diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index 34c9f98077b..e7a63fbe215 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -6501,6 +6501,7 @@
         "supports_web_search": true
     },
     "azure/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.25e-06,
         "cache_read_input_token_cost": 5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1e-06,
         "cache_read_input_token_cost_priority": 1e-06,
@@ -6551,6 +6552,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.25e-06,
         "cache_read_input_token_cost": 5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1e-06,
         "cache_read_input_token_cost_priority": 1e-06,
@@ -6602,6 +6604,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.5e-06,
         "cache_read_input_token_cost": 2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4e-07,
         "cache_read_input_token_cost_priority": 4e-07,
@@ -6653,6 +6656,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.5e-07,
         "cache_read_input_token_cost": 2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4e-08,
         "cache_read_input_token_cost_priority": 4e-08,
@@ -6704,6 +6708,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6751,6 +6756,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6799,6 +6805,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.75e-06,
         "cache_read_input_token_cost": 2.2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-07,
         "cache_read_input_token_cost_priority": 5.5e-07,
@@ -6847,6 +6854,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.75e-07,
         "cache_read_input_token_cost": 2.2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-08,
         "cache_read_input_token_cost_priority": 5.5e-08,
@@ -6895,6 +6903,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6942,6 +6951,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6990,6 +7000,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.75e-06,
         "cache_read_input_token_cost": 2.2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-07,
         "cache_read_input_token_cost_priority": 5.5e-07,
@@ -7038,6 +7049,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.75e-07,
         "cache_read_input_token_cost": 2.2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-08,
         "cache_read_input_token_cost_priority": 5.5e-08,
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index 34c9f98077b..e7a63fbe215 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -6501,6 +6501,7 @@
         "supports_web_search": true
     },
     "azure/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.25e-06,
         "cache_read_input_token_cost": 5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1e-06,
         "cache_read_input_token_cost_priority": 1e-06,
@@ -6551,6 +6552,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.25e-06,
         "cache_read_input_token_cost": 5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1e-06,
         "cache_read_input_token_cost_priority": 1e-06,
@@ -6602,6 +6604,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.5e-06,
         "cache_read_input_token_cost": 2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4e-07,
         "cache_read_input_token_cost_priority": 4e-07,
@@ -6653,6 +6656,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.5e-07,
         "cache_read_input_token_cost": 2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4e-08,
         "cache_read_input_token_cost_priority": 4e-08,
@@ -6704,6 +6708,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6751,6 +6756,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6799,6 +6805,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.75e-06,
         "cache_read_input_token_cost": 2.2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-07,
         "cache_read_input_token_cost_priority": 5.5e-07,
@@ -6847,6 +6854,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/us/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.75e-07,
         "cache_read_input_token_cost": 2.2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-08,
         "cache_read_input_token_cost_priority": 5.5e-08,
@@ -6895,6 +6903,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6942,6 +6951,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-sol": {
+        "cache_creation_input_token_cost": 6.875e-06,
         "cache_read_input_token_cost": 5.5e-07,
         "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
         "cache_read_input_token_cost_priority": 1.375e-06,
@@ -6990,6 +7000,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-terra": {
+        "cache_creation_input_token_cost": 2.75e-06,
         "cache_read_input_token_cost": 2.2e-07,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-07,
         "cache_read_input_token_cost_priority": 5.5e-07,
@@ -7038,6 +7049,7 @@
         "supports_minimal_reasoning_effort": false
     },
     "azure/eu/gpt-5.6-luna": {
+        "cache_creation_input_token_cost": 2.75e-07,
         "cache_read_input_token_cost": 2.2e-08,
         "cache_read_input_token_cost_above_272k_tokens": 4.4e-08,
         "cache_read_input_token_cost_priority": 5.5e-08,
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 06be96fefdf..86ca6eb4239 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
@@ -1057,19 +1057,24 @@ def test_generic_cost_per_token_gpt56_terra_cache_costs_by_tier_and_context(
 
