diff --git a/src/i18n/messages/en.json b/src/i18n/messages/en.json
index a6269686d6..1dcf49ab91 100644
--- a/src/i18n/messages/en.json
+++ b/src/i18n/messages/en.json
@@ -914,26 +914,6 @@
         "when": "Use when long sessions must survive account rotation without losing the working context.",
         "avoid": "Avoid when account switching is rare or when you do not want extra summarization requests.",
         "example": "Example: Codex sessions that rotate across multiple accounts near quota exhaustion."
-      },
-      "fill-first": {
-        "when": "Use when you want to fully exhaust one provider's quota before moving to the next.",
-        "avoid": "Avoid when you need request-level load balancing between providers.",
-        "example": "Example: Use up all $200 of Deepgram credits before falling back to Groq."
-      },
-      "auto": {
-        "when": "Use when you have one preferred model and only want fallback on failure.",
-        "avoid": "Avoid when you need balanced load between models.",
-        "example": "Example: Primary coding model with cheaper backup for outages."
-      },
-      "lkgp": {
-        "when": "Use when you have one preferred model and only want fallback on failure.",
-        "avoid": "Avoid when you need balanced load between models.",
-        "example": "Example: Primary coding model with cheaper backup for outages."
-      },
-      "context-optimized": {
-        "when": "Use when you have one preferred model and only want fallback on failure.",
-        "avoid": "Avoid when you need balanced load between models.",
-        "example": "Example: Primary coding model with cheaper backup for outages."
       }
     },
     "advancedHelp": {
@@ -1013,25 +993,6 @@
     "candidatePoolHint": "Select which providers this engine should evaluate. Leave empty to use all active providers.",
     "candidatePoolEmpty": "No active providers available yet.",
     "candidatePoolAllProviders": "All providers",
-    "builderStagesDescription": "Move through the stages in order to define the combo, build the steps, choose the routing strategy and review the result.",
-    "builderStepsDescription": "Build each combo step in sequence: provider, model, then account. This allows repeating the same provider and model with different accounts.",
-    "selectProvider": "Select Provider",
-    "selectProviderPlaceholder": "Select provider",
-    "selectModel": "Select Model",
-    "selectModelPlaceholder": "Select provider first",
-    "selectAccount": "Select Account",
-    "autoSelectAccount": "Auto-select account at runtime",
-    "previewNextStep": "Choose provider and model to preview the next step.",
-    "addStep": "Add Step",
-    "comboReference": "Combo Reference",
-    "selectComboToReference": "Select an existing combo to reference",
-    "addComboReference": "Add Combo Reference",
-    "browseLegacyCatalog": "Browse legacy model catalog",
-    "addStepBeforeContinue": "Add at least one step before continuing to the next stage.",
-    "modePackPerformance": "Performance",
-    "modePackBudget": "Budget",
-    "modePackBalanced": "Balanced",
-    "modePackCustom": "Custom",
     "modePackLabel": "Mode Pack",
     "routerStrategyLabel": "Router Strategy",
     "strategyRules": "Rules (6-Factor Scoring)",
@@ -1091,54 +1052,12 @@
         "tip2": "Keep a quality fallback for hard prompts.",
         "tip3": "Use for batch/background jobs where cost is the main KPI."
       },
-      "fill-first": {
-        "title": "Quota drain strategy",
-        "description": "Exhausts one provider's quota before moving to the next in chain.",
-        "tip1": "Order models by free quota size — biggest first.",
-        "tip2": "Enable health checks to skip drained providers.",
-        "tip3": "Ideal for free-tier stacking (Deepgram → Groq → NIM)."
-      },
-      "p2c": {
-        "title": "Power-of-Two-Choices",
-        "description": "Picks the less-loaded of two random candidates per request — low latency at scale.",
-        "tip1": "Use with 4+ models for best effect.",
-        "tip2": "Requires latency telemetry enabled in Settings.",
-        "tip3": "Great replacement for round-robin in high-throughput combos."
-      },
-      "context-relay": {
-        "title": "Session continuity first",
-        "description": "Best when account rotation is expected and the next account must inherit a condensed task summary.",
-        "tip1": "Use with providers that rotate accounts for the same model family.",
-        "tip2": "Keep the handoff threshold below the hard quota cutoff to give the summary time to generate.",
-        "tip3": "Set a dedicated summary model only when the primary model is too expensive or unstable."
-      },
       "strict-random": {
         "title": "Shuffle deck distribution",
         "description": "Each model is used exactly once per cycle before reshuffling.",
         "tip1": "Use at least 2 models for meaningful distribution.",
         "tip2": "Works best with equivalent-performance models.",
         "tip3": "Ideal for load balancing across multiple API accounts."
