diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 4229d80a671..adc4a557e63 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -554,6 +554,7 @@ "supports_vision": true }, "amazon.nova-sonic-v1:0": { + "deprecation_date": "2026-09-14", "input_cost_per_audio_token": 3.4e-06, "input_cost_per_token": 6e-08, "litellm_provider": "bedrock", @@ -3045,6 +3046,7 @@ "prompt_cache_min_tokens": 2048 }, "azure_ai/claude-fable-5": { + "deprecation_date": "2027-12-05", "supports_mid_conversation_system": true, "input_cost_per_token": 1e-05, "output_cost_per_token": 5e-05, @@ -3078,6 +3080,7 @@ "prompt_cache_min_tokens": 512 }, "azure_ai/claude-opus-5": { + "deprecation_date": "2027-07-08", "supports_mid_conversation_system": true, "supports_adaptive_thinking": true, "input_cost_per_token": 5e-06, @@ -3110,6 +3113,7 @@ "prompt_cache_min_tokens": 512 }, "azure_ai/claude-opus-4-8": { + "deprecation_date": "2027-09-01", "supports_mid_conversation_system": true, "supports_adaptive_thinking": true, "input_cost_per_token": 5e-06, @@ -3188,6 +3192,7 @@ "prompt_cache_min_tokens": 1024 }, "azure_ai/claude-sonnet-5": { + "deprecation_date": "2027-06-30", "supports_mid_conversation_system": true, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_1hr": 4e-06, @@ -15101,6 +15106,60 @@ "supports_tool_choice": true, "supports_vision": true }, + "databricks/databricks-deepseek-v4-flash-0731": { + "cache_read_input_token_cost": 2.8e-08, + "input_cost_per_token": 1.4e-07, + "input_dbu_cost_per_token": 2e-06, + "litellm_provider": "databricks", + "max_input_tokens": 1000000, + "max_output_tokens": 393216, + "max_tokens": 393216, + "metadata": { + "notes": "Input/output cost per token is dbu cost * $0.070. Billing reads the per-token dollar fields; the '*_dbu_cost_per_token' fields are the published Databricks rates, kept for reference. Context/max output are the DeepSeek-published model limits (1M context, 384K max output)." + }, + "mode": "chat", + "output_cost_per_token": 2.8e-07, + "output_dbu_cost_per_token": 4e-06, + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supported_modalities": [ + "text" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_tool_choice": true, + "supports_vision": false + }, + "databricks/databricks-deepseek-v4-pro-0813": { + "cache_read_input_token_cost": 1.3202e-07, + "input_cost_per_token": 1.31999e-06, + "input_dbu_cost_per_token": 1.8857e-05, + "litellm_provider": "databricks", + "max_input_tokens": 1000000, + "max_output_tokens": 393216, + "max_tokens": 393216, + "metadata": { + "notes": "Input/output cost per token is dbu cost * $0.070. Billing reads the per-token dollar fields; the '*_dbu_cost_per_token' fields are the published Databricks rates, kept for reference. Context/max output are the DeepSeek-published model limits (1M context, 384K max output)." + }, + "mode": "chat", + "output_cost_per_token": 3.95997e-06, + "output_dbu_cost_per_token": 5.6571e-05, + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supported_modalities": [ + "text" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_tool_choice": true, + "supports_vision": false + }, "databricks/databricks-gemini-2-5-flash": { "cache_creation_input_token_cost": 3.0002e-07, "cache_read_input_token_cost": 3.0002e-08, @@ -20631,6 +20690,7 @@ "supports_image_size": false }, "gemini-live-2.5-flash-native-audio": { + "deprecation_date": "2026-12-13", "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "vertex_ai-language-models", @@ -52237,14 +52297,14 @@ "supports_vision": true }, "fireworks_ai/deepseek-v4-flash-0731": { - "cache_read_input_token_cost": 2.8e-08, - "input_cost_per_token": 1.4e-07, + "cache_read_input_token_cost": 7e-09, + "input_cost_per_token": 2.2e-07, "litellm_provider": "fireworks_ai", "max_input_tokens": 1048576, "max_output_tokens": 131072, "max_tokens": 131072, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 6.6e-07, "source": "https://docs.fireworks.ai/serverless/pricing", "supports_function_calling": true, "supports_reasoning": true, diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 4229d80a671..adc4a557e63 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -554,6 +554,7 @@ "supports_vision": true }, "amazon.nova-sonic-v1:0": { + "deprecation_date": "2026-09-14", "input_cost_per_audio_token": 3.4e-06, "input_cost_per_token": 6e-08, "litellm_provider": "bedrock", @@ -3045,6 +3046,7 @@ "prompt_cache_min_tokens": 2048 }, "azure_ai/claude-fable-5": { + "deprecation_date": "2027-12-05", "supports_mid_conversation_system": true, "input_cost_per_token": 1e-05, "output_cost_per_token": 5e-05, @@ -3078,6 +3080,7 @@ "prompt_cache_min_tokens": 512 }, "azure_ai/claude-opus-5": { + "deprecation_date": "2027-07-08", "supports_mid_conversation_system": true, "supports_adaptive_thinking": true, "input_cost_per_token": 5e-06, @@ -3110,6 +3113,7 @@ "prompt_cache_min_tokens": 512 }, "azure_ai/claude-opus-4-8": { + "deprecation_date": "2027-09-01", "supports_mid_conversation_system": true, "supports_adaptive_thinking": true, "input_cost_per_token": 