 
 @pytest.mark.parametrize(
-    "model,input_cost,output_cost,cache_read_cost",
+    "model,input_cost,output_cost,cache_write_cost,cache_read_cost",
     [
-        ("azure/gpt-5.6", 5e-6, 3e-5, 5e-7),
-        ("azure/gpt-5.6-sol", 5e-6, 3e-5, 5e-7),
-        ("azure/gpt-5.6-terra", 2e-6, 1.2e-5, 2e-7),
-        ("azure/gpt-5.6-luna", 2e-7, 1.2e-6, 2e-8),
-        ("azure/us/gpt-5.6", 5.5e-6, 3.3e-5, 5.5e-7),
-        ("azure/eu/gpt-5.6-terra", 2.2e-6, 1.32e-5, 2.2e-7),
-        ("azure/eu/gpt-5.6-luna", 2.2e-7, 1.32e-6, 2.2e-8),
+        ("azure/gpt-5.6", 5e-6, 3e-5, 6.25e-6, 5e-7),
+        ("azure/gpt-5.6-sol", 5e-6, 3e-5, 6.25e-6, 5e-7),
+        ("azure/gpt-5.6-terra", 2e-6, 1.2e-5, 2.5e-6, 2e-7),
+        ("azure/gpt-5.6-luna", 2e-7, 1.2e-6, 2.5e-7, 2e-8),
+        ("azure/us/gpt-5.6", 5.5e-6, 3.3e-5, 6.875e-6, 5.5e-7),
+        ("azure/us/gpt-5.6-sol", 5.5e-6, 3.3e-5, 6.875e-6, 5.5e-7),
+        ("azure/us/gpt-5.6-terra", 2.2e-6, 1.32e-5, 2.75e-6, 2.2e-7),
+        ("azure/us/gpt-5.6-luna", 2.2e-7, 1.32e-6, 2.75e-7, 2.2e-8),
+        ("azure/eu/gpt-5.6", 5.5e-6, 3.3e-5, 6.875e-6, 5.5e-7),
+        ("azure/eu/gpt-5.6-sol", 5.5e-6, 3.3e-5, 6.875e-6, 5.5e-7),
+        ("azure/eu/gpt-5.6-terra", 2.2e-6, 1.32e-5, 2.75e-6, 2.2e-7),
+        ("azure/eu/gpt-5.6-luna", 2.2e-7, 1.32e-6, 2.75e-7, 2.2e-8),
     ],
 )
 def test_generic_cost_per_token_azure_gpt56(
-    model, input_cost, output_cost, cache_read_cost
+    model, input_cost, output_cost, cache_write_cost, cache_read_cost
 ):
     """Azure gpt-5.6 (global + us/eu regional): pricing mirrors the openai
     family for global deployments and carries the standard 10% regional uplift.
@@ -1081,22 +1086,31 @@ def test_generic_cost_per_token_azure_gpt56(
     assert model_cost_map["litellm_provider"] == "azure"
     assert model_cost_map["input_cost_per_token"] == input_cost
     assert model_cost_map["output_cost_per_token"] == output_cost
+    assert model_cost_map["cache_creation_input_token_cost"] == cache_write_cost
     assert model_cost_map["cache_read_input_token_cost"] == cache_read_cost
 
     prompt_tokens = 1000
     completion_tokens = 500
+    cache_write_tokens = 800
     usage = Usage(
         prompt_tokens=prompt_tokens,
         completion_tokens=completion_tokens,
         total_tokens=prompt_tokens + completion_tokens,
+        prompt_tokens_details=PromptTokensDetailsWrapper(
+            cache_write_tokens=cache_write_tokens
+        ),
     )
     prompt_cost, completion_cost = generic_cost_per_token(
         model=model,
         usage=usage,
         custom_llm_provider="azure",
     )
-    assert round(prompt_cost, 10) == round(input_cost * prompt_tokens, 10)
-    assert round(completion_cost, 10) == round(output_cost * completion_tokens, 10)
+    expected_prompt_cost = (
+        (prompt_tokens - cache_write_tokens) * input_cost
+        + cache_write_tokens * cache_write_cost
+    )
+    assert prompt_cost == pytest.approx(expected_prompt_cost)
+    assert completion_cost == pytest.approx(output_cost * completion_tokens)
 
 
 @pytest.mark.parametrize(