diff --git a/src/polygateway/client.py b/src/polygateway/client.py
index 1bfbd0b..dac5f5c 100644
--- a/src/polygateway/client.py
+++ b/src/polygateway/client.py
@@ -134,6 +134,7 @@ class GatewayClient:
         retry: RetryPolicy,
         backpressure: BackpressurePolicy,
         quota_full: str = "wait",
+        circuit_open: str = "fail_fast",
         telemetry: TelemetryRecorder | None = None,
         pricing: PricingTable | None = None,
         text_cap: int | None = None,
@@ -162,6 +163,7 @@ class GatewayClient:
             retry=retry,
             backpressure=backpressure,
             quota_full=quota_full,
+            circuit_open=circuit_open,
             cooldown_memo=SourceCooldownMemo(now=now),
             # AIMD ceiling 尊重源级静态并发上限(独立核验 I1: 不得静默钳制大于 64 的配置)
             pacer=AdaptivePacer(
@@ -313,6 +315,7 @@ class GatewayClient:
             retry=settings.retry,
             backpressure=settings.backpressure,
             quota_full=settings.quota_full,
+            circuit_open=settings.circuit_open,
             telemetry=telemetry if telemetry is not None else _build_telemetry(settings),
             pricing=PricingTable.from_file(settings.pricing_path)
             if settings.pricing_path is not None
diff --git a/src/polygateway/config.py b/src/polygateway/config.py
index 8dfe5b2..4f1b839 100644
--- a/src/polygateway/config.py
+++ b/src/polygateway/config.py
@@ -50,6 +50,9 @@ _SOURCE_FIELDS: dict[str, tuple[str, str]] = {
 _RESERVED_SEGMENTS = frozenset({"GLOBAL", "RETRY", "BREAKER", "BACKPRESSURE"})
 _SELECTORS = frozenset({"round_robin", "least_inflight", "health_aware"})
 _QUOTA_FULL = frozenset({"wait", "fail_fast"})
+# 熔断全拒时的处置(issue #14);值域与 _QUOTA_FULL 相同但语义不同——配额满是
+# "排队等自己的份额"(必然轮到),熔断开路是"等源恢复"(未必恢复),故分列两键
+_CIRCUIT_OPEN = frozenset({"wait", "fail_fast"})
 # 后端合法域: env 解析与构造期校验共用一份定义,避免两处分叉
 _LIMITER_BACKENDS = frozenset({"memory", "redis"})
 _BREAKER_BACKENDS = frozenset({"memory", "redis"})
@@ -128,6 +131,9 @@ class GatewaySettings:
     backpressure: BackpressurePolicy
     selector: str
     quota_full: str
+    # 熔断全拒时是当场判死还是等冷却过去(issue #14);缺省 fail_fast 保持
+    # 存量下游的控制流不变,单源 scope 应显式配 wait
+    circuit_open: str
     limiter_backend: str
     breaker_backend: str
     cache_backend: str
@@ -203,6 +209,7 @@ class GatewaySettings:
             ("telemetry_backend", _TELEMETRY_BACKENDS),
             ("selector", _SELECTORS),
             ("quota_full", _QUOTA_FULL),
+            ("circuit_open", _CIRCUIT_OPEN),
         ):
             value = getattr(self, field)
             if value not in allowed:
@@ -320,6 +327,9 @@ class GatewaySettings:
             backpressure=_load_backpressure(scope_u, env),
             selector=_load_choice(env, f"{scope_u}__SELECTOR", _SELECTORS, "health_aware"),
             quota_full=_load_choice(env, f"{scope_u}__QUOTA_FULL", _QUOTA_FULL, "wait"),
+            circuit_open=_load_choice(
+                env, f"{scope_u}__CIRCUIT_OPEN", _CIRCUIT_OPEN, "fail_fast"
+            ),
             **_load_pgw(env),
         )
 
diff --git a/src/polygateway/embedding.py b/src/polygateway/embedding.py
index dcf8c2a..f06b97c 100644
--- a/src/polygateway/embedding.py
+++ b/src/polygateway/embedding.py
@@ -101,6 +101,7 @@ class EmbeddingClient:
         retry: RetryPolicy,
         backpressure: BackpressurePolicy,
         quota_full: str = "wait",
+        circuit_open: str = "fail_fast",
         telemetry: TelemetryRecorder | None = None,
         pricing: PricingTable | None = None,
         text_cap: int | None = None,