5e-06, @@ -3188,6 +3192,7 @@ "prompt_cache_min_tokens": 1024 }, "azure_ai/claude-sonnet-5": { + "deprecation_date": "2027-06-30", "supports_mid_conversation_system": true, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_1hr": 4e-06, @@ -15101,6 +15106,60 @@ "supports_tool_choice": true, "supports_vision": true }, + "databricks/databricks-deepseek-v4-flash-0731": { + "cache_read_input_token_cost": 2.8e-08, + "input_cost_per_token": 1.4e-07, + "input_dbu_cost_per_token": 2e-06, + "litellm_provider": "databricks", + "max_input_tokens": 1000000, + "max_output_tokens": 393216, + "max_tokens": 393216, + "metadata": { + "notes": "Input/output cost per token is dbu cost * $0.070. Billing reads the per-token dollar fields; the '*_dbu_cost_per_token' fields are the published Databricks rates, kept for reference. Context/max output are the DeepSeek-published model limits (1M context, 384K max output)." + }, + "mode": "chat", + "output_cost_per_token": 2.8e-07, + "output_dbu_cost_per_token": 4e-06, + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supported_modalities": [ + "text" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_tool_choice": true, + "supports_vision": false + }, + "databricks/databricks-deepseek-v4-pro-0813": { + "cache_read_input_token_cost": 1.3202e-07, + "input_cost_per_token": 1.31999e-06, + "input_dbu_cost_per_token": 1.8857e-05, + "litellm_provider": "databricks", + "max_input_tokens": 1000000, + "max_output_tokens": 393216, + "max_tokens": 393216, + "metadata": { + "notes": "Input/output cost per token is dbu cost * $0.070. Billing reads the per-token dollar fields; the '*_dbu_cost_per_token' fields are the published Databricks rates, kept for reference. Context/max output are the DeepSeek-published model limits (1M context, 384K max output)." + }, + "mode": "chat", + "output_cost_per_token": 3.95997e-06, + "output_dbu_cost_per_token": 5.6571e-05, + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supported_modalities": [ + "text" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_tool_choice": true, + "supports_vision": false + }, "databricks/databricks-gemini-2-5-flash": { "cache_creation_input_token_cost": 3.0002e-07, "cache_read_input_token_cost": 3.0002e-08, @@ -20631,6 +20690,7 @@ "supports_image_size": false }, "gemini-live-2.5-flash-native-audio": { + "deprecation_date": "2026-12-13", "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "vertex_ai-language-models", @@ -52237,14 +52297,14 @@ "supports_vision": true }, "fireworks_ai/deepseek-v4-flash-0731": { - "cache_read_input_token_cost": 2.8e-08, - "input_cost_per_token": 1.4e-07, + "cache_read_input_token_cost": 7e-09, + "input_cost_per_token": 2.2e-07, "litellm_provider": "fireworks_ai", "max_input_tokens": 1048576, "max_output_tokens": 131072, "max_tokens": 131072, "mode": "chat", - "output_cost_per_token": 2.8e-07, + "output_cost_per_token": 6.6e-07, "source": "https://docs.fireworks.ai/serverless/pricing", "supports_function_calling": true, "supports_reasoning": true, diff --git a/tests/test_litellm/test_fireworks_serverless_model_costs.py b/tests/test_litellm/test_fireworks_serverless_model_costs.py index 0458af0da0e..a7a9e0fc37d 100644 --- a/tests/test_litellm/test_fireworks_serverless_model_costs.py +++ b/tests/test_litellm/test_fireworks_serverless_model_costs.py @@ -84,3 +84,41 @@ def test_bare_fireworks_ids_resolve_through_prefixed_entries(): assert info["output_cost_per_token"] == pytest.approx(expected["output_cost_per_token"]) assert info["max_input_tokens"] == expected["max_input_tokens"] assert info["max_output_tokens"] == expected["max_output_tokens"] + + +TWIN_PINNED_PRICES = { + "deepseek-v4-flash-0731": { + "input_cost_per_token": 2.2e-07, + "cache_read_input_token_cost": 7e-09, + "output_cost_per_token": 6.6e-07, + }, +} + + +def test_deepseek_v4_flash_0731_twins_pin_published_pricing(model_data): + """Both 0731 entries carry the price published at docs.fireworks.ai/serverless/pricing.""" + for bare_suffix, expected in TWIN_PINNED_PRICES.items(): + for key in ( + f"fireworks_ai/{bare_suffix}", + f"fireworks_ai/accounts/fireworks/models/{bare_suffix}", + ): + entry = model_data[key] + for field, value in expected.items(): + assert entry[field] == pytest.approx(value), f"{key}.{field}" + + +def test_fireworks_account_prefixed_twins_agree_on_price(model_data): + """Every accounts/fireworks/models/X entry prices identically to its bare fireworks_ai/X twin.""" + prefix = "fireworks_ai/accounts/fireworks/models/" + pairs_checked = 0 + for key, entry in model_data.items(): + if not key.startswith(prefix): + continue + bare_key = f"fireworks_ai/{key[len(prefix):]}" + bare_entry = model_data.get(bare_key) + if bare_entry is None: + continue + pairs_checked += 1 + for field in sorted({f for f in (*entry, *bare_entry) if "cost" in f}): + assert entry.get(field) == bare_entry.get(field), f"{key} vs {bare_key}: {field}" + assert pairs_checked >= 20