-      },
-      "auto": {
-        "title": "Fail-safe baseline",
-        "description": "Use one primary model and keep fallback chain short and reliable.",
-        "tip1": "Put your most reliable model first.",
-        "tip2": "Keep 1-2 backup models with similar quality.",
-        "tip3": "Use safe retries to absorb transient provider failures."
-      },
-      "lkgp": {
-        "title": "Fail-safe baseline",
-        "description": "Use one primary model and keep fallback chain short and reliable.",
-        "tip1": "Put your most reliable model first.",
-        "tip2": "Keep 1-2 backup models with similar quality.",
-        "tip3": "Use safe retries to absorb transient provider failures."
-      },
-      "context-optimized": {
-        "title": "Fail-safe baseline",
-        "description": "Use one primary model and keep fallback chain short and reliable.",
-        "tip1": "Put your most reliable model first.",
-        "tip2": "Keep 1-2 backup models with similar quality.",
-        "tip3": "Use safe retries to absorb transient provider failures."
       }
     },
     "templateFreeStack": "Free Stack ($0)",
@@ -3399,258 +3318,5 @@
     "deduplicatedRequests": "Deduplicated Requests",
     "savedCalls": "Saved API Calls",
     "totalProcessed": "Total Requests Processed"
-  },
-  "proxyConfigModal": {
-    "levelGlobal": "Global",
-    "levelProvider": "Provider",
-    "levelCombo": "Combo",
-    "levelKey": "Key",
-    "levelDirect": "Direct (none)",
-    "titleGlobal": "Global Proxy Configuration",
-    "titleLevel": "{level} Proxy — {label}",
-    "loading": "Loading proxy configuration...",
-    "inheritingFrom": "Inheriting from",
-    "source": "Source",
-    "savedProxy": "Saved Proxy",
-    "custom": "Custom",
-    "selectSavedProxyPlaceholder": "Select saved proxy...",
-    "proxyType": "Proxy Type",
-    "host": "Host",
-    "hostPlaceholder": "1.2.3.4 or proxy.example.com",
-    "port": "Port",
-    "authOptional": "Authentication (optional)",
-    "username": "Username",
-    "usernamePlaceholder": "Username",
-    "password": "Password",
-    "passwordPlaceholder": "Password",
-    "connected": "Connected",
-    "ip": "IP:",
-    "connectionFailed": "Connection failed",
-    "testConnection": "Test Connection",
-    "clear": "Clear",
-    "cancel": "Cancel",
-    "save": "Save",
-    "errorSelectSavedProxy": "Select a saved proxy before saving.",
-    "errorSelectProxyFirst": "Select a saved proxy first.",
-    "errorProxyNotFound": "Selected proxy not found.",
-    "errorClearSavedProxy": "Failed to clear saved proxy",
-    "errorSaveProxy": "Failed to save proxy configuration",
-    "errorClearProxy": "Failed to clear proxy configuration",
-    "errorSocks5Hidden": "SOCKS5 is configured but hidden because NEXT_PUBLIC_ENABLE_SOCKS5_PROXY=false."
-  },
-  "oauthModal": {
-    "title": "Connect {providerName}",
-    "waiting": "Waiting for Authorization",
-    "completeAuthInPopup": "Complete the authorization in the popup window.",
-    "popupClosedHint": "If the popup closes without redirecting back (e.g. Qoder), this dialog will automatically switch to manual URL input mode.",
-    "popupBlocked": "Popup blocked? Enter URL manually",
-    "deviceCodeVisitUrl": "Visit the URL below and enter the code:",
-    "deviceCodeVerificationUrl": "Verification URL",
-    "deviceCodeYourCode": "Your Code",
-    "deviceCodeWaiting": "Waiting for authorization...",
-    "googleOAuthWarning": "Remote access + Google OAuth: The default credentials only accept redirects to <code>localhost</code>. After authorizing, your browser will try to open <code>localhost</code> — copy that full URL and paste it below. For fully remote use without this manual step, <a>configure your own OAuth credentials</a>.",
-    "remoteAccessInfo": "Remote access: Since you're accessing OmniRoute remotely, after authorizing you'll see an error page (localhost not found). That's expected — just copy the full URL from your browser's address bar and paste it below.",
-    "step1OpenUrl": "Step 1: Open this URL in your browser",
-    "copy": "Copy",
-    "step2PasteCallback": "Step 2: Paste the callback URL or auth code here",
-    "step2Hint": "After authorization, paste the full callback URL. For Claude Code and Cline, you can also paste the Authentication Code directly, for example <code>code#state</code>.",
