diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index 2788ba69dab..200641c7388 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -20431,17 +20431,17 @@
         "supports_minimal_reasoning_effort": true
     },
     "gpt-5.1-mini": {
-        "cache_read_input_token_cost": 1.25e-07,
-        "cache_read_input_token_cost_priority": 2.5e-07,
-        "input_cost_per_token": 1.25e-06,
-        "input_cost_per_token_priority": 2.5e-06,
+        "cache_read_input_token_cost": 3.75e-08,
+        "cache_read_input_token_cost_priority": 7.5e-08,
+        "input_cost_per_token": 3.75e-07,
+        "input_cost_per_token_priority": 7.5e-07,
         "litellm_provider": "openai",
         "max_input_tokens": 272000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
-        "output_cost_per_token": 1e-05,
-        "output_cost_per_token_priority": 2e-05,
+        "output_cost_per_token": 3e-06,
+        "output_cost_per_token_priority": 6e-06,
         "supported_endpoints": [
             "/v1/chat/completions",
             "/v1/responses"
@@ -20470,17 +20470,17 @@
         "supports_minimal_reasoning_effort": true
     },
     "gpt-5.1-nano": {
-        "cache_read_input_token_cost": 1.25e-07,
-        "cache_read_input_token_cost_priority": 2.5e-07,
-        "input_cost_per_token": 1.25e-06,
-        "input_cost_per_token_priority": 2.5e-06,
+        "cache_read_input_token_cost": 1e-08,
+        "cache_read_input_token_cost_priority": 2e-08,
+        "input_cost_per_token": 1e-07,
+        "input_cost_per_token_priority": 2e-07,
         "litellm_provider": "openai",
         "max_input_tokens": 272000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
-        "output_cost_per_token": 1e-05,
-        "output_cost_per_token_priority": 2e-05,
+        "output_cost_per_token": 8e-07,
+        "output_cost_per_token_priority": 1.6e-06,
         "supported_endpoints": [
             "/v1/chat/completions",
             "/v1/responses"
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index d47fa9166aa..ae3101747e1 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -20467,17 +20467,17 @@
         "supports_minimal_reasoning_effort": true
     },
     "gpt-5.1-mini": {
-        "cache_read_input_token_cost": 1.25e-07,
-        "cache_read_input_token_cost_priority": 2.5e-07,
-        "input_cost_per_token": 1.25e-06,
-        "input_cost_per_token_priority": 2.5e-06,
+        "cache_read_input_token_cost": 3.75e-08,
+        "cache_read_input_token_cost_priority": 7.5e-08,
+        "input_cost_per_token": 3.75e-07,
+        "input_cost_per_token_priority": 7.5e-07,
         "litellm_provider": "openai",
         "max_input_tokens": 272000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
-        "output_cost_per_token": 1e-05,
-        "output_cost_per_token_priority": 2e-05,
+        "output_cost_per_token": 3e-06,
+        "output_cost_per_token_priority": 6e-06,
         "supported_endpoints": [
             "/v1/chat/completions",
             "/v1/responses"
@@ -20506,17 +20506,17 @@
         "supports_minimal_reasoning_effort": true
     },
     "gpt-5.1-nano": {
-        "cache_read_input_token_cost": 1.25e-07,
-        "cache_read_input_token_cost_priority": 2.5e-07,
-        "input_cost_per_token": 1.25e-06,
-        "input_cost_per_token_priority": 2.5e-06,
+        "cache_read_input_token_cost": 1e-08,
+        "cache_read_input_token_cost_priority": 2e-08,
+        "input_cost_per_token": 1e-07,
+        "input_cost_per_token_priority": 2e-07,
         "litellm_provider": "openai",
         "max_input_tokens": 272000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
-        "output_cost_per_token": 1e-05,
-        "output_cost_per_token_priority": 2e-05,
+        "output_cost_per_token": 8e-07,
+        "output_cost_per_token_priority": 1.6e-06,
         "supported_endpoints": [
             "/v1/chat/completions",
             "/v1/responses"
diff --git a/tests/test_litellm/llms/azure/chat/test_azure_gpt5_transformation.py b/tests/test_litellm/llms/azure/chat/test_azure_gpt5_transformation.py
index 081a7048e75..f6af3a5faa3 100644
--- a/tests/test_litellm/llms/azure/chat/test_azure_gpt5_transformation.py
+++ b/tests/test_litellm/llms/azure/chat/test_azure_gpt5_transformation.py
@@ -16,7 +16,6 @@ def use_local_model_cost_map(monkeypatch: pytest.MonkeyPatch):
     monkeypatch.setattr(
         litellm, "model_cost", get_model_cost_map(url=litellm.model_cost_map_url)
     )
-    litellm.add_known_models(model_cost_map=litellm.model_cost)
 
 
 def test_azure_gpt5_supports_reasoning_effort(config: AzureOpenAIGPT5Config):