeen bool
@@ -1941,10 +2026,10 @@ const (
 // 由 BeforeTurn 在每个 turn 开始时冻结，AfterTurn 的用量提交读取它；turn 在
 // 连接内串行推进，互斥锁只为跨用量提交 goroutine 的读取安全。
 //
-// ws_v2 passthrough ingress 没有 BeforeTurn，因此本值会保持零；AfterTurn 必须
-// 以 TurnStarted 已记录的所属 turn 开始时刻为回退，而不是用建连或记录时刻。
-// 这样每个 passthrough turn 都按自己的开始时刻计价，但不改变其仅在建连时执行
-// 准入门、没有 turn 级利润复核的既有行为。
+// 零值语义（重要）：首轮准入由握手路径完成，不调用 BeforeTurn，因此首轮保持
+// 零值并回退到 TurnStarted 记录的首轮开始时刻。后续 turn 在 response.create
+// 写入上游前调用 BeforeTurn，按当时的利润门复核并冻结定价。绝不能用建连时刻
+// 初始化，否则会把长连接的所有 turn 钉死在建连时的峰谷因子。
 type openAIWSTurnPricing struct {
 	mu sync.Mutex
 	at time.Time
@@ -2005,7 +2090,7 @@ func (h *OpenAIGatewayHandler) acquireResponsesAccountSlot(
 	return h.acquireOpenAIAccountSlot(c, groupID, sessionHash, selection, reqStream, streamStarted, reqLog, nil)
 }

-type openAISlotErrorWriter func(status int, errType, message string)
+type openAISlotErrorWriter func(status int, errType, code, message string)

 // acquireOpenAIAccountSlot centralizes scheduler selection admission. The
 // optional error writer lets non-Responses endpoints retain their wire format
@@ -2021,13 +2106,13 @@ func (h *OpenAIGatewayHandler) acquireOpenAIAccountSlot(
 	writeError openAISlotErrorWriter,
 ) (func(), openAISlotAcquireResult) {
 	if writeError == nil {
-		writeError = func(status int, errType, message string) {
-			h.handleStreamingAwareError(c, status, errType, message, *streamStarted)
+		writeError = func(status int, errType, code, message string) {
+			h.handleStreamingAwareErrorWithCode(c, status, errType, code, message, *streamStarted, false)
 		}
 	}
 	if selection == nil || selection.Account == nil {
 		markOpsRoutingCapacityLimited(c)
-		writeError(http.StatusServiceUnavailable, "api_error", "No available accounts")
+		writeError(http.StatusServiceUnavailable, "api_error", "", "No available accounts")
 		return nil, openAISlotAcquireFailed
 	}

@@ -2057,7 +2142,7 @@ func (h *OpenAIGatewayHandler) acquireOpenAIAccountSlot(
 	}
 	if selection.WaitPlan == nil {
 		markOpsRoutingCapacityLimited(c)
-		writeError(http.StatusServiceUnavailable, "api_error", "No available accounts")
+		writeError(http.StatusServiceUnavailable, "api_error", "", "No available accounts")
 		return nil, openAISlotAcquireFailed
 	}

@@ -2068,8 +2153,8 @@ func (h *OpenAIGatewayHandler) acquireOpenAIAccountSlot(
 	)
 	if err != nil {
 		reqLog.Warn("openai.account_slot_quick_acquire_failed", zap.Int64("account_id", account.ID), zap.Error(err))
-		status, errType, message := concurrencyErrorResponse(err, "account")
-		writeError(status, errType, message)
+		status, errType, code, message := concurrencyErrorResponse(err, "account")
+		writeError(status, errType, code, message)
 		return nil, openAISlotAcquireFailed
 	}
 	if fastAcquired {
@@ -2099,7 +2184,7 @@ func (h *OpenAIGatewayHandler) acquireOpenAIAccountSlot(
 			zap.Int64("account_id", account.ID),
 			zap.Int("max_waiting", selection.WaitPlan.MaxWaiting),
 		)
-		writeError(http.StatusTooManyRequests, "rate_limit_error", "Too many pending requests, please retry later")
+		writeError(http.StatusTooManyRequests, "rate_limit_error", gatewayQueueFullCode, "Too many pending requests, please retry later")
 		return nil, openAISlotAcquireFailed
 	}

@@ -2122,8 +2207,8 @@ func (h *OpenAIGatewayHandler) acquireOpenAIAccountSlot(
 	)
 	if err != nil {
 		reqLog.Warn("openai.account_slot_acquire_failed", zap.Int64("account_id", account.ID), zap.Error(err))
-		status, errType, message := concurrencyErrorResponse(err, "account")
-		writeError(status, errType, message)
+		status, errType, code, message := concurrencyErrorResponse(err, "account")
+		writeError(status, errType, code, message)
 		return nil, openAISlotAcquireFailed
 	}

@@ -2321,6 +2406,7 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 		return
 	}
 	cyberBlockedThisConn := false
+	cyberBlockPendingAfterFailover := false
 	var cyberTurnBodiesMu sync.Mutex
 	cyberTurnBodies := map[int][]byte{1: append([]byte(nil), firstMessage...)}
 	setCyberTurnBody := func(turn int, payload []byte) {
@@ -2338,10 +2424,7 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {

