diff --git a/litellm/litellm_core_utils/llm_cost_calc/utils.py b/litellm/litellm_core_utils/llm_cost_calc/utils.py
index cf1b1962df8..e24fa004448 100644
--- a/litellm/litellm_core_utils/llm_cost_calc/utils.py
+++ b/litellm/litellm_core_utils/llm_cost_calc/utils.py
@@ -74,12 +74,9 @@ def _uses_inclusive_token_thresholds(custom_llm_provider: str | None) -> bool:
 
 
 def apply_provider_cache_read_default(model_info: ModelInfo, custom_llm_provider: str | None) -> ModelInfo:
-    """Dispatch to the provider's cache-read pricing default; providers without one keep their entry as is."""
-    match custom_llm_provider:
-        case "fireworks_ai":
-            return with_default_cache_read_rate(model_info)
-        case _:
-            return model_info
+    if custom_llm_provider == "fireworks_ai":
+        return with_default_cache_read_rate(model_info)
+    return model_info
 
 
 def _get_token_detail_value(details: object, key: str) -> int | None:
diff --git a/litellm/llms/fireworks_ai/cache_pricing.py b/litellm/llms/fireworks_ai/cache_pricing.py
index c28a8684c66..f5e49cad01a 100644
--- a/litellm/llms/fireworks_ai/cache_pricing.py
+++ b/litellm/llms/fireworks_ai/cache_pricing.py
@@ -1,7 +1,3 @@
-"""
-Fireworks AI serverless cache-read pricing defaults.
-"""
-
 from typing import (
     Final,
     cast,  # noqa: TID251  # the derived entry is a dict copy of a ReadOnly TypedDict; no cast-free way to retype it
@@ -11,16 +7,24 @@ from litellm.constants import FIREWORKS_AI_DEFAULT_CACHE_READ_RATE_RATIO
 from litellm.types.utils import ModelInfo
 
 
+def _as_rate(value: object) -> float | None:
+    if isinstance(value, bool) or not isinstance(value, (int, float, str)):
+        return None
+    try:
+        return float(value)
+    except ValueError:
+        return None
+
+
 def with_default_cache_read_rate(model_info: ModelInfo) -> ModelInfo:
-    """Entries without a cache-read rate get the documented discount off the input rate; the shared map is
-    never mutated, so a copy carries it."""
-    input_rate: Final = model_info.get("input_cost_per_token")
+    input_rate: Final = _as_rate(model_info.get("input_cost_per_token"))
     if model_info.get("cache_read_input_token_cost") is not None or input_rate is None:
         return model_info
     cache_read_rate: Final = input_rate * FIREWORKS_AI_DEFAULT_CACHE_READ_RATE_RATIO
     off_peak: Final = model_info.get("off_peak_pricing")
     if off_peak is None or "cache_read_input_token_cost" in off_peak:
         return cast(ModelInfo, {**model_info, "cache_read_input_token_cost": cache_read_rate})
+    off_peak_input_rate: Final = _as_rate(off_peak.get("input_cost_per_token"))
     return cast(
         ModelInfo,
         {
@@ -29,8 +33,8 @@ def with_default_cache_read_rate(model_info: ModelInfo) -> ModelInfo:
             "off_peak_pricing": {
                 **off_peak,
                 "cache_read_input_token_cost": (
-                    off_peak["input_cost_per_token"] * FIREWORKS_AI_DEFAULT_CACHE_READ_RATE_RATIO
-                    if "input_cost_per_token" in off_peak
+                    off_peak_input_rate * FIREWORKS_AI_DEFAULT_CACHE_READ_RATE_RATIO
+                    if off_peak_input_rate is not None
                     else cache_read_rate
                 ),
             },
diff --git a/tests/test_litellm/llms/fireworks_ai/test_fireworks_ai_cache_pricing.py b/tests/test_litellm/llms/fireworks_ai/test_fireworks_ai_cache_pricing.py
index 2ab273b27c6..c21943cfe75 100644
--- a/tests/test_litellm/llms/fireworks_ai/test_fireworks_ai_cache_pricing.py
+++ b/tests/test_litellm/llms/fireworks_ai/test_fireworks_ai_cache_pricing.py
@@ -48,3 +48,18 @@ def test_off_peak_window_without_its_own_input_rate_reuses_the_standard_derived_
     derived = with_default_cache_read_rate(model_info)
 
     assert derived["off_peak_pricing"]["cache_read_input_token_cost"] == derived["cache_read_input_token_cost"]
+
+
+def test_string_rates_from_config_are_coerced_before_the_discount_is_applied() -> None:
+    model_info: ModelInfo = {
+        "input_cost_per_token": "2e-6",
+        "off_peak_pricing": {
+            "hours_utc": "14:00-00:00",
+            "input_cost_per_token": "1e-6",
+        },
+    }
+
+    derived = with_default_cache_read_rate(model_info)
+
+    assert derived["cache_read_input_token_cost"] == pytest.approx(1e-6)
+    assert derived["off_peak_pricing"]["cache_read_input_token_cost"] == pytest.approx(5e-7)