diff --git a/src/hooks/runtime-fallback/auto-retry-signal.test.ts b/src/hooks/runtime-fallback/auto-retry-signal.test.ts
index e734b309b..e485fbd6c 100644
--- a/src/hooks/runtime-fallback/auto-retry-signal.test.ts
+++ b/src/hooks/runtime-fallback/auto-retry-signal.test.ts
@@ -15,6 +15,7 @@ describe("extractAutoRetrySignal", () => {
     //#then
     expect(signal).toBeDefined()
     expect(signal?.signal).toContain("exceeded")
+    expect(signal?.signal).toContain("usage quota")
   })
 
   test("detects standard 'quota exceeded' signal", () => {
diff --git a/src/hooks/runtime-fallback/constants.ts b/src/hooks/runtime-fallback/constants.ts
index e835b4099..f407ffea0 100644
--- a/src/hooks/runtime-fallback/constants.ts
+++ b/src/hooks/runtime-fallback/constants.ts
@@ -28,7 +28,7 @@ export const RETRYABLE_ERROR_PATTERNS = [
   /quota\s+will\s+reset\s+after/i,
   /quota.?exceeded/i,
   /exceeded.*quota/i,
-  /usage.?quota/i,
+  /usage\s*quota/i,
   /exhausted\s+your\s+capacity/i,
   /all\s+credentials\s+for\s+model/i,
   /cool(?:ing)?\s+down/i,
diff --git a/src/hooks/runtime-fallback/error-classifier.ts b/src/hooks/runtime-fallback/error-classifier.ts
index 614023f1c..3bb46454c 100644
--- a/src/hooks/runtime-fallback/error-classifier.ts
+++ b/src/hooks/runtime-fallback/error-classifier.ts
@@ -126,6 +126,8 @@ export function classifyErrorType(error: unknown): string | undefined {
     errorName?.includes("insufficientquota") ||
     errorName?.includes("billingerror") ||
     /quota.?exceeded/i.test(message) ||
+    /exceeded.*quota/i.test(message) ||
+    /usage\s*quota/i.test(message) ||
     /subscription.*quota/i.test(message) ||
     /insufficient.?(?:quota|balance|funds?)/i.test(message) ||
     /billing.?(?:hard.?)?limit/i.test(message) ||
diff --git a/src/hooks/runtime-fallback/quota-error-classifier.regression.test.ts b/src/hooks/runtime-fallback/quota-error-classifier.regression.test.ts
index 737db208c..5878e8f2a 100644
--- a/src/hooks/runtime-fallback/quota-error-classifier.regression.test.ts
+++ b/src/hooks/runtime-fallback/quota-error-classifier.regression.test.ts
@@ -57,7 +57,7 @@ describe("runtime-fallback quota error regressions", () => {
     expect(retryable).toBe(true)
   })
 
-  test("classifies Volcano Engine 'exceeded the usage quota' as retryable", () => {
+  test("classifies Volcano Engine 'exceeded the usage quota' as quota_exceeded and retryable", () => {
     //#given
     const error = {
       name: "SessionRetry",
@@ -65,9 +65,11 @@ describe("runtime-fallback quota error regressions", () => {
     }
 
     //#when
+    const errorType = classifyErrorType(error)
     const retryable = isRetryableError(error, [429, 500, 502, 503, 504])
 
     //#then
+    expect(errorType).toBe("quota_exceeded")
     // Volcano Engine quota errors trigger fallback to the next model
     expect(retryable).toBe(true)
   })