diff --git a/src/app/v1/_lib/proxy/response-handler.ts b/src/app/v1/_lib/proxy/response-handler.ts
index e2dd19edc..7f0b4c7d0 100644
--- a/src/app/v1/_lib/proxy/response-handler.ts
+++ b/src/app/v1/_lib/proxy/response-handler.ts
@@ -1257,8 +1257,11 @@ export class ProxyResponseHandler {
               if (resolvedPricing) {
                 ensurePricingResolutionSpecialSetting(session, resolvedPricing);
                 const longContextPricing =
-                  matchLongContextPricing(billableUsageMetrics, resolvedPricing.priceData)
-                    ?.pricing ?? null;
+                  matchLongContextPricing(
+                    billableUsageMetrics,
+                    resolvedPricing.priceData,
+                    priorityServiceTierApplied ? "priority" : null
+                  )?.pricing ?? null;
                 const cost = calculateRequestCost(
                   billableUsageMetrics,
                   resolvedPricing.priceData,
@@ -2373,8 +2376,11 @@ export class ProxyResponseHandler {
               if (resolvedPricing) {
                 ensurePricingResolutionSpecialSetting(session, resolvedPricing);
                 const longContextPricing =
-                  matchLongContextPricing(billableUsageForCost, resolvedPricing.priceData)
-                    ?.pricing ?? null;
+                  matchLongContextPricing(
+                    billableUsageForCost,
+                    resolvedPricing.priceData,
+                    priorityServiceTierApplied ? "priority" : null
+                  )?.pricing ?? null;
                 const cost = calculateRequestCost(
                   billableUsageForCost,
                   resolvedPricing.priceData,
@@ -3466,7 +3472,11 @@ async function updateRequestCostFromUsage(
     }
 
     const longContextPricing =
-      matchLongContextPricing(usage, resolvedPricing.priceData)?.pricing ?? null;
+      matchLongContextPricing(
+        usage,
+        resolvedPricing.priceData,
+        priorityServiceTierApplied ? "priority" : null
+      )?.pricing ?? null;
     const cost = calculateRequestCost(
       usage,
       resolvedPricing.priceData,
@@ -3669,8 +3679,11 @@ export async function finalizeRequestStats(
         if (resolvedPricing) {
           ensurePricingResolutionSpecialSetting(session, resolvedPricing);
           const longContextPricing =
-            matchLongContextPricing(billableNormalizedUsage, resolvedPricing.priceData)?.pricing ??
-            null;
+            matchLongContextPricing(
+              billableNormalizedUsage,
+              resolvedPricing.priceData,
+              priorityServiceTierApplied ? "priority" : null
+            )?.pricing ?? null;
           const cost = calculateRequestCost(
             billableNormalizedUsage,
             resolvedPricing.priceData,
@@ -3778,7 +3791,11 @@ async function trackCostToRedis(
     ensurePricingResolutionSpecialSetting(session, resolvedPricing);
     const longContextPricing =
       longContextPricingOverride === undefined
-        ? (matchLongContextPricing(usage, resolvedPricing.priceData)?.pricing ?? null)
+        ? (matchLongContextPricing(
+            usage,
+            resolvedPricing.priceData,
+            priorityServiceTierApplied ? "priority" : null
+          )?.pricing ?? null)
         : longContextPricingOverride;
 
     const cost = calculateRequestCost(
diff --git a/src/lib/utils/cost-calculation.ts b/src/lib/utils/cost-calculation.ts
index 5bfee1c26..7ee803952 100644
--- a/src/lib/utils/cost-calculation.ts
+++ b/src/lib/utils/cost-calculation.ts
@@ -1,5 +1,5 @@
 import { CONTEXT_1M_TOKEN_THRESHOLD } from "@/lib/special-attributes";
-import type { ModelPriceData } from "@/types/model-price";
+import type { ModelPriceData, ServiceTierPricing } from "@/types/model-price";
 import { COST_SCALE, Decimal, toDecimal } from "./currency";
 
 const OPENAI_LONG_CONTEXT_TOKEN_THRESHOLD = 272000;
@@ -40,6 +40,8 @@ type RequestCostBreakdownOptions = Omit<
   "multiplier" | "groupMultiplier"
 >;
 
+type ServiceTierName = "priority";
+
 export interface LongContextPricingMatch {
   thresholdTokens: number;
   scope: "request" | "session";
@@ -111,6 +113,47 @@ function multiplyCost(quantity: number | undefined, unitCost: number | undefined
   return qtyDecimal.mul(costDecimal);
 }
 
