diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index 931d4bdeabf..ddcfb8bb034 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -61618,13 +61618,16 @@
     },
     "tokenlab/gemini-3.1-pro-preview": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 1e-06,
+        "input_cost_per_token_above_200k_tokens": 2e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1048576,
         "max_output_tokens": 65536,
         "max_tokens": 65536,
         "mode": "chat",
         "output_cost_per_token": 6e-06,
+        "output_cost_per_token_above_200k_tokens": 9e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62332,13 +62335,16 @@
     },
     "tokenlab/gpt-5.4": {
         "cache_read_input_token_cost": 1.75e-07,
+        "cache_read_input_token_cost_above_272k_tokens": 3.5e-07,
         "input_cost_per_token": 1.75e-06,
+        "input_cost_per_token_above_272k_tokens": 3.5e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 400000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 1.05e-05,
+        "output_cost_per_token_above_272k_tokens": 1.575e-05,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62446,13 +62452,16 @@
     },
     "tokenlab/gpt-5.5": {
         "cache_read_input_token_cost": 1.5e-07,
+        "cache_read_input_token_cost_above_272k_tokens": 3e-07,
         "input_cost_per_token": 1.5e-06,
+        "input_cost_per_token_above_272k_tokens": 3e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 9e-06,
+        "output_cost_per_token_above_272k_tokens": 1.35e-05,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62476,12 +62485,14 @@
     },
     "tokenlab/gpt-5.5-pro": {
         "input_cost_per_token": 3e-05,
+        "input_cost_per_token_above_272k_tokens": 6e-05,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1050000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 0.00018,
+        "output_cost_per_token_above_272k_tokens": 0.00027,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62781,13 +62792,16 @@
     },
     "tokenlab/grok-4.20": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62811,13 +62825,16 @@
     },
     "tokenlab/grok-4.20-multi-agent": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62841,13 +62858,16 @@
     },
     "tokenlab/grok-4.20-non-reasoning": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62870,13 +62890,16 @@
     },
     "tokenlab/grok-4.3": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62998,13 +63021,16 @@
     },
     "tokenlab/grok-build-0.1": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 5e-07,
+        "input_cost_per_token_above_200k_tokens": 1e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 256000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1e-06,
+        "output_cost_per_token_above_200k_tokens": 2e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index 931d4bdeabf..ddcfb8bb034 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -61618,13 +61618,16 @@
     },
     "tokenlab/gemini-3.1-pro-preview": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 1e-06,
+        "input_cost_per_token_above_200k_tokens": 2e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1048576,
         "max_output_tokens": 65536,
         "max_tokens": 65536,
         "mode": "chat",
         "output_cost_per_token": 6e-06,
+        "output_cost_per_token_above_200k_tokens": 9e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62332,13 +62335,16 @@
     },
     "tokenlab/gpt-5.4": {
         "cache_read_input_token_cost": 1.75e-07,
+        "cache_read_input_token_cost_above_272k_tokens": 3.5e-07,
         "input_cost_per_token": 1.75e-06,
+        "input_cost_per_token_above_272k_tokens": 3.5e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 400000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 1.05e-05,
+        "output_cost_per_token_above_272k_tokens": 1.575e-05,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62446,13 +62452,16 @@
     },
     "tokenlab/gpt-5.5": {
         "cache_read_input_token_cost": 1.5e-07,
+        "cache_read_input_token_cost_above_272k_tokens": 3e-07,
         "input_cost_per_token": 1.5e-06,
+        "input_cost_per_token_above_272k_tokens": 3e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 9e-06,