 	// 解析渠道级模型映射
 	channelMappingWS, _ := h.gatewayService.ResolveChannelMappingAndRestrict(ctx, apiKey.GroupID, reqModel)
-	wsForwardModel := reqModel
-	if channelMappingWS.Mapped && strings.TrimSpace(channelMappingWS.MappedModel) != "" {
-		wsForwardModel = strings.TrimSpace(channelMappingWS.MappedModel)
-	}
+	wsForwardModel := openAIChannelForwardModel(channelMappingWS, reqModel)

 	var currentUserRelease func()
 	var currentAccountRelease func()
@@ -2504,7 +2587,7 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 			apiKey.GroupID,
 			previousResponseID,
 			sessionHash,
-			reqModel,
+			wsForwardModel,
 			failedAccountIDs,
 			requiredTransport,
 			requiredCapability,
@@ -2656,8 +2739,8 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 		// turn-tagged slot preserves the exact mapping used for the in-flight request.
 		var turnChannelMapping atomic.Pointer[openAIWSTurnChannelMappingSnapshot]
 		turnChannelMapping.Store(&openAIWSTurnChannelMappingSnapshot{turn: 1, mapping: channelMappingWS})
-		// turn 级定价：BeforeTurn 重新冻结 pricingAt 并按最新门复核当前账号；
-		// passthrough 没有 BeforeTurn 时，AfterTurn 回退到 TurnStarted 的所属 turn 时刻。
+		// turn 级定价：首轮回退到 TurnStarted 的所属 turn 时刻；后续 turn 由
+		// BeforeTurn 重新冻结 pricingAt 并按最新门复核当前账号。
 		var turnPricing openAIWSTurnPricing
 		hooks := &service.OpenAIWSIngressHooks{
 			ClientLifecycleContext:      clientLifecycleCtx,
@@ -2671,10 +2754,9 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 				c.Set(securityAuditWSTurnContextKey, turn)
 				service.BeginOpsStreamTurn(c, turn)
 				setCyberTurnBody(turn, payload)
-				// Passthrough ingress intentionally skips BeforeTurn, so enforce only
-				// the connection-level cyber session gate here as well. Native ingress
-				// visits this hook first and gets the same side-effect-free close error;
-				// its original BeforeTurn guard remains as defense in depth.
+				// 连接级 cyber session gate 也在 BeforeRequest 先执行，使 native 与
+				// passthrough ingress 都能在 BeforeTurn 及上游写入前无副作用地拒绝。
+				// BeforeTurn 中保留同一检查作为防御式兜底。
 				if cyberBlockedThisConn {
 					return service.NewOpenAIWSClientCloseError(coderws.StatusPolicyViolation, cyberSessionBlockedClientMsg, nil)
 				}
@@ -2775,10 +2857,12 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 			AfterTurn: func(turn int, result *service.OpenAIForwardResult, turnErr error) {
 				turnStart := getTurnStart(turn)
 				cyberBlockBody := takeCyberTurnBody(turn)
-				// F1: cyber 标记按 turn 生命周期清理——defer 保证任意早返回路径都执行；
-				// CyberBlocked 必须在 submit 前同步预捕获（task 闭包由 worker 池异步执行，
-				// 届时 defer 已清除标记）。
-				defer clearCyberPolicyTurnState(c)
+				// 每次 attempt 都清 cyber mark；failover 链结束前保留 recorded guard，
+				// 避免同一逻辑 turn 换号后重复落风控。CyberBlocked 必须在 submit 前
+				// 同步预捕获（task 闭包由 worker 池异步执行，届时 mark 已清除）。
+				defer func() {
+					clearCyberPolicyAttemptState(c, !cyberBlockPendingAfterFailover)
+				}()
 				releaseTurnSlots()
 				turnRequestedModel := reqModel
 				turnUpstreamModel := ""
@@ -2800,10 +2884,14 @@ func (h *OpenAIGatewayHandler) ResponsesWebSocket(c *gin.Context) {
 					turnUpstreamModel = turnRequestedModel
 				}
 				turnUsageFields := turnMapping.ToUsageFields(turnRequestedModel, turnUpstreamModel)
+				cyberMarked := service.GetOpsCyberPolicy(c) != nil
 				h.recordCyberPolicyIfMarked(c, apiKey, account, subscription, turnRequestedModel, turnErr != nil, cyberBlockBody, turnUsageFields, requestPayloadHash)
-				if service.GetOpsCyberPolicy(c) != nil {
-					cyberBlockedThisConn = true
-				}
+				cyberBlockedThisConn, cyberBlockPendingAfterFailover = advanceOpenAIWSCyberBlockState(
+					cyberBlockedThisConn,
+					cyberBlockPendingAfterFailover,
+					cyberMarked,
+					turnErr,
+				)
 				if turnErr != nil {
 					if result == nil || result.ImageCount <= 0 {
 						return
@@ -3194,14 +3282,14 @@ func (h *OpenAIGatewayHandler) acquireImageGenerationSlot(c *gin.Context, stream
 	if acquired {
 		return release, true
 	}
-	h.handleStreamingAwareError(c, http.StatusTooManyRequests, "rate_limit_error", "Image generation concurrency limit exceeded, please retry later", streamStarted)
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error", gatewayConcurrencyLimitCode, "Image generation concurrency limit exceeded, please retry later", streamStarted, false)
 	return nil, false
 }