+function getServiceTierPricing(
+  priceData: ModelPriceData,
+  serviceTier?: ServiceTierName | null
+): ServiceTierPricing | null {
+  if (!serviceTier) {
+    return null;
+  }
+
+  const pricing = priceData.service_tier_pricing;
+  if (!pricing || typeof pricing !== "object" || Array.isArray(pricing)) {
+    return null;
+  }
+
+  const tierPricing = pricing[serviceTier];
+  if (!tierPricing || typeof tierPricing !== "object" || Array.isArray(tierPricing)) {
+    return null;
+  }
+
+  return tierPricing;
+}
+
+function getPriorityServiceTierPricing(
+  priceData: ModelPriceData,
+  priorityServiceTierApplied: boolean
+): ServiceTierPricing | null {
+  return priorityServiceTierApplied ? getServiceTierPricing(priceData, "priority") : null;
+}
+
+function resolvePriorityAwareRate(
+  priorityServiceTierApplied: boolean,
+  serviceTierRate: number | undefined,
+  legacyPriorityRate: number | undefined,
+  baseRate: number | undefined
+): number | undefined {
+  if (!priorityServiceTierApplied) {
+    return baseRate;
+  }
+
+  return serviceTierRate ?? legacyPriorityRate ?? baseRate;
+}
+
 function resolveLongContextThreshold(priceData: ModelPriceData): number {
   const has272kFields =
     typeof priceData.input_cost_per_token_above_272k_tokens === "number" ||
@@ -131,9 +174,11 @@ function resolveLongContextThreshold(priceData: ModelPriceData): number {
 }
 
 export function resolveLongContextPricing(
-  priceData: ModelPriceData
+  priceData: ModelPriceData,
+  serviceTier?: ServiceTierName | null
 ): ResolvedLongContextPricing | null {
-  const pricing = priceData.long_context_pricing;
+  const serviceTierPricing = getServiceTierPricing(priceData, serviceTier);
+  const pricing = serviceTierPricing?.long_context_pricing ?? priceData.long_context_pricing;
   if (!pricing) {
     return null;
   }
@@ -143,17 +188,35 @@ export function resolveLongContextPricing(
     return null;
   }
 
-  const baseInputCost = priceData.input_cost_per_token;
-  const baseOutputCost = priceData.output_cost_per_token;
+  const priorityServiceTierApplied = serviceTier === "priority";
+  const baseInputCost = resolvePriorityAwareRate(
+    priorityServiceTierApplied,
+    serviceTierPricing?.input_cost_per_token,
+    priceData.input_cost_per_token_priority,
+    priceData.input_cost_per_token
+  );
+  const baseOutputCost = resolvePriorityAwareRate(
+    priorityServiceTierApplied,
+    serviceTierPricing?.output_cost_per_token,
+    priceData.output_cost_per_token_priority,
+    priceData.output_cost_per_token
+  );
   const baseCacheCreation5mCost =
+    serviceTierPricing?.cache_creation_input_token_cost ??
     priceData.cache_creation_input_token_cost ??
     (baseInputCost != null ? baseInputCost * 1.25 : undefined);
   const baseCacheCreation1hCost =
+    serviceTierPricing?.cache_creation_input_token_cost_above_1hr ??
     priceData.cache_creation_input_token_cost_above_1hr ??
     (baseInputCost != null ? baseInputCost * 2 : undefined) ??
     baseCacheCreation5mCost;
   const baseCacheReadCost =
-    priceData.cache_read_input_token_cost ??
+    resolvePriorityAwareRate(
+      priorityServiceTierApplied,
+      serviceTierPricing?.cache_read_input_token_cost,
+      priceData.cache_read_input_token_cost_priority,
+      priceData.cache_read_input_token_cost
+    ) ??
     (baseInputCost != null
       ? baseInputCost * 0.1
       : baseOutputCost != null
@@ -301,11 +364,18 @@ function resolvePriorityAwareLongContextRate(
     above272kPriority?: number;
     above200k?: number;
     above200kPriority?: number;
+    serviceTierAbove272k?: number;
+    serviceTierAbove200k?: number;
   }
 ): number | undefined {
   if (priorityServiceTierApplied) {
     return (
-      fields.above272kPriority ?? fields.above200kPriority ?? fields.above272k ?? fields.above200k
+      fields.serviceTierAbove272k ??
+      fields.serviceTierAbove200k ??
+      fields.above272kPriority ??
+      fields.above200kPriority ??
+      fields.above272k ??
+      fields.above200k
     );
   }
 
@@ -332,9 +402,10 @@ export function getLongContextTriggerInputTokens(
 
 export function matchLongContextPricing(
   usage: UsageMetrics,
-  priceData: ModelPriceData
+  priceData: ModelPriceData,
+  serviceTier?: ServiceTierName | null
 ): LongContextPricingMatch | null {
-  const pricing = resolveLongContextPricing(priceData);
+  const pricing = resolveLongContextPricing(priceData, serviceTier);
   if (!pricing) {
     return null;
   }
@@ -387,17 +458,24 @@ export function calculateRequestCostBreakdown(
 
