From 0f15f7857f37c27e9b639ee53b68869075faa4e8 Mon Sep 17 00:00:00 2001
From: mgalbato <37748295+mgalbato@users.noreply.github.com>
Date: Tue, 23 Jun 2026 16:38:35 -0400
Subject: [PATCH] test(model_prices): update data residency tests to use
 gpt-5.4 as the uplift model

The tests were using gpt-5 which no longer carries the regional processing
uplift after correcting which models have it. Switch to gpt-5.4 (released
2026-03-05, the cutoff date) and add a regression parametrize covering
all pre-cutoff models to pin that they stay uplift-free.
---
 .../llm_cost_calc/test_llm_cost_calc_utils.py | 34 ++++++++++++++-----
 1 file changed, 26 insertions(+), 8 deletions(-)

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 9b3152fae07..7f3d5a959a1 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
@@ -1499,19 +1499,20 @@ def _local_model_cost_map():
 
 @pytest.mark.parametrize("data_residency", ["eu", "us"])
 def test_data_residency_applies_uplift(data_residency, _local_model_cost_map):
-    """gpt-5 should apply the regional processing uplift multiplier when
-    data_residency is set."""
+    """gpt-5.4 should apply the regional processing uplift multiplier when
+    data_residency is set. gpt-5.4+ (released 2026-03-05) carry the 10% uplift;
+    gpt-5 and older models do not."""
     from litellm.types.utils import Usage
 
     usage = Usage(prompt_tokens=1000, completion_tokens=500, total_tokens=1500)
 
     base = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
     )
     regional = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
         data_residency=data_residency,
@@ -1526,6 +1527,23 @@ def test_data_residency_applies_uplift(data_residency, _local_model_cost_map):
     assert regional[1] == pytest.approx(base[1] * 1.10, rel=1e-9)
 
 
+@pytest.mark.parametrize("model", ["gpt-5", "gpt-5-mini", "gpt-5-nano", "gpt-5-pro", "gpt-4o", "gpt-4.1"])
+def test_data_residency_no_uplift_for_pre_march_2026_models(model, _local_model_cost_map):
+    """Models released before 2026-03-05 must not have the regional uplift."""
+    from litellm.types.utils import Usage
+
+    usage = Usage(prompt_tokens=1000, completion_tokens=500, total_tokens=1500)
+
+    base = generic_cost_per_token(model=model, usage=usage, custom_llm_provider="openai")
+    regional = generic_cost_per_token(
+        model=model, usage=usage, custom_llm_provider="openai", data_residency="eu"
+    )
+
+    assert base == regional, (
+        f"{model} should not have a regional uplift, but cost changed with data_residency"
+    )
+
+
 def test_data_residency_no_uplift_for_unmarked_model(_local_model_cost_map):
     """A model without a regional_processing_uplift_multiplier_* entry should
     fall back to base pricing, not error."""
@@ -1555,12 +1573,12 @@ def test_data_residency_none_no_uplift(_local_model_cost_map):
     usage = Usage(prompt_tokens=1000, completion_tokens=500, total_tokens=1500)
 
     base = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
     )
     explicit_none = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
         data_residency=None,
@@ -1576,13 +1594,13 @@ def test_data_residency_composes_with_service_tier(_local_model_cost_map):
     usage = Usage(prompt_tokens=1000, completion_tokens=500, total_tokens=1500)
 
     priority_base = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
         service_tier="priority",
     )
     priority_eu = generic_cost_per_token(
-        model="gpt-5",
+        model="gpt-5.4",
         usage=usage,
         custom_llm_provider="openai",
         service_tier="priority",