 // handleConcurrencyError handles concurrency-related acquire errors.
 func (h *OpenAIGatewayHandler) handleConcurrencyError(c *gin.Context, err error, slotType string, streamStarted bool) {
-	status, errType, message := concurrencyErrorResponse(err, slotType)
-	h.handleStreamingAwareError(c, status, errType, message, streamStarted)
+	status, errType, code, message := concurrencyErrorResponse(err, slotType)
+	h.handleStreamingAwareErrorWithCode(c, status, errType, code, message, streamStarted, false)
 }

 func (h *OpenAIGatewayHandler) handleFailoverExhausted(c *gin.Context, failoverErr *service.UpstreamFailoverError, streamStarted bool) {
@@ -3244,6 +3332,12 @@ func (h *OpenAIGatewayHandler) handleFailoverExhausted(c *gin.Context, failoverE
 	}
 	statusCode := failoverErr.StatusCode
 	responseBody := failoverErr.ResponseBody
+	if statusCode == http.StatusBadRequest && service.IsOpenAICompatibleModelNotFound400(responseBody) && !streamStarted {
+		upstreamMsg := service.SanitizeUpstreamErrorMessage(service.ExtractUpstreamErrorMessage(responseBody))
+		service.SetOpsUpstreamError(c, statusCode, upstreamMsg, "")
+		service.WriteOpenAIUpstreamClientError(c, statusCode, responseBody, upstreamMsg)
+		return
+	}
 	if service.IsOpenAISilentRefusalErrorBody(responseBody) {
 		service.SetOpsUpstreamError(c, statusCode, service.OpenAISilentRefusalClientMessage(), "")
 		h.handleStreamingAwareError(c, http.StatusBadGateway, "upstream_error", service.OpenAISilentRefusalClientMessage(), streamStarted)
@@ -3381,7 +3475,7 @@ func (h *OpenAIGatewayHandler) handleStreamingAwareErrorWithCode(
 		// 通用 `event: error` 帧不被识别为终止事件，会导致
 		// "stream closed before response.completed"。
 		if inboundIsResponses(c) {
-			if writeResponsesFailedSSE(c, errType, message) {
+			if writeResponsesFailedSSE(c, errType, code, message) {
 				return
 			}
 		}
@@ -3548,7 +3642,7 @@ func (h *OpenAIGatewayHandler) errorResponse(c *gin.Context, status int, errType
 	// 提交的 SSE 流交错，必须降级为 response.failed 终止事件（#3887）。
 	if service.StopOpenAICompactSSEKeepaliveCommitted(c) {
 		service.MarkOpsStreamError(c, errType, message, status)
-		if writeResponsesFailedSSE(c, errType, message) {
+		if writeResponsesFailedSSE(c, errType, "", message) {
 			return
 		}
 	}
@@ -3881,7 +3975,7 @@ func (h *OpenAIGatewayHandler) rejectIfCyberSessionBlocked(c *gin.Context, apiKe
 	// 写 JSON（#3887）。
 	if service.StopOpenAICompactSSEKeepaliveCommitted(c) {
 		service.MarkOpsStreamError(c, "permission_error", cyberSessionBlockedClientMsg, http.StatusForbidden)
-		if writeResponsesFailedSSE(c, "permission_error", cyberSessionBlockedClientMsg) {
+		if writeResponsesFailedSSE(c, "permission_error", "", cyberSessionBlockedClientMsg) {
 			h.enqueueCyberSessionBlockedOpsEntry(c, apiKey, model, key)
 			return true
 		}
@@ -4125,15 +4219,20 @@ func (h *OpenAIGatewayHandler) recordCyberPolicyIfMarked(c *gin.Context, apiKey
 	}()
 }

-// clearCyberPolicyTurnState resets the cyber mark and the per-request recorded
-// guard. WS-only: called at the END of AfterTurn, after recordCyberPolicyIfMarked
-// and RecordUsage (which reads CyberBlocked) have both consumed the mark.
+// clearCyberPolicyTurnState resets the cyber mark and recorded guard after a
+// logical WS turn has finished.
 func clearCyberPolicyTurnState(c *gin.Context) {
+	clearCyberPolicyAttemptState(c, true)
+}
+
+func clearCyberPolicyAttemptState(c *gin.Context, resetRecorded bool) {
 	if c == nil {
 		return
 	}
 	service.ClearOpsCyberPolicy(c)
-	c.Set(cyberPolicyRecordedKey, false)
+	if resetRecorded {
+		c.Set(cyberPolicyRecordedKey, false)
+	}
 }

 func summarizeWSCloseErrorForLog(err error) (string, string) {
@@ -1530,7 +1530,7 @@ func TestOpenAIResponsesWebSocket_PassthroughTracksModelPerTurn(t *testing.T) {
 		"each turn must be billed with its own channel-mapped model")
 }

-func TestOpenAIResponsesWebSocket_UnchangedChannelTargetOutsideAccountMappingKeysRemainsValid(t *testing.T) {
+func TestOpenAIResponsesWebSocket_ChannelMappedTargetSelectsAccountWithoutRequestedAlias(t *testing.T) {
 	got := runOpenAIResponsesWebSocketUsageLogCase(t, openAIResponsesWSUsageLogCase{
 		firstPayload:  `{"type":"response.create","model":"public-alias","stream":false}`,
 		secondPayload: `{"type":"response.create","stream":false}`,
@@ -1538,7 +1538,7 @@ func TestOpenAIResponsesWebSocket_UnchangedChannelTargetOutsideAccountMappingKey
 			"public-alias": "gpt-5.6-sol",
 		},
 		accountModelMapping: map[string]any{
-			"public-alias": "gpt-5.6-terra",
+			"gpt-5.6-sol": "gpt-5.6-sol",
 		},
 	})