   const baseInputCostPerToken = priceData.input_cost_per_token;
   const baseOutputCostPerToken = priceData.output_cost_per_token;
-  const inputCostPerToken =
-    options.priorityServiceTierApplied &&
-    typeof priceData.input_cost_per_token_priority === "number"
-      ? priceData.input_cost_per_token_priority
-      : baseInputCostPerToken;
-  const outputCostPerToken =
-    options.priorityServiceTierApplied &&
-    typeof priceData.output_cost_per_token_priority === "number"
-      ? priceData.output_cost_per_token_priority
-      : baseOutputCostPerToken;
-  const inputCostPerRequest = priceData.input_cost_per_request;
+  const priorityTierPricing = getPriorityServiceTierPricing(
+    priceData,
+    options.priorityServiceTierApplied
+  );
+  const inputCostPerToken = resolvePriorityAwareRate(
+    options.priorityServiceTierApplied,
+    priorityTierPricing?.input_cost_per_token,
+    priceData.input_cost_per_token_priority,
+    baseInputCostPerToken
+  );
+  const outputCostPerToken = resolvePriorityAwareRate(
+    options.priorityServiceTierApplied,
+    priorityTierPricing?.output_cost_per_token,
+    priceData.output_cost_per_token_priority,
+    baseOutputCostPerToken
+  );
+  const inputCostPerRequest =
+    priorityTierPricing?.input_cost_per_request ?? priceData.input_cost_per_request;
   const longContextPricing = options.longContextPricing;
 
   // Per-request cost -> input bucket
@@ -413,23 +491,27 @@ export function calculateRequestCostBreakdown(
   }
 
   const cacheCreation5mCost =
+    priorityTierPricing?.cache_creation_input_token_cost ??
     priceData.cache_creation_input_token_cost ??
-    (baseInputCostPerToken != null ? baseInputCostPerToken * 1.25 : undefined);
+    (inputCostPerToken != null ? inputCostPerToken * 1.25 : undefined);
 
   const cacheCreation1hCost =
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr ??
     priceData.cache_creation_input_token_cost_above_1hr ??
-    (baseInputCostPerToken != null ? baseInputCostPerToken * 2 : undefined) ??
+    (inputCostPerToken != null ? inputCostPerToken * 2 : undefined) ??
     cacheCreation5mCost;
 
   const cacheReadCost =
-    (options.priorityServiceTierApplied &&
-    typeof priceData.cache_read_input_token_cost_priority === "number"
-      ? priceData.cache_read_input_token_cost_priority
-      : priceData.cache_read_input_token_cost) ??
-    (baseInputCostPerToken != null
-      ? baseInputCostPerToken * 0.1
-      : baseOutputCostPerToken != null
-        ? baseOutputCostPerToken * 0.1
+    resolvePriorityAwareRate(
+      options.priorityServiceTierApplied,
+      priorityTierPricing?.cache_read_input_token_cost,
+      priceData.cache_read_input_token_cost_priority,
+      priceData.cache_read_input_token_cost
+    ) ??
+    (inputCostPerToken != null
+      ? inputCostPerToken * 0.1
+      : outputCostPerToken != null
+        ? outputCostPerToken * 0.1
         : undefined);
 
   // Derive cache creation tokens by TTL
@@ -457,6 +539,8 @@ export function calculateRequestCostBreakdown(
       above272kPriority: priceData.input_cost_per_token_above_272k_tokens_priority,
       above200k: priceData.input_cost_per_token_above_200k_tokens,
       above200kPriority: priceData.input_cost_per_token_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.input_cost_per_token_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.input_cost_per_token_above_200k_tokens,
     }
   );
   const outputAboveThreshold = resolvePriorityAwareLongContextRate(
@@ -466,12 +550,18 @@ export function calculateRequestCostBreakdown(
       above272kPriority: priceData.output_cost_per_token_above_272k_tokens_priority,
       above200k: priceData.output_cost_per_token_above_200k_tokens,
       above200kPriority: priceData.output_cost_per_token_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.output_cost_per_token_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.output_cost_per_token_above_200k_tokens,
     }
   );
   const cacheCreationAboveThreshold =
+    priorityTierPricing?.cache_creation_input_token_cost_above_272k_tokens ??
+    priorityTierPricing?.cache_creation_input_token_cost_above_200k_tokens ??
     priceData.cache_creation_input_token_cost_above_272k_tokens ??
     priceData.cache_creation_input_token_cost_above_200k_tokens;
   const cacheCreation1hAboveThreshold =
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr_above_272k_tokens ??
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr_above_200k_tokens ??
     priceData.cache_creation_input_token_cost_above_1hr_above_272k_tokens ??
     priceData.cache_creation_input_token_cost_above_1hr_above_200k_tokens ??
     cacheCreationAboveThreshold;
@@ -482,6 +572,8 @@ export function calculateRequestCostBreakdown(
       above272kPriority: priceData.cache_read_input_token_cost_above_272k_tokens_priority,
       above200k: priceData.cache_read_input_token_cost_above_200k_tokens,
       above200kPriority: priceData.cache_read_input_token_cost_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.cache_read_input_token_cost_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.cache_read_input_token_cost_above_200k_tokens,
     }
   );
   const longContextThreshold = resolveLongContextThreshold(priceData);
@@ -656,17 +748,24 @@ export function calculateRequestCost(
 
