From 6bb677d30f672c1e498b8eacbbf28d1ab42d3f6e Mon Sep 17 00:00:00 2001 From: mateo Date: Thu, 20 Aug 2026 21:44:32 +0000 Subject: [PATCH] fix(model-costs): correct gpt-5.6 input token cap Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- ...odel_prices_and_context_window_backup.json | 32 +++++++++---------- model_prices_and_context_window.json | 32 +++++++++---------- .../llm_cost_calc/test_llm_cost_calc_utils.py | 3 +- 3 files changed, 34 insertions(+), 33 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index b9c8824aa67..78869e05dfe 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -6510,7 +6510,7 @@ "input_cost_per_token_priority": 1e-05, "input_cost_per_token_above_272k_tokens_priority": 2e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6561,7 +6561,7 @@ "input_cost_per_token_priority": 1e-05, "input_cost_per_token_above_272k_tokens_priority": 2e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6612,7 +6612,7 @@ "input_cost_per_token_priority": 4e-06, "input_cost_per_token_above_272k_tokens_priority": 8e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6663,7 +6663,7 @@ "input_cost_per_token_priority": 4e-07, "input_cost_per_token_above_272k_tokens_priority": 8e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6711,7 +6711,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6759,7 +6759,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6807,7 +6807,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-06, "input_cost_per_token_priority": 5.5e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6855,7 +6855,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-07, "input_cost_per_token_priority": 5.5e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6902,7 +6902,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6950,7 +6950,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6998,7 +6998,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-06, "input_cost_per_token_priority": 5.5e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -7046,7 +7046,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-07, "input_cost_per_token_priority": 5.5e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25333,7 +25333,7 @@ "input_cost_per_token_flex": 2.5e-06, "input_cost_per_token_priority": 1e-05, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25396,7 +25396,7 @@ "input_cost_per_token_flex": 2.5e-06, "input_cost_per_token_priority": 1e-05, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25459,7 +25459,7 @@ "input_cost_per_token_flex": 1e-06, "input_cost_per_token_priority": 4e-06, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25522,7 +25522,7 @@ "input_cost_per_token_flex": 1e-07, "input_cost_per_token_priority": 4e-07, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index b9c8824aa67..78869e05dfe 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -6510,7 +6510,7 @@ "input_cost_per_token_priority": 1e-05, "input_cost_per_token_above_272k_tokens_priority": 2e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6561,7 +6561,7 @@ "input_cost_per_token_priority": 1e-05, "input_cost_per_token_above_272k_tokens_priority": 2e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6612,7 +6612,7 @@ "input_cost_per_token_priority": 4e-06, "input_cost_per_token_above_272k_tokens_priority": 8e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6663,7 +6663,7 @@ "input_cost_per_token_priority": 4e-07, "input_cost_per_token_above_272k_tokens_priority": 8e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6711,7 +6711,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6759,7 +6759,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6807,7 +6807,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-06, "input_cost_per_token_priority": 5.5e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6855,7 +6855,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-07, "input_cost_per_token_priority": 5.5e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6902,7 +6902,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6950,7 +6950,7 @@ "input_cost_per_token_above_272k_tokens": 1.1e-05, "input_cost_per_token_priority": 1.375e-05, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -6998,7 +6998,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-06, "input_cost_per_token_priority": 5.5e-06, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -7046,7 +7046,7 @@ "input_cost_per_token_above_272k_tokens": 4.4e-07, "input_cost_per_token_priority": 5.5e-07, "litellm_provider": "azure", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25333,7 +25333,7 @@ "input_cost_per_token_flex": 2.5e-06, "input_cost_per_token_priority": 1e-05, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25396,7 +25396,7 @@ "input_cost_per_token_flex": 2.5e-06, "input_cost_per_token_priority": 1e-05, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25459,7 +25459,7 @@ "input_cost_per_token_flex": 1e-06, "input_cost_per_token_priority": 4e-06, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", @@ -25522,7 +25522,7 @@ "input_cost_per_token_flex": 1e-07, "input_cost_per_token_priority": 4e-07, "litellm_provider": "openai", - "max_input_tokens": 1050000, + "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", 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..afe8e1d37a2 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 @@ -941,7 +941,7 @@ def test_generic_cost_per_token_gpt56( assert model_cost_map["cache_creation_input_token_cost"] == pytest.approx( input_cost * 1.25 ) - assert model_cost_map["max_input_tokens"] == 1050000 + assert model_cost_map["max_input_tokens"] == 922000 assert model_cost_map["input_cost_per_token_above_272k_tokens"] == pytest.approx( input_cost * 2 ) @@ -1082,6 +1082,7 @@ def test_generic_cost_per_token_azure_gpt56( 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_read_input_token_cost"] == cache_read_cost + assert model_cost_map["max_input_tokens"] == 922000 prompt_tokens = 1000 completion_tokens = 500