@@ -1710,6 +1710,40 @@ func TestOpenAIAccountScheduleModelUsesActualOrSharedResolver(t *testing.T) {
 	require.Equal(t, "attempt-actual", openAIAccountScheduleModel(c, account, "public", true, nil))
 }

+func TestOpenAIChannelForwardModelForScheduler(t *testing.T) {
+	tests := []struct {
+		name      string
+		mapping   service.ChannelMappingResult
+		requested string
+		want      string
+	}{
+		{
+			name:      "mapped model is used for capability filtering",
+			mapping:   service.ChannelMappingResult{Mapped: true, MappedModel: "  gpt-forward  "},
+			requested: "client-alias",
+			want:      "gpt-forward",
+		},
+		{
+			name:      "unmapped request preserves client model",
+			mapping:   service.ChannelMappingResult{MappedModel: "client-model"},
+			requested: "client-model",
+			want:      "client-model",
+		},
+		{
+			name:      "empty mapped model safely preserves client model",
+			mapping:   service.ChannelMappingResult{Mapped: true, MappedModel: "  "},
+			requested: "client-model",
+			want:      "client-model",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			require.Equal(t, tt.want, openAIChannelForwardModel(tt.mapping, tt.requested))
+		})
+	}
+}
+
 func TestShouldReportOpenAIWSProxyAccountFailure(t *testing.T) {
 	t.Run("unsupported client model switch does not penalize account", func(t *testing.T) {
 		err := fmt.Errorf("wrapped ingress turn: %w", newOpenAIWSUnsupportedModelSwitchError("gpt-unsupported"))
@@ -0,0 +1,52 @@
+package handler
+
+import (
+	"errors"
+	"net/http"
+
+	infraerrors "github.com/Wei-Shaw/sub2api/internal/pkg/errors"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+)
+
+func (h *GatewayHandler) pinnedOpenAIModels(c *gin.Context, group *service.Group) {
+	if c.Request.Context().Err() != nil {
+		return
+	}
+	if h.openAIGatewayService == nil {
+		writeOpenAIModelsError(c, http.StatusInternalServerError, "api_error", "OpenAI model discovery is not configured")
+		return
+	}
+	response, account, err := h.openAIGatewayService.FetchPinnedOpenAIModelsList(
+		c.Request.Context(), group, h.maxAccountSwitches, c.GetHeader("If-None-Match"),
+	)
+	if c.Request.Context().Err() != nil {
+		return
+	}
+	if err != nil {
+		if errors.Is(err, service.ErrNoPinnedCodexModelsAccounts) {
+			writeOpenAIModelsError(c, http.StatusServiceUnavailable, "upstream_error", "No available OpenAI model discovery accounts")
+			return
+		}
+		writeOpenAIModelsError(c, infraerrors.Code(err), "upstream_error", infraerrors.Message(err))
+		return
+	}
+	setOpsSelectedAccount(c, account.ID, account.Platform)
+	writeOpenAIModelsResponse(c, response)
+}
+
+func writeOpenAIModelsError(c *gin.Context, status int, errorType, message string) {
+	c.JSON(status, gin.H{"error": gin.H{"type": errorType, "message": message}})
+}
+
+func writeOpenAIModelsResponse(c *gin.Context, manifest *service.OpenAIModelsResponse) {
+	if manifest.ETag != "" {
+		c.Header("ETag", manifest.ETag)
+	}
+	if manifest.NotModified {
+		c.Status(http.StatusNotModified)
+		c.Writer.WriteHeaderNow()
+		return
+	}
+	c.Data(http.StatusOK, "application/json", manifest.Body)
+}
@@ -0,0 +1,217 @@
+package handler
+
+import (
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+	"time"
+
+	middleware2 "github.com/Wei-Shaw/sub2api/internal/server/middleware"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+func performOrdinaryPinnedModelsRequest(t *testing.T, codex *OpenAIGatewayHandler, group *service.Group, path, etag string) *httptest.ResponseRecorder {
+	t.Helper()
+	h := &GatewayHandler{openAIGatewayService: codex.gatewayService, maxAccountSwitches: 3}
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+	c.Request = httptest.NewRequest(http.MethodGet, path, nil)
+	c.Request.Header.Set("If-None-Match", etag)
+	c.Set(string(middleware2.ContextKeyAPIKey), &service.APIKey{GroupID: &group.ID, Group: group})
+	h.Models(c)
+	return recorder
+}
+
+func ordinaryPinnedModelIDs(t *testing.T, recorder *httptest.ResponseRecorder) []string {
+	t.Helper()
+	var response gatewayModelsResponseForTest
+	require.NoError(t, json.Unmarshal(recorder.Body.Bytes(), &response))
+	require.NotNil(t, response.Data, "an empty catalog must be an array, not null")
+	return modelIDsForTest(response.Data)
+}
+
+func TestOrdinaryPinnedModelsUsesSelectedAccountsAndFinalETag(t *testing.T) {
+	accounts := []service.Account{
+		newPinnedCodexAccount(1, service.StatusActive, true, false),
+		newPinnedCodexAccount(2, service.StatusActive, true, true),
+		newPinnedCodexAccount(3, service.StatusActive, true, false),
+	}
+	upstream := &codexModelsPinnedHTTPUpstream{bodies: map[int64]string{
+		1: `{"data":[{"id":"unselected"}]}`,
+		2: `{"data":[{"id":"special-model","owned_by":"first","created":123},{"id":"gpt-image-1"}]}`,
+		3: `{"data":[{"id":"special-model","owned_by":"second"},{"id":"text-embedding-3-large"}]}`,
+	}}
+	h := newPinnedCodexTestHandler(accounts, upstream, 3)
+	group := &service.Group{ID: 91, Platform: service.PlatformOpenAI,
+		CodexModelsManifestConfig: service.GroupCodexModelsManifestConfig{Enabled: true, AccountIDs: []int64{2, 3}}}
