diff --git a/litellm/litellm_core_utils/llm_cost_calc/utils.py b/litellm/litellm_core_utils/llm_cost_calc/utils.py
index bdbaee00c19..7ba932458e5 100644
--- a/litellm/litellm_core_utils/llm_cost_calc/utils.py
+++ b/litellm/litellm_core_utils/llm_cost_calc/utils.py
@@ -126,7 +126,7 @@ def select_cost_metric_for_model(
     """
     if model_info.get("input_cost_per_character"):
         return "cost_per_character"
-    elif model_info.get("input_cost_per_token"):
+    elif model_info.get("input_cost_per_token") is not None:
         return "cost_per_token"
     else:
         raise ValueError(
diff --git a/tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py b/tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py
index ce90719789a..bd78702d6cb 100644
--- a/tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py
+++ b/tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py
@@ -3841,3 +3841,18 @@ def test_generic_cost_per_token_grok_46_long_context(_local_model_cost_map):
     )
     assert prompt_cost == pytest.approx(200_000 * 4e-06 + 50_000 * 1e-06)
     assert completion_cost == pytest.approx(1_000 * 1.2e-05)
+
+
+def test_select_cost_metric_for_model_treats_zero_token_cost_as_priced():
+    from litellm.litellm_core_utils.llm_cost_calc.utils import select_cost_metric_for_model
+
+    free_model = {"key": "free-model", "input_cost_per_token": 0.0}
+    assert select_cost_metric_for_model(free_model) == "cost_per_token"
+
+    char_model = {"key": "char-model", "input_cost_per_character": 1e-6, "input_cost_per_token": 0.0}
+    assert select_cost_metric_for_model(char_model) == "cost_per_character"
+
+    import pytest
+
+    with pytest.raises(ValueError):
+        select_cost_metric_for_model({"key": "no-pricing"})