   const baseInputCostPerToken = priceData.input_cost_per_token;
   const baseOutputCostPerToken = priceData.output_cost_per_token;
-  const inputCostPerToken =
-    options.priorityServiceTierApplied &&
-    typeof priceData.input_cost_per_token_priority === "number"
-      ? priceData.input_cost_per_token_priority
-      : baseInputCostPerToken;
-  const outputCostPerToken =
-    options.priorityServiceTierApplied &&
-    typeof priceData.output_cost_per_token_priority === "number"
-      ? priceData.output_cost_per_token_priority
-      : baseOutputCostPerToken;
-  const inputCostPerRequest = priceData.input_cost_per_request;
+  const priorityTierPricing = getPriorityServiceTierPricing(
+    priceData,
+    options.priorityServiceTierApplied
+  );
+  const inputCostPerToken = resolvePriorityAwareRate(
+    options.priorityServiceTierApplied,
+    priorityTierPricing?.input_cost_per_token,
+    priceData.input_cost_per_token_priority,
+    baseInputCostPerToken
+  );
+  const outputCostPerToken = resolvePriorityAwareRate(
+    options.priorityServiceTierApplied,
+    priorityTierPricing?.output_cost_per_token,
+    priceData.output_cost_per_token_priority,
+    baseOutputCostPerToken
+  );
+  const inputCostPerRequest =
+    priorityTierPricing?.input_cost_per_request ?? priceData.input_cost_per_request;
   const longContextPricing = options.longContextPricing;
 
   if (
@@ -681,23 +780,27 @@ export function calculateRequestCost(
   }
 
   const cacheCreation5mCost =
+    priorityTierPricing?.cache_creation_input_token_cost ??
     priceData.cache_creation_input_token_cost ??
-    (baseInputCostPerToken != null ? baseInputCostPerToken * 1.25 : undefined);
+    (inputCostPerToken != null ? inputCostPerToken * 1.25 : undefined);
 
   const cacheCreation1hCost =
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr ??
     priceData.cache_creation_input_token_cost_above_1hr ??
-    (baseInputCostPerToken != null ? baseInputCostPerToken * 2 : undefined) ??
+    (inputCostPerToken != null ? inputCostPerToken * 2 : undefined) ??
     cacheCreation5mCost;
 
   const cacheReadCost =
-    (options.priorityServiceTierApplied &&
-    typeof priceData.cache_read_input_token_cost_priority === "number"
-      ? priceData.cache_read_input_token_cost_priority
-      : priceData.cache_read_input_token_cost) ??
-    (baseInputCostPerToken != null
-      ? baseInputCostPerToken * 0.1
-      : baseOutputCostPerToken != null
-        ? baseOutputCostPerToken * 0.1
+    resolvePriorityAwareRate(
+      options.priorityServiceTierApplied,
+      priorityTierPricing?.cache_read_input_token_cost,
+      priceData.cache_read_input_token_cost_priority,
+      priceData.cache_read_input_token_cost
+    ) ??
+    (inputCostPerToken != null
+      ? inputCostPerToken * 0.1
+      : outputCostPerToken != null
+        ? outputCostPerToken * 0.1
         : undefined);
 
   // Derive cache creation tokens by TTL
@@ -725,6 +828,8 @@ export function calculateRequestCost(
       above272kPriority: priceData.input_cost_per_token_above_272k_tokens_priority,
       above200k: priceData.input_cost_per_token_above_200k_tokens,
       above200kPriority: priceData.input_cost_per_token_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.input_cost_per_token_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.input_cost_per_token_above_200k_tokens,
     }
   );
   const outputAboveThreshold = resolvePriorityAwareLongContextRate(
@@ -734,12 +839,18 @@ export function calculateRequestCost(
       above272kPriority: priceData.output_cost_per_token_above_272k_tokens_priority,
       above200k: priceData.output_cost_per_token_above_200k_tokens,
       above200kPriority: priceData.output_cost_per_token_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.output_cost_per_token_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.output_cost_per_token_above_200k_tokens,
     }
   );
   const cacheCreationAboveThreshold =
+    priorityTierPricing?.cache_creation_input_token_cost_above_272k_tokens ??
+    priorityTierPricing?.cache_creation_input_token_cost_above_200k_tokens ??
     priceData.cache_creation_input_token_cost_above_272k_tokens ??
     priceData.cache_creation_input_token_cost_above_200k_tokens;
   const cacheCreation1hAboveThreshold =
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr_above_272k_tokens ??
+    priorityTierPricing?.cache_creation_input_token_cost_above_1hr_above_200k_tokens ??
     priceData.cache_creation_input_token_cost_above_1hr_above_272k_tokens ??
     priceData.cache_creation_input_token_cost_above_1hr_above_200k_tokens ??
     cacheCreationAboveThreshold;
@@ -750,6 +861,8 @@ export function calculateRequestCost(
       above272kPriority: priceData.cache_read_input_token_cost_above_272k_tokens_priority,
       above200k: priceData.cache_read_input_token_cost_above_200k_tokens,
       above200kPriority: priceData.cache_read_input_token_cost_above_200k_tokens_priority,
+      serviceTierAbove272k: priorityTierPricing?.cache_read_input_token_cost_above_272k_tokens,
+      serviceTierAbove200k: priorityTierPricing?.cache_read_input_token_cost_above_200k_tokens,
     }
   );
   const longContextThreshold = resolveLongContextThreshold(priceData);
diff --git a/src/lib/utils/model-price-fields.ts b/src/lib/utils/model-price-fields.ts
index dea76b635..24fab89d9 100644
--- a/src/lib/utils/model-price-fields.ts
+++ b/src/lib/utils/model-price-fields.ts
@@ -59,6 +59,11 @@ const SUPPORTED_LONG_CONTEXT_KEYS = new Set([
   "cache_read_input_token_cost",
 ]);
 