+	first := performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+	require.Equal(t, http.StatusOK, first.Code, first.Body.String())
+	require.Equal(t, []string{"special-model", "gpt-image-1", "text-embedding-3-large"}, ordinaryPinnedModelIDs(t, first))
+	var response gatewayModelsResponseForTest
+	require.NoError(t, json.Unmarshal(first.Body.Bytes(), &response))
+	require.Equal(t, "model", response.Data[0].Object)
+	require.Equal(t, "first", response.Data[0].OwnedBy)
+	require.EqualValues(t, 123, response.Data[0].Created)
+	require.Equal(t, []int64{2, 3}, upstream.accountIDs())
+	alias := performOrdinaryPinnedModelsRequest(t, h, group, "/models", first.Header().Get("ETag"))
+	require.Equal(t, http.StatusNotModified, alias.Code)
+	require.Empty(t, alias.Body.String())
+	require.Equal(t, []int64{2, 3}, upstream.accountIDs(), "fresh requests must not hit upstream")
+
+	otherGroup := *group
+	otherGroup.ID++
+	otherGroup.ModelAllowlist = service.GroupModelAllowlist{Enabled: true, Models: []string{"text-embedding-3-large", "special-model"}}
+	filtered := performOrdinaryPinnedModelsRequest(t, h, &otherGroup, "/v1/models", first.Header().Get("ETag"))
+	require.Equal(t, http.StatusOK, filtered.Code)
+	require.Equal(t, []string{"text-embedding-3-large", "special-model"}, ordinaryPinnedModelIDs(t, filtered))
+	require.NotEqual(t, first.Header().Get("ETag"), filtered.Header().Get("ETag"))
+	again := performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+	require.Equal(t, first.Body.String(), again.Body.String(), "group filters must not modify account cache")
+	require.Equal(t, []int64{2, 3}, upstream.accountIDs())
+}
+
+func TestOrdinaryPinnedModelsFailureAndEmptyPolicies(t *testing.T) {
+	for _, tc := range []struct {
+		name       string
+		accountIDs []int64
+		bodies     map[int64]string
+		statuses   map[int64]int
+		fallback   bool
+		selected   []string
+		wantStatus int
+		wantIDs    []string
+	}{
+		{name: "partial failure", accountIDs: []int64{2, 3}, bodies: map[int64]string{2: `{"data":[{"id":"ok"}]}`}, statuses: map[int64]int{3: 503}, wantStatus: 200, wantIDs: []string{"ok"}},
+		{name: "all failed", accountIDs: []int64{2, 3}, statuses: map[int64]int{2: 429, 3: 504}, wantStatus: 502},
+		{name: "missing members", accountIDs: []int64{99}, wantStatus: 503},
+		{name: "empty config", wantStatus: 503},
+		{name: "fallback from missing members", accountIDs: []int64{99}, fallback: true, wantStatus: 200, wantIDs: []string{"from-scheduler"}},
+		{name: "fallback from failure", accountIDs: []int64{2}, statuses: map[int64]int{2: 503}, fallback: true, wantStatus: 200, wantIDs: []string{"from-scheduler"}},
+		{name: "valid empty", accountIDs: []int64{2}, bodies: map[int64]string{2: `{"data":[]}`}, fallback: true, wantStatus: 200},
+		{name: "filtered empty", accountIDs: []int64{2}, bodies: map[int64]string{2: `{"data":[{"id":"ok"}]}`}, selected: []string{"absent"}, fallback: true, wantStatus: 200},
+		{name: "invalid envelope", accountIDs: []int64{2}, bodies: map[int64]string{2: `{"error":"denied"}`}, wantStatus: 502},
+		{name: "null data", accountIDs: []int64{2}, bodies: map[int64]string{2: `{"data":null}`}, wantStatus: 502},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			accounts := []service.Account{
+				newPinnedCodexAccount(1, service.StatusActive, true, false),
+				newPinnedCodexAccount(2, service.StatusActive, true, false),
+				newPinnedCodexAccount(3, service.StatusActive, true, false),
+			}
+			bodies := map[int64]string{1: `{"data":[{"id":"from-scheduler"}]}`}
+			for id, body := range tc.bodies {
+				bodies[id] = body
+			}
+			upstream := &codexModelsPinnedHTTPUpstream{bodies: bodies, statuses: tc.statuses}
+			h := newPinnedCodexTestHandler(accounts, upstream, 3)
+			group := &service.Group{ID: 92, Platform: service.PlatformOpenAI,
+				ModelAllowlist:            service.GroupModelAllowlist{Enabled: len(tc.selected) > 0, Models: tc.selected},
+				CodexModelsManifestConfig: service.GroupCodexModelsManifestConfig{Enabled: true, AccountIDs: tc.accountIDs, FallbackToScheduler: tc.fallback}}
+			recorder := performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+			require.Equal(t, tc.wantStatus, recorder.Code, recorder.Body.String())
+			if tc.wantStatus == 200 {
+				require.Equal(t, len(tc.wantIDs), len(ordinaryPinnedModelIDs(t, recorder)))
+				if len(tc.wantIDs) > 0 {
+					require.Equal(t, tc.wantIDs, ordinaryPinnedModelIDs(t, recorder))
+				} else {
+					require.NotContains(t, upstream.accountIDs(), int64(1), "empty success must not trigger fallback")
+				}
+			}
+		})
+	}
+}
+
+func TestOrdinaryPinnedModelsSkipsPersistentlyUnavailableAccounts(t *testing.T) {
+	accounts := []service.Account{
+		newPinnedCodexAccount(1, service.StatusActive, true, true),
+		newPinnedCodexAccount(2, service.StatusActive, false, false),
+		newPinnedCodexAccount(3, service.StatusDisabled, true, false),
+		newPinnedCodexAccount(4, service.StatusActive, true, false),
+	}