@@ -113,8 +114,6 @@ class EmbeddingClient:
     ) -> None:
         if batch_size < 1:
             raise ValueError("batch_size 必须 ≥ 1")
-        if quota_full not in ("wait", "fail_fast"):
-            raise ValueError(f"quota_full 必须是 wait|fail_fast: {quota_full!r}")
         if expected_dim is not None and expected_dim < 1:
             raise ValueError("expected_dim 必须 ≥ 1")
         self._scope = scope
@@ -145,6 +144,7 @@ class EmbeddingClient:
             breaker=self._breaker,
             backpressure=backpressure,
             quota_full=quota_full,
+            circuit_open=circuit_open,
             now=now,
             sleep=sleep,
             rng=rng,
@@ -500,6 +500,7 @@ class EmbeddingClient:
             retry=gw.retry,
             backpressure=gw.backpressure,
             quota_full=gw.quota_full,
+            circuit_open=gw.circuit_open,
             telemetry=telemetry if telemetry is not None else _build_telemetry(gw),
             pricing=PricingTable.from_file(gw.pricing_path)
             if gw.pricing_path is not None
diff --git a/src/polygateway/middleware/admission.py b/src/polygateway/middleware/admission.py
index a817523..8997784 100644
--- a/src/polygateway/middleware/admission.py
+++ b/src/polygateway/middleware/admission.py
@@ -29,6 +29,9 @@ from loguru import logger
 from polygateway.errors import AllSourcesExhausted, CircuitOpenError
 from polygateway.sources import SourceCooldownMemo
 
+# 两个准入策略键共用的值域;校验只此一处,不在各客户端重复
+_POLICIES = frozenset({"wait", "fail_fast"})
+
 if TYPE_CHECKING:
     from collections.abc import Callable
 
@@ -137,6 +140,7 @@ class SourceAdmission:
         breaker: BreakerGate,
         backpressure: BackpressurePolicy,
         quota_full: str,
+        circuit_open: str,
         memo: SourceCooldownMemo | None = None,
         pacer: AdaptivePacer | None = None,
         health_view: Callable[[str], float] | None = None,
@@ -144,8 +148,9 @@ class SourceAdmission:
         sleep: Callable[[float], object] = asyncio.sleep,
         rng: Callable[[], float] = random.random,
     ) -> None:
-        if quota_full not in ("wait", "fail_fast"):
-            raise ValueError(f"quota_full 必须是 wait|fail_fast: {quota_full!r}")
+        for name, value in (("quota_full", quota_full), ("circuit_open", circuit_open)):
+            if value not in _POLICIES:
+                raise ValueError(f"{name} 必须是 wait|fail_fast: {value!r}")
         self._scope = scope
         self._sources = sources
         self._selector = selector
@@ -153,6 +158,7 @@ class SourceAdmission:
         self._breaker = breaker
         self._bp = backpressure
         self._quota_full = quota_full
+        self._circuit_open = circuit_open
         self._memo = memo or SourceCooldownMemo(now=now)
         self._pacer = pacer
         self._health_view = health_view
diff --git a/src/polygateway/middleware/retry.py b/src/polygateway/middleware/retry.py
index c30f2ab..0e9b00c 100644
--- a/src/polygateway/middleware/retry.py
+++ b/src/polygateway/middleware/retry.py
@@ -178,6 +178,7 @@ class RetryMW:
         retry: RetryPolicy,
         backpressure: BackpressurePolicy,
         quota_full: str = "wait",
+        circuit_open: str = "fail_fast",
         cooldown_memo: SourceCooldownMemo | None = None,
         pacer: AdaptivePacer | None = None,
         emitter: object | None = None,
@@ -185,8 +186,6 @@ class RetryMW:
         sleep: Callable[[float], Awaitable[None]] = asyncio.sleep,
         rng: Callable[[], float] = random.random,