+const SUPPORTED_SERVICE_TIER_KEYS = new Set([
+  ...SUPPORTED_TOP_LEVEL_BILLING_KEYS,
+  "long_context_pricing",
+]);
+
 const CORE_TOP_LEVEL_FIELDS = new Set([
   "mode",
   "display_name",
@@ -124,7 +129,8 @@ export function isPriceLikeFieldPath(path: string): boolean {
 function classifyField(path: string, key: string): ModelPriceFieldKind {
   if (
     SUPPORTED_TOP_LEVEL_BILLING_KEYS.has(key) ||
-    (path.startsWith("long_context_pricing.") && SUPPORTED_LONG_CONTEXT_KEYS.has(key))
+    (path.startsWith("long_context_pricing.") && SUPPORTED_LONG_CONTEXT_KEYS.has(key)) ||
+    (path.startsWith("service_tier_pricing.") && SUPPORTED_SERVICE_TIER_KEYS.has(key))
   ) {
     return "supported";
   }
@@ -150,7 +156,11 @@ function isCoreField(
     return true;
   }
 
-  return CORE_TOP_LEVEL_FIELDS.has(key) || path.startsWith("long_context_pricing.");
+  return (
+    CORE_TOP_LEVEL_FIELDS.has(key) ||
+    path.startsWith("long_context_pricing.") ||
+    path.startsWith("service_tier_pricing.")
+  );
 }
 
 function pushEntries(
diff --git a/src/lib/utils/pricing-resolution.ts b/src/lib/utils/pricing-resolution.ts
index 4f3f7d4e0..d1d656476 100644
--- a/src/lib/utils/pricing-resolution.ts
+++ b/src/lib/utils/pricing-resolution.ts
@@ -68,7 +68,7 @@ const PROVIDER_DETAIL_FIELDS = [
   "input_cost_per_image_token",
 ] as const;
 
-const DETAIL_SCORE_OBJECT_FIELDS = ["long_context_pricing"] as const;
+const DETAIL_SCORE_OBJECT_FIELDS = ["long_context_pricing", "service_tier_pricing"] as const;
 
 const DETAIL_TIE_BREAK_ORDER = [
   "openrouter",
diff --git a/src/proxy.ts b/src/proxy.ts
index 65b1b4628..86471fed3 100644
--- a/src/proxy.ts
+++ b/src/proxy.ts
@@ -1,6 +1,6 @@
 import { type NextRequest, NextResponse } from "next/server";
 import createMiddleware from "next-intl/middleware";
-import { type Locale, localeCookieName } from "@/i18n/config";
+import { defaultLocale, type Locale, localeCookieName } from "@/i18n/config";
 import { getLocaleFromValue, normalizePathnameForLocaleNavigation } from "@/i18n/pathname";
 import { routing } from "@/i18n/routing";
 import { AUTH_COOKIE_NAME } from "@/lib/auth";
@@ -18,6 +18,7 @@ const PUBLIC_PATH_PATTERNS = [
 ];
 
 const API_PROXY_PATH = "/v1";
+const SYSTEM_STATUS_ALIAS_PATH = "/system-status";
 
 function matchesPublicPath(pathname: string, pattern: string) {
   return pathname === pattern || pathname.startsWith(`${pattern}/`);
@@ -46,6 +47,21 @@ function proxyHandler(request: NextRequest) {
     return NextResponse.next();
   }
 
+  if (pathname === SYSTEM_STATUS_ALIAS_PATH) {
+    const url = request.nextUrl.clone();
+    url.pathname = `/${defaultLocale}/status`;
+    return NextResponse.redirect(url);
+  }
+
+  const localePrefixedSystemStatusAlias = routing.locales.find(
+    (locale) => pathname === `/${locale}${SYSTEM_STATUS_ALIAS_PATH}`
+  );
+  if (localePrefixedSystemStatusAlias) {
+    const url = request.nextUrl.clone();
+    url.pathname = `/${localePrefixedSystemStatusAlias}/status`;
+    return NextResponse.redirect(url);
+  }
+
   const isLocalePrefixedPublicStatusPath = routing.locales.some(
     (locale) => pathname === `/${locale}/status` || pathname.startsWith(`/${locale}/status/`)
   );
diff --git a/src/types/model-price.ts b/src/types/model-price.ts
index 734897aac..035a98759 100644
--- a/src/types/model-price.ts
+++ b/src/types/model-price.ts
@@ -18,6 +18,27 @@ export interface LongContextPricing {
   cache_read_input_token_cost?: number;
 }
 