+	expired := time.Now().Add(-time.Hour)
+	accounts[3].ExpiresAt = &expired
+	accounts[3].AutoPauseOnExpired = true
+	overloaded := time.Now().Add(time.Hour)
+	accounts[0].OverloadUntil = &overloaded
+	accounts[0].TempUnschedulableUntil = &overloaded
+	upstream := &codexModelsPinnedHTTPUpstream{bodies: map[int64]string{1: `{"data":[{"id":"available"}]}`}}
+	h := newPinnedCodexTestHandler(accounts, upstream, 3)
+	group := &service.Group{ID: 93, Platform: service.PlatformOpenAI,
+		CodexModelsManifestConfig: service.GroupCodexModelsManifestConfig{Enabled: true, AccountIDs: []int64{1, 2, 3, 4, 99}}}
+	recorder := performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+	require.Equal(t, http.StatusOK, recorder.Code, recorder.Body.String())
+	require.Equal(t, []int64{1}, upstream.accountIDs())
+}
+
+func TestPinnedModelsAllowlistExpandsWildcardsForBothRepresentations(t *testing.T) {
+	for _, codex := range []bool{false, true} {
+		t.Run(map[bool]string{false: "ordinary", true: "codex"}[codex], func(t *testing.T) {
+			accounts := []service.Account{newPinnedCodexAccount(1, service.StatusActive, true, false)}
+			accounts[0].Credentials["model_mapping"] = map[string]any{
+				"public-a": "gpt-5.5", "public-b": "gpt-5.5", "blocked": "gpt-5.5",
+			}
+			body := `{"data":[{"id":"gpt-5.5","owned_by":"provider"}]}`
+			if codex {
+				body = `{"models":[{"slug":"gpt-5.5","context_window":424242}]}`
+			}
+			upstream := &codexModelsPinnedHTTPUpstream{bodies: map[int64]string{1: body}}
+			h := newPinnedCodexTestHandler(accounts, upstream, 3)
+			group := &service.Group{ID: 95, Platform: service.PlatformOpenAI,
+				ModelAllowlist:            service.GroupModelAllowlist{Enabled: true, Models: []string{"public-b", "public-*"}},
+				CodexModelsManifestConfig: service.GroupCodexModelsManifestConfig{Enabled: true, AccountIDs: []int64{1}}}
+			if codex {
+				recorder := performPinnedCodexModelsRequest(t, h, group, "")
+				require.Equal(t, http.StatusOK, recorder.Code, recorder.Body.String())
+				require.Equal(t, []string{"public-b", "public-a"}, codexHandlerManifestSlugs(t, recorder))
+				require.Contains(t, recorder.Body.String(), `"context_window":424242`)
+			} else {
+				recorder := performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+				require.Equal(t, http.StatusOK, recorder.Code, recorder.Body.String())
+				require.Equal(t, []string{"public-b", "public-a"}, ordinaryPinnedModelIDs(t, recorder))
+				require.Contains(t, recorder.Body.String(), `"owned_by":"provider"`)
+			}
+		})
+	}
+}
+
+func TestPinnedModelsMappingFollowsUpstreamDiscoveryForBothRepresentations(t *testing.T) {
+	for _, codex := range []bool{false, true} {
+		t.Run(map[bool]string{false: "ordinary", true: "codex"}[codex], func(t *testing.T) {
+			accounts := []service.Account{
+				newPinnedCodexAccount(1, service.StatusActive, true, false),
+				newPinnedCodexAccount(2, service.StatusActive, true, false),
+			}
+			accounts[0].Credentials["model_mapping"] = map[string]any{"unselected-alias": "gpt-5.5"}
+			accounts[1].Credentials["model_mapping"] = map[string]any{
+				"public-alias": "gpt-5.5", "missing-alias": "missing-upstream", "custom-*": "gpt-5.5",
+			}
+			body := `{"data":[{"id":"gpt-5.5","owned_by":"provider","created":123},{"id":"unmapped-upstream"}]}`
+			if codex {
+				body = `{"models":[{"slug":"gpt-5.5","context_window":424242},{"slug":"unmapped-upstream"}]}`
+			}
+			upstream := &codexModelsPinnedHTTPUpstream{bodies: map[int64]string{2: body}}
+			h := newPinnedCodexTestHandler(accounts, upstream, 3)
+			group := &service.Group{ID: 94, Platform: service.PlatformOpenAI,
+				ModelAllowlist:            service.GroupModelAllowlist{Enabled: true, Models: []string{"custom-concrete", "public-alias", "unselected-alias", "missing-alias"}},
+				CodexModelsManifestConfig: service.GroupCodexModelsManifestConfig{Enabled: true, AccountIDs: []int64{2}}}
+			var recorder *httptest.ResponseRecorder
+			if codex {
+				recorder = performPinnedCodexModelsRequest(t, h, group, "")
+			} else {
+				recorder = performOrdinaryPinnedModelsRequest(t, h, group, "/v1/models", "")
+			}
+			require.Equal(t, http.StatusOK, recorder.Code, recorder.Body.String())
+			require.Equal(t, []int64{2}, upstream.accountIDs(), "mappings must not bypass pinned discovery")
+			if codex {
+				require.Equal(t, []string{"custom-concrete", "public-alias"}, codexHandlerManifestSlugs(t, recorder))
+				require.Contains(t, recorder.Body.String(), `"context_window":424242`)
+			} else {
+				require.Equal(t, []string{"custom-concrete", "public-alias"}, ordinaryPinnedModelIDs(t, recorder))
+				require.Contains(t, recorder.Body.String(), `"owned_by":"provider"`)
+			}
+		})
+	}
+}
@@ -17,7 +17,7 @@ import (
 // 上以 OpenAI 兼容错误结构返回 HTTP 400。这些用例在 handler 的 service_tier
 // 校验处短路，不会进入账号选择/重试。
 //
-// 合法值（fast/priority/flex/auto/default/scale）与省略/null 的接受语义由
+// 合法值（fast/priority/flex/auto/default/scale/ultrafast）与省略/null 的接受语义由
 // service 层纯校验函数 TestValidateOpenAIServiceTierField 覆盖，避免 handler
 // 测试走入真实账号选择/重试路径。