     ) -> None:
-        if quota_full not in ("wait", "fail_fast"):
-            raise ValueError(f"quota_full 必须是 wait|fail_fast: {quota_full!r}")
         self._scope = scope
         self._sources = list(sources)
         # 记账写回与 pacer 结算仍在 `_attempt` 内,故这三者由本类持有并与
@@ -212,6 +211,7 @@ class RetryMW:
             breaker=self._breaker,
             backpressure=backpressure,
             quota_full=quota_full,
+            circuit_open=circuit_open,
             memo=cooldown_memo,
             pacer=self._pacer,
             health_view=self._outcome_sink.health if self._outcome_sink else None,
diff --git a/src/polygateway/ocr.py b/src/polygateway/ocr.py
index bd06daa..4f6da20 100644
--- a/src/polygateway/ocr.py
+++ b/src/polygateway/ocr.py
@@ -107,14 +107,13 @@ class OcrClient:
         retry: RetryPolicy,
         backpressure: BackpressurePolicy,
         quota_full: str = "wait",
+        circuit_open: str = "fail_fast",
         telemetry: TelemetryRecorder | None = None,
         text_cap: int | None = None,
         now: Callable[[], float] = time.monotonic,
         sleep: Callable[[float], Awaitable[None]] = asyncio.sleep,
         rng: Callable[[], float] = random.random,
     ) -> None:
-        if quota_full not in ("wait", "fail_fast"):
-            raise ValueError(f"quota_full 必须是 wait|fail_fast: {quota_full!r}")
         self._scope = scope
         # MonkeyOCR 只发 multipart 表单,带 extra_body 的源必须先剥离,否则
         # 遥测会记录一个从未发出的采样参数(issue #4 决策 G)
@@ -139,6 +138,7 @@ class OcrClient:
             breaker=self._breaker,
             backpressure=backpressure,
             quota_full=quota_full,
+            circuit_open=circuit_open,
             now=now,
             sleep=sleep,
             rng=rng,
@@ -513,6 +513,7 @@ class OcrClient:
             retry=gw.retry,
             backpressure=gw.backpressure,
             quota_full=gw.quota_full,
+            circuit_open=gw.circuit_open,
             telemetry=telemetry if telemetry is not None else _build_telemetry(gw),
             # OCR 行与 chat 行写同一张 llm_calls;漏传这一条,同表内就一半受控
             # 一半不受控(issue #12)
diff --git a/tests/unit/test_config.py b/tests/unit/test_config.py
index 4f9772f..74011eb 100644
--- a/tests/unit/test_config.py
+++ b/tests/unit/test_config.py
@@ -170,6 +170,20 @@ class TestResilienceKeys:
         with pytest.raises(ValueError, match="probe"):
             GatewaySettings.from_env("LLM", env=_env(**{"LLM__BREAKER__PROBE_TTL_S": "45"}))
 
+    def test_circuit_open_defaults_to_fail_fast(self):
+        """issue #14: 熔断拒绝的处置策略。
+
+        缺省**不跟随** quota_full 的 wait——把最坏墙钟从毫秒抬到 stall 窗口
+        是"快速失败 → 长时间挂起"这个最危险的方向,不能强加给存量下游。
+        """
+        assert GatewaySettings.from_env("LLM", env=_env()).circuit_open == "fail_fast"
+        waiting = GatewaySettings.from_env(
+            "LLM", env=_env(**{"LLM__CIRCUIT_OPEN": "wait"})
+        )
+        assert waiting.circuit_open == "wait"
+        with pytest.raises(ValueError, match="CIRCUIT_OPEN"):
+            GatewaySettings.from_env("LLM", env=_env(**{"LLM__CIRCUIT_OPEN": "block"}))
+
     def test_selector_and_quota_full(self):
         # M2.5: 缺省选源改 health_aware(生产级默认);显式配置者不变
         s = GatewaySettings.from_env("LLM", env=_env())
@@ -589,6 +603,7 @@ class TestCrossFieldInvariants:
             ("telemetry_backend", "redis"),
             ("selector", "random"),
             ("quota_full", "block"),
+            ("circuit_open", "block"),
         ],
     )
     def test_enum_field_rejects_value_outside_domain(self, field, bad_value):