+export interface ServiceTierPricing {
+  input_cost_per_token?: number;
+  output_cost_per_token?: number;
+  cache_creation_input_token_cost?: number;
+  cache_creation_input_token_cost_above_1hr?: number;
+  cache_read_input_token_cost?: number;
+  input_cost_per_token_above_200k_tokens?: number;
+  output_cost_per_token_above_200k_tokens?: number;
+  cache_creation_input_token_cost_above_200k_tokens?: number;
+  cache_read_input_token_cost_above_200k_tokens?: number;
+  cache_creation_input_token_cost_above_1hr_above_200k_tokens?: number;
+  input_cost_per_token_above_272k_tokens?: number;
+  output_cost_per_token_above_272k_tokens?: number;
+  cache_creation_input_token_cost_above_272k_tokens?: number;
+  cache_read_input_token_cost_above_272k_tokens?: number;
+  cache_creation_input_token_cost_above_1hr_above_272k_tokens?: number;
+  input_cost_per_request?: number;
+  long_context_pricing?: LongContextPricing;
+  [key: string]: unknown;
+}
+
 export interface ModelPriceData {
   // 基础价格信息
   input_cost_per_token?: number;
@@ -73,6 +94,9 @@ export interface ModelPriceData {
   // 长上下文价格（例如 GPT-5.4 超过 272K 后的 premium 费率）
   long_context_pricing?: LongContextPricing;
 
+  // 服务等级定价（例如 OpenAI service_tier="priority"）
+  service_tier_pricing?: Record<string, ServiceTierPricing>;
+
   // 模型能力信息
   display_name?: string;
   litellm_provider?: string;
diff --git a/tests/integration/billing-model-source.test.ts b/tests/integration/billing-model-source.test.ts
index d9ef3e2d9..f543de375 100644
--- a/tests/integration/billing-model-source.test.ts
+++ b/tests/integration/billing-model-source.test.ts
@@ -630,6 +630,74 @@ describe("Billing model source - Redis session cost vs DB cost", () => {
     expect(rateLimitCosts[0]).toBe(64);
   });
 
+  it("codex fast: uses model-specific service_tier_pricing priority rates", async () => {
+    vi.mocked(getSystemSettings).mockResolvedValue(makeSystemSettings("redirected"));
+    vi.mocked(updateMessageRequestDetails).mockResolvedValue(undefined);
+    vi.mocked(updateMessageRequestDuration).mockResolvedValue(undefined);
+    vi.mocked(SessionManager.storeSessionResponse).mockResolvedValue(undefined);
+    vi.mocked(RateLimitService.trackUserDailyCost).mockResolvedValue(undefined);
+    vi.mocked(SessionTracker.refreshSession).mockResolvedValue(undefined);
+
+    vi.mocked(findLatestPriceByModel).mockImplementation(async (modelName: string) => {
+      if (modelName === "gpt-5.5") {
+        return makePriceRecord(modelName, {
+          mode: "responses",
+          model_family: "gpt",
+          litellm_provider: "chatgpt",
+          pricing: {
+            openai: {
+              input_cost_per_token: 0.000005,
+              output_cost_per_token: 0.00003,
+              cache_read_input_token_cost: 0.0000005,
+              input_cost_per_token_priority: 0.00001,
+              output_cost_per_token_priority: 0.00006,
+              cache_read_input_token_cost_priority: 0.000001,
+              service_tier_pricing: {
+                priority: {
+                  input_cost_per_token: 0.0000125,
+                  output_cost_per_token: 0.000075,
+                  cache_read_input_token_cost: 0.00000125,
+                },
+              },
+            },
+          },
+        });
+      }
+      return null;
+    });
+
+    const dbCosts: string[] = [];
+    vi.mocked(updateMessageRequestCostWithBreakdown).mockImplementation(
+      async (_id: number, costUsd: unknown) => {
+        dbCosts.push(String(costUsd));
+      }
+    );
+    const rateLimitCosts = captureRateLimitCosts();
+
+    const session = createSession({
+      originalModel: "gpt-5.5",
+      redirectedModel: "gpt-5.5",
+      sessionId: "sess-gpt55-priority-service-tier-pricing",
+      messageId: 3202,
+      providerOverrides: {
+        name: "ChatGPT",
+        url: "https://chatgpt.com/backend-api/codex",
+        providerType: "codex",
+      },
+      requestMessage: { service_tier: "priority" },
+    });
+
+    const response = createNonStreamResponse({
+      input_tokens: 1_000_000,
+      output_tokens: 1_000_000,
+    });
+    await ProxyResponseHandler.dispatch(session, response);
+    await drainAsyncTasks();
+
+    expect(dbCosts[0]).toBe("87.5");
+    expect(rateLimitCosts[0]).toBe(87.5);
+  });
+
   it("codex fast: uses long-context priority pricing when request is priority and response omits service_tier", async () => {
     vi.mocked(getSystemSettings).mockResolvedValue(makeSystemSettings("redirected"));
     vi.mocked(updateMessageRequestDetails).mockResolvedValue(undefined);
diff --git a/tests/unit/lib/cost-calculation-breakdown.test.ts b/tests/unit/lib/cost-calculation-breakdown.test.ts
index 8ea9a47dd..33266a021 100644
--- a/tests/unit/lib/cost-calculation-breakdown.test.ts
+++ b/tests/unit/lib/cost-calculation-breakdown.test.ts
@@ -134,6 +134,35 @@ describe("calculateRequestCostBreakdown", () => {
     expect(result.total).toBe(1904154);
   });
 