@@ -56,7 +56,7 @@ type responsesFailedEvent struct {
 // 返回 false 表示 writer 不支持 Flusher，无法以 SSE 形式回报错误；
 // 此时 caller 也无法回退到 JSON（HTTP 200 已固化），通常意味着连接已经损坏，
 // 应当让请求处理函数 return，由上层关闭连接。
-func writeResponsesFailedSSE(c *gin.Context, errType, message string) bool {
+func writeResponsesFailedSSE(c *gin.Context, errType, code, message string) bool {
 	flusher, ok := c.Writer.(http.Flusher)
 	if !ok {
 		return false
@@ -72,7 +72,7 @@ func writeResponsesFailedSSE(c *gin.Context, errType, message string) bool {
 			Status:    "failed",
 			Output:    []any{},
 			Error: responsesFailedError{
-				Code:    mapResponsesErrorCode(errType),
+				Code:    mapResponsesErrorCode(errType, code),
 				Message: message,
 			},
 		},
@@ -155,7 +155,10 @@ func requestModel(c *gin.Context) string {

 // mapResponsesErrorCode 把内部 errType 映射为 Responses 协议常见的 error.code。
 // 无明确映射时原样返回，保证至少可读。
-func mapResponsesErrorCode(errType string) string {
+func mapResponsesErrorCode(errType, code string) string {
+	if code != "" {
+		return code
+	}
 	switch errType {
 	case "rate_limit_error":
 		return "rate_limit_exceeded"
@@ -12,6 +12,7 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
+	"github.com/tidwall/gjson"
 )

 // Regression for the production incident on 2026-05-24 around 9:13 CST:
@@ -77,6 +78,60 @@ func TestOpenAIHandleStreamingAwareError_ResponsesStreamingEmitsResponseFailed(t
 	assert.Equal(t, "Concurrency limit exceeded for user, please retry later", errObj["message"])
 }