+        "output_cost_per_token_above_272k_tokens": 1.35e-05,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62476,12 +62485,14 @@
     },
     "tokenlab/gpt-5.5-pro": {
         "input_cost_per_token": 3e-05,
+        "input_cost_per_token_above_272k_tokens": 6e-05,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1050000,
         "max_output_tokens": 128000,
         "max_tokens": 128000,
         "mode": "chat",
         "output_cost_per_token": 0.00018,
+        "output_cost_per_token_above_272k_tokens": 0.00027,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62781,13 +62792,16 @@
     },
     "tokenlab/grok-4.20": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62811,13 +62825,16 @@
     },
     "tokenlab/grok-4.20-multi-agent": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62841,13 +62858,16 @@
     },
     "tokenlab/grok-4.20-non-reasoning": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62870,13 +62890,16 @@
     },
     "tokenlab/grok-4.3": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 6.25e-07,
+        "input_cost_per_token_above_200k_tokens": 1.25e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 1000000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1.25e-06,
+        "output_cost_per_token_above_200k_tokens": 2.5e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
@@ -62998,13 +63021,16 @@
     },
     "tokenlab/grok-build-0.1": {
         "cache_read_input_token_cost": 1e-07,
+        "cache_read_input_token_cost_above_200k_tokens": 2e-07,
         "input_cost_per_token": 5e-07,
+        "input_cost_per_token_above_200k_tokens": 1e-06,
         "litellm_provider": "tokenlab",
         "max_input_tokens": 256000,
         "max_output_tokens": 131072,
         "max_tokens": 131072,
         "mode": "chat",
         "output_cost_per_token": 1e-06,
+        "output_cost_per_token_above_200k_tokens": 2e-06,
         "source": "https://api.tokenlab.sh/v1/models",
         "supported_endpoints": [
             "/v1/chat/completions",
diff --git a/tests/unit/llms/openai_like/test_tokenlab_provider.py b/tests/unit/llms/openai_like/test_tokenlab_provider.py
index 55348399306..2f5cb7bc3da 100644
--- a/tests/unit/llms/openai_like/test_tokenlab_provider.py
+++ b/tests/unit/llms/openai_like/test_tokenlab_provider.py
@@ -160,15 +160,33 @@ class TestTokenLabCostMap:
     def test_chat_cost_is_wired(self):
         prompt_cost, completion_cost = litellm.cost_per_token(
             model="tokenlab/gpt-5.5",
-            prompt_tokens=1_000_000,
-            completion_tokens=1_000_000,
+            prompt_tokens=100_000,
+            completion_tokens=100_000,
         )
-        assert prompt_cost == pytest.approx(litellm.model_cost["tokenlab/gpt-5.5"]["input_cost_per_token"] * 1_000_000)
+        assert prompt_cost == pytest.approx(litellm.model_cost["tokenlab/gpt-5.5"]["input_cost_per_token"] * 100_000)
         assert completion_cost == pytest.approx(
-            litellm.model_cost["tokenlab/gpt-5.5"]["output_cost_per_token"] * 1_000_000
+            litellm.model_cost["tokenlab/gpt-5.5"]["output_cost_per_token"] * 100_000
         )
         assert prompt_cost > 0 and completion_cost > 0
 
+    def test_chat_cost_uses_the_long_context_tier(self):
+        assert (
+            litellm.model_cost["tokenlab/gpt-5.5"]["input_cost_per_token_above_272k_tokens"]
+            > litellm.model_cost["tokenlab/gpt-5.5"]["input_cost_per_token"]
+        )
+
+        prompt_cost, completion_cost = litellm.cost_per_token(
+            model="tokenlab/gpt-5.5",
+            prompt_tokens=300_000,
+            completion_tokens=100_000,
+        )
+        assert prompt_cost == pytest.approx(
+            litellm.model_cost["tokenlab/gpt-5.5"]["input_cost_per_token_above_272k_tokens"] * 300_000
+        )
+        assert completion_cost == pytest.approx(
+            litellm.model_cost["tokenlab/gpt-5.5"]["output_cost_per_token_above_272k_tokens"] * 100_000
+        )
+
     def test_embedding_cost_is_wired(self):
         prompt_cost, completion_cost = litellm.cost_per_token(
             model="tokenlab/text-embedding-3-small",