+  test("uses service_tier_pricing.priority in breakdown before legacy priority fields", () => {
+    const result = calculateRequestCostBreakdown(
+      {
+        input_tokens: 2,
+        output_tokens: 3,
+        cache_read_input_tokens: 5,
+      },
+      makePriceData({
+        input_cost_per_token_priority: 2,
+        output_cost_per_token_priority: 20,
+        cache_read_input_token_cost_priority: 0.2,
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 3,
+            output_cost_per_token: 30,
+            cache_read_input_token_cost: 0.3,
+          },
+        },
+      }),
+      false,
+      true
+    );
+
+    expect(result.input).toBe(6);
+    expect(result.output).toBe(90);
+    expect(result.cache_read).toBe(1.5);
+    expect(result.total).toBe(97.5);
+  });
+
   test("falls back to regular long-context pricing in breakdown when priority long-context fields are absent", () => {
     const result = calculateRequestCostBreakdown(
       {
diff --git a/tests/unit/lib/cost-calculation-priority.test.ts b/tests/unit/lib/cost-calculation-priority.test.ts
index 0342b94be..1794f2420 100644
--- a/tests/unit/lib/cost-calculation-priority.test.ts
+++ b/tests/unit/lib/cost-calculation-priority.test.ts
@@ -1,5 +1,5 @@
 import { describe, expect, test } from "vitest";
-import { calculateRequestCost } from "@/lib/utils/cost-calculation";
+import { calculateRequestCost, matchLongContextPricing } from "@/lib/utils/cost-calculation";
 import type { ModelPriceData } from "@/types/model-price";
 
 function makePriceData(overrides: Partial<ModelPriceData> = {}): ModelPriceData {
@@ -16,6 +16,79 @@ function makePriceData(overrides: Partial<ModelPriceData> = {}): ModelPriceData
 }
 
 describe("calculateRequestCost priority service tier", () => {
+  test("uses service_tier_pricing.priority before legacy priority fields", () => {
+    const cost = calculateRequestCost(
+      { input_tokens: 2, output_tokens: 3, cache_read_input_tokens: 5 },
+      makePriceData({
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 3,
+            output_cost_per_token: 30,
+            cache_read_input_token_cost: 0.3,
+          },
+        },
+      }),
+      1,
+      false,
+      true
+    );
+
+    expect(Number(cost.toString())).toBe(97.5);
+  });
+
+  test("keeps service_tier_pricing.priority scoped to priority requests", () => {
+    const cost = calculateRequestCost(
+      { input_tokens: 2, output_tokens: 3, cache_read_input_tokens: 5 },
+      makePriceData({
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 3,
+            output_cost_per_token: 30,
+            cache_read_input_token_cost: 0.3,
+          },
+        },
+      }),
+      1,
+      false,
+      false
+    );
+
+    expect(Number(cost.toString())).toBe(32.5);
+  });
+
+  test("allows different models to define different priority tier prices", () => {
+    const usage = { input_tokens: 1, output_tokens: 1, cache_read_input_tokens: 1 };
+    const gpt55Cost = calculateRequestCost(
+      usage,
+      makePriceData({
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 0.0000125,
+            output_cost_per_token: 0.000075,
+            cache_read_input_token_cost: 0.00000125,
+          },
+        },
+      }),
+      { priorityServiceTierApplied: true }
+    );
+    const gpt54Cost = calculateRequestCost(
+      usage,
+      makePriceData({
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 0.000005,
+            output_cost_per_token: 0.00003,
+            cache_read_input_token_cost: 0.0000005,
+          },
+        },
+      }),
+      { priorityServiceTierApplied: true }
+    );
+
+    expect(gpt55Cost.toNumber()).toBe(0.00008875);
+    expect(gpt54Cost.toNumber()).toBe(0.0000355);
+  });
+
   test("uses priority pricing fields when priority service tier is applied", () => {
     const cost = calculateRequestCost(
       { input_tokens: 2, output_tokens: 3, cache_read_input_tokens: 5 },
@@ -115,4 +188,32 @@ describe("calculateRequestCost priority service tier", () => {
 