+func TestOpenAIAdmissionError_SynchronousNonResponsesIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointChatCompletions)
+	h := &OpenAIGatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayQueueFullCode, "Too many pending requests, please retry later", false, false)
+
+	assert.Equal(t, http.StatusTooManyRequests, w.Code)
+	assert.Equal(t, "rate_limit_error", gjson.GetBytes(w.Body.Bytes(), "error.type").String())
+	assert.Equal(t, gatewayQueueFullCode, gjson.GetBytes(w.Body.Bytes(), "error.code").String())
+}
+
+func TestOpenAIAdmissionError_SynchronousResponsesIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointResponses)
+	h := &OpenAIGatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayConcurrencyLimitCode, "Concurrency limit exceeded for account, please retry later", false, false)
+
+	assert.Equal(t, http.StatusTooManyRequests, w.Code)
+	assert.Equal(t, "rate_limit_error", gjson.GetBytes(w.Body.Bytes(), "error.type").String())
+	assert.Equal(t, gatewayConcurrencyLimitCode, gjson.GetBytes(w.Body.Bytes(), "error.code").String())
+}
+
+func TestOpenAIAdmissionError_ResponsesFailedIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointResponses)
+	h := &OpenAIGatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayConcurrencyLimitCode, "Concurrency limit exceeded for account, please retry later", true, false)
+
+	_, errObj := parseResponsesFailedSSE(t, w.Body.String())
+	assert.Equal(t, gatewayConcurrencyLimitCode, errObj["code"])
+}
+
+func TestOpenAIAdmissionError_StreamingNonResponsesIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointChatCompletions)
+	h := &OpenAIGatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayQueueFullCode, "Too many pending requests, please retry later", true, false)
+
+	body := w.Body.String()
+	assert.True(t, strings.HasPrefix(body, "event: error\n"))
+	payload := body[strings.Index(body, "{"):]
+	assert.Equal(t, "rate_limit_error", gjson.Get(payload, "error.type").String())
+	assert.Equal(t, gatewayQueueFullCode, gjson.Get(payload, "error.code").String())
+}
+
+func TestOpenAIUpstreamRateLimitWithoutLocalCodeRemainsUnchanged(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointChatCompletions)
+	h := &OpenAIGatewayHandler{}
+	h.handleStreamingAwareError(c, http.StatusTooManyRequests, "rate_limit_error", "upstream limited", false)
+
+	assert.Equal(t, "rate_limit_error", gjson.GetBytes(w.Body.Bytes(), "error.type").String())
+	assert.False(t, gjson.GetBytes(w.Body.Bytes(), "error.code").Exists())
+}
+
 // 当 setOpsRequestContext 写过 model，合成事件应回填该字段（与 codebase 已有 makeResponsesCompletedEvent 对齐）。
 func TestOpenAIHandleStreamingAwareError_ResponsesStreamingIncludesModel(t *testing.T) {
 	c, w := newGinContextForEndpoint(t, EndpointResponses)
@@ -164,6 +219,41 @@ func TestGatewayHandleStreamingAwareError_ResponsesStreamingEmitsResponseFailed(
 	assert.Equal(t, "upstream gone", errObj["message"])
 }

+func TestGatewayAdmissionError_SynchronousIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointMessages)
+	h := &GatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayQueueFullCode, "Too many pending requests, please retry later", false)
+
+	assert.Equal(t, http.StatusTooManyRequests, w.Code)
+	assert.Equal(t, "error", gjson.GetBytes(w.Body.Bytes(), "type").String())
+	assert.Equal(t, "rate_limit_error", gjson.GetBytes(w.Body.Bytes(), "error.type").String())
+	assert.Equal(t, gatewayQueueFullCode, gjson.GetBytes(w.Body.Bytes(), "error.code").String())
+}
+
+func TestGatewayAdmissionError_MessagesStreamingIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, EndpointMessages)
+	h := &GatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayConcurrencyLimitCode, "Concurrency limit exceeded for account, please retry later", true)
+
+	body := w.Body.String()
+	assert.True(t, strings.HasPrefix(body, `data: {"type":"error"`))
+	payload := body[strings.Index(body, "{"):]
+	assert.Equal(t, "rate_limit_error", gjson.Get(payload, "error.type").String())
+	assert.Equal(t, gatewayConcurrencyLimitCode, gjson.Get(payload, "error.code").String())
+}
+
+func TestGatewayAdmissionError_BareResponsesFailedIncludesGatewayCode(t *testing.T) {
+	c, w := newGinContextForEndpoint(t, "/responses")
+	h := &GatewayHandler{}
+	h.handleStreamingAwareErrorWithCode(c, http.StatusTooManyRequests, "rate_limit_error",
+		gatewayQueueFullCode, "Too many pending requests, please retry later", true)
+
+	_, errObj := parseResponsesFailedSSE(t, w.Body.String())
+	assert.Equal(t, gatewayQueueFullCode, errObj["code"])
+}
+
 // Gateway handler: /v1/messages preserves the legacy data:{type:error,...} format
 // (Anthropic spec accepts a type:"error" stream event).
 func TestGatewayHandleStreamingAwareError_MessagesStreamingKeepsLegacy(t *testing.T) {
@@ -264,6 +354,7 @@ func TestMapResponsesErrorCode(t *testing.T) {
 		{"custom_thing", "custom_thing"},
 	}
 	for _, tc := range cases {
-		assert.Equal(t, tc.out, mapResponsesErrorCode(tc.in), "in=%q", tc.in)
+		assert.Equal(t, tc.out, mapResponsesErrorCode(tc.in, ""), "in=%q", tc.in)
 	}
+	assert.Equal(t, gatewayQueueFullCode, mapResponsesErrorCode("rate_limit_error", gatewayQueueFullCode))
 }
@@ -250,7 +250,7 @@ var ProviderSet = wire.NewSet(
 	// Admin handlers
 	admin.NewDashboardHandler,
 	admin.NewUserHandler,
-	admin.NewGroupHandler,
+	admin.NewGroupHandlerWithConfig,
 	admin.ProvideAccountHandler,
 	admin.NewAnnouncementHandler,
 	admin.NewDataManagementHandler,