-    "connect": "Connect",
-    "cancel": "Cancel",
-    "success": "Connected Successfully!",
-    "successMessage": "Your {providerName} account has been connected.",
-    "done": "Done",
-    "error": "Connection Failed",
-    "tryAgain": "Try Again"
-  },
-  "cursorAuthModal": {
-    "title": "Connect Cursor IDE",
-    "autoDetecting": "Auto-detecting tokens...",
-    "readingFromCursor": "Reading from Cursor IDE or cursor-agent",
-    "tokensAutoDetected": "Tokens auto-detected from Cursor IDE successfully!",
-    "cursorNotDetected": "Cursor IDE not detected. Please paste your tokens manually.",
-    "accessToken": "Access Token",
-    "required": "*",
-    "accessTokenPlaceholder": "Access token will be auto-filled...",
-    "machineId": "Machine ID",
-    "optional": "(optional)",
-    "machineIdPlaceholder": "Machine ID will be auto-filled...",
-    "importing": "Importing...",
-    "importToken": "Import Token",
-    "cancel": "Cancel",
-    "errorAutoDetect": "Could not auto-detect tokens",
-    "errorAutoDetectFailed": "Failed to auto-detect tokens",
-    "errorEnterToken": "Please enter an access token",
-    "errorImportFailed": "Import failed"
-  },
-  "pricingModal": {
-    "title": "Pricing Configuration",
-    "loading": "Loading pricing data...",
-    "pricingRatesFormat": "Pricing Rates Format",
-    "ratesDescription": "All rates are in <strong>dollars per million tokens</strong> ($/1M tokens). Example: Input rate of 2.50 means $2.50 per 1,000,000 input tokens.",
-    "model": "Model",
-    "input": "Input",
-    "output": "Output",
-    "cached": "Cached",
-    "reasoning": "Reasoning",
-    "cacheCreation": "Cache Creation",
-    "noPricingData": "No pricing data available",
-    "resetToDefaults": "Reset to Defaults",
-    "cancel": "Cancel",
-    "saving": "Saving...",
-    "saveChanges": "Save Changes",
-    "resetConfirm": "Reset all pricing to defaults? This cannot be undone.",
-    "errorSaveFailed": "Failed to save pricing",
-    "errorResetFailed": "Failed to reset pricing"
-  },
-  "proxyRegistry": {
-    "title": "Proxy Registry",
-    "description": "Store reusable proxies and track assignments.",
-    "importLegacy": "Import Legacy",
-    "bulkAssign": "Bulk Assign",
-    "addProxy": "Add Proxy",
-    "loading": "Loading proxies...",
-    "noProxies": "No saved proxies yet.",
-    "tableName": "Name",
-    "tableEndpoint": "Endpoint",
-    "tableStatus": "Status",
-    "tableHealth": "Health (24h)",
-    "tableUsage": "Usage",
-    "tableActions": "Actions",
-    "test": "Test",
-    "edit": "Edit",
-    "delete": "Delete",
-    "modalCreateTitle": "Create Proxy",
-    "modalEditTitle": "Edit Proxy",
-    "labelName": "Name",
-    "labelType": "Type",
-    "labelHost": "Host",
-    "labelPort": "Port",
-    "labelUsername": "Username",
-    "labelPassword": "Password",
-    "labelRegion": "Region",
-    "labelStatus": "Status",
-    "labelNotes": "Notes",
-    "usernamePlaceholderEdit": "Leave blank to keep current username",
-    "passwordPlaceholderEdit": "Leave blank to keep current password",
-    "statusActive": "active",
-    "statusInactive": "inactive",
-    "cancel": "Cancel",
-    "save": "Save",
-    "bulkModalTitle": "Bulk Proxy Assignment",
-    "bulkLabelScope": "Scope",
-    "bulkLabelProxy": "Proxy",
-    "bulkClearAssignment": "(clear assignment)",
-    "bulkLabelScopeIds": "Scope IDs (comma or newline)",
-    "bulkScopeIdsPlaceholder": "provider-openai,provider-anthropic",
-    "bulkApply": "Apply",
-    "errorLoadFailed": "Failed to load proxy registry",
-    "errorNameHostRequired": "Name and host are required",
-    "errorSaveFailed": "Failed to save proxy",
-    "errorDeleteFailed": "Failed to delete proxy",
-    "errorForceDeleteConfirm": "This proxy is still assigned. Force delete and remove all assignments?",
-    "errorMigrateFailed": "Failed to migrate legacy proxy config",
-    "errorBulkFailed": "Failed to run bulk assignment",
-    "success": "✓",
-    "failure": "✗",
-    "failed": "failed",
-    "successRate": "{rate}% success",
-    "avgLatency": "{latency} ms avg",
-    "assignmentsCount": "{count} assignment(s)",
-    "noData": "—",
-    "testSuccess": "✓ {ip}",
-    "testLatency": "{latency}ms",