     expect(Number(cost.toString())).toBe(1904147);
   });
+
+  test("uses service_tier_pricing.priority long_context_pricing when matched", () => {
+    const usage = {
+      input_tokens: 101,
+      output_tokens: 2,
+    };
+    const priceData = makePriceData({
+      service_tier_pricing: {
+        priority: {
+          input_cost_per_token: 4,
+          output_cost_per_token: 40,
+          long_context_pricing: {
+            threshold_tokens: 100,
+            input_multiplier: 2,
+            output_multiplier: 2,
+          },
+        },
+      },
+    });
+    const match = matchLongContextPricing(usage, priceData, "priority");
+    const cost = calculateRequestCost(usage, priceData, {
+      priorityServiceTierApplied: true,
+      longContextPricing: match?.pricing ?? null,
+    });
+
+    expect(match).not.toBeNull();
+    expect(Number(cost.toString())).toBe(968);
+  });
 });
diff --git a/tests/unit/lib/model-price-fields.test.ts b/tests/unit/lib/model-price-fields.test.ts
index ef7647b84..a917f8cb0 100644
--- a/tests/unit/lib/model-price-fields.test.ts
+++ b/tests/unit/lib/model-price-fields.test.ts
@@ -23,6 +23,13 @@ describe("model-price-fields", () => {
         threshold_tokens: 128000,
         input_cost_per_token: 0.000005,
       },
+      service_tier_pricing: {
+        priority: {
+          input_cost_per_token: 0.0000125,
+          output_cost_per_token: 0.000075,
+          cache_read_input_token_cost: 0.00000125,
+        },
+      },
       pricing: {
         openai: {
           input_cost_per_token: 0.0000025,
@@ -45,6 +52,10 @@ describe("model-price-fields", () => {
     expect(
       entries.find((entry) => entry.path === "long_context_pricing.input_cost_per_token")?.kind
     ).toBe("supported");
+    expect(
+      entries.find((entry) => entry.path === "service_tier_pricing.priority.input_cost_per_token")
+        ?.kind
+    ).toBe("supported");
   });
 
   test("extracts editable extra price data by excluding managed fields", () => {
diff --git a/tests/unit/lib/price-data-price-like-fields.test.ts b/tests/unit/lib/price-data-price-like-fields.test.ts
index 440e4427e..4e46538b3 100644
--- a/tests/unit/lib/price-data-price-like-fields.test.ts
+++ b/tests/unit/lib/price-data-price-like-fields.test.ts
@@ -23,6 +23,18 @@ describe("hasValidPriceData: generic price-like fields", () => {
     ).toBe(true);
   });
 
+  test("treats service tier pricing as valid price data", () => {
+    expect(
+      hasValidPriceData({
+        service_tier_pricing: {
+          priority: {
+            input_cost_per_token: 0.0000125,
+          },
+        },
+      })
+    ).toBe(true);
+  });
+
   test("ignores non price-like numeric metadata", () => {
     expect(
       hasValidPriceData({
diff --git a/tests/unit/lib/utils/pricing-resolution.test.ts b/tests/unit/lib/utils/pricing-resolution.test.ts
index d9196449e..0ff1d884d 100644
--- a/tests/unit/lib/utils/pricing-resolution.test.ts
+++ b/tests/unit/lib/utils/pricing-resolution.test.ts
@@ -225,4 +225,50 @@ describe("resolvePricingForModelRecords", () => {
       input_cost_per_token: 0.000005,
     });
   });
+
+  test("provider detail scoring counts service_tier_pricing", () => {
+    const cloudRecord = makeRecord("gpt-5.5", {
+      mode: "responses",
+      model_family: "gpt",
+      pricing: {
+        fallback: {
+          input_cost_per_token: 0.000005,
+          output_cost_per_token: 0.00003,
+        },
+        openai: {
+          input_cost_per_token: 0.000005,
+          output_cost_per_token: 0.00003,
+          service_tier_pricing: {
+            priority: {
+              input_cost_per_token: 0.0000125,
+              output_cost_per_token: 0.000075,
+              cache_read_input_token_cost: 0.00000125,
+            },
+          },
+        },
+      },
+    });
+
+    const resolved = resolvePricingForModelRecords({
+      provider: {
+        id: 5,
+        name: "OpenAI",
+        url: "https://api.openai.com/v1/responses",
+      } as never,
+      primaryModelName: "gpt-5.5",
+      fallbackModelName: null,
+      primaryRecord: cloudRecord,
+      fallbackRecord: null,
+    });
+
+    expect(resolved).not.toBeNull();
+    expect(resolved?.resolvedPricingProviderKey).toBe("openai");
+    expect(resolved?.priceData.service_tier_pricing).toEqual({
+      priority: {
+        input_cost_per_token: 0.0000125,
+        output_cost_per_token: 0.000075,
+        cache_read_input_token_cost: 0.00000125,
+      },
+    });
+  });
 });