-    "testFailure": "✗ {error}"
-  },
-  "playground": {
-    "title": "Model Playground",
-    "description": "Test any model directly from the dashboard. Pick a provider, model, and endpoint type, then send a request to see the raw response.",
-    "endpoint": "Endpoint",
-    "provider": "Provider",
-    "model": "Model",
-    "accountKey": "Account / Key",
-    "autoAccounts": "Auto ({count} accounts)",
-    "noAccounts": "No accounts",
-    "send": "Send",
-    "cancel": "Cancel",
-    "audioFile": "Audio File",
-    "attachImages": "Attach Images (Vision)",
-    "multipartFormData": "multipart/form-data",
-    "upToImages": "up to 4 images",
-    "selectAudioFile": "Select an audio file to transcribe (mp3, wav, m4a, ogg, flac…)",
-    "clearAll": "Clear all",
-    "request": "Request",
-    "response": "Response",
-    "transcription": "Transcription",
-    "copy": "Copy",
-    "resetToDefault": "Reset to default",
-    "downloadAudio": "Download audio",
-    "copyText": "Copy text",
-    "transcriptionHint": "Transcription uses multipart/form-data. Upload the audio file above — JSON below controls extra params (model, language).",
-    "imagesGenerated": "{count} image{count, plural, one {} other {s}} generated",
-    "generatedImage": "Generated image {index}",
-    "save": "Save",
-    "endpointOptions": {
-      "chat": "Chat Completions",
-      "responses": "Responses",
-      "images": "Image Generation",
-      "embeddings": "Embeddings",
-      "speech": "Text to Speech",
-      "transcription": "Audio Transcription",
-      "video": "Video Generation",
-      "music": "Music Generation",
-      "rerank": "Rerank",
-      "search": "Web Search"
-    }
-  },
-  "requestLogger": {
-    "recording": "Recording",
-    "paused": "Paused",
-    "pipelineLogsOn": "Pipeline Logs On",
-    "pipelineLogsOff": "Pipeline Logs Off",
-    "updatingPipelineLogs": "Updating pipeline logs...",
-    "searchPlaceholder": "Search model, provider, account, API key, combo...",
-    "allProviders": "All Providers",
-    "allModels": "All Models",
-    "allAccounts": "All Accounts",
-    "allApiKeys": "All API Keys",
-    "total": "total",
-    "ok": "OK",
-    "err": "ERR",
-    "combo": "combo",
-    "keys": "keys",
-    "shown": "shown",
-    "sortNewest": "Newest",
-    "sortOldest": "Oldest",
-    "sortTokensDesc": "Tokens ↓",
-    "sortTokensAsc": "Tokens ↑",
-    "sortDurationDesc": "Duration ↓",
-    "sortDurationAsc": "Duration ↑",
-    "sortStatusDesc": "Status ↓",
-    "sortStatusAsc": "Status ↑",
-    "sortModelAsc": "Model A-Z",
-    "sortModelDesc": "Model Z-A",
-    "statusFilters": {
-      "all": "All",
-      "error": "Errors",
-      "success": "Success",
-      "combo": "Combo"
-    },
-    "columns": {
-      "status": "Status",
-      "model": "Model",
-      "requested": "Requested",
-      "provider": "Provider",
-      "protocol": "Req Protocol",
-      "account": "Account",
-      "apiKey": "API Key",
-      "combo": "Combo",
-      "tokens": "Tokens",
-      "duration": "Duration",
-      "time": "Time"
-    },
-    "loadingLogs": "Loading logs...",
-    "noLogs": "No logs recorded yet. Make some API calls to see them here.",
-    "noMatchingLogs": "No logs match the current filters.",
-    "callLogsInfo": "Call logs are also saved as JSON files to {dataDir} and rotated by {retentionDays} and {maxEntries}."
   }
 }
diff --git a/src/i18n/messages/zh-CN.json b/src/i18n/messages/zh-CN.json
index f28f0c058f..171bd2152e 100644
--- a/src/i18n/messages/zh-CN.json
+++ b/src/i18n/messages/zh-CN.json
@@ -1163,6 +1163,28 @@
     "reorderHandle": "拖拽排序",
     "failedReorder": "模型重排序失败",
     "builderFlowTitle": "组合构建流程",
+    "builderStage": {
+      "basics": {
+        "label": "基础",
+        "description": "名称和起始模板"
+      },
+      "steps": {
+        "label": "步骤",
+        "description": "提供商、模型和账户选择"
+      },
+      "strategy": {
+        "label": "策略",
+        "description": "路由行为和高级设置"
+      },
+      "intelligent": {
+        "label": "智能路由",
+        "description": "自动路由候选池、预设和评分"
+      },
+      "review": {
+        "label": "审查",
+        "description": "保存前的最终验证"
+      }
+    },
     "builderStageVisited": "阶段已完成",
     "builderStageCurrent": "当前阶段",
     "builderStagePending": "待处理",