From 57faa623e3fd3c301f737a2ad79ebd6f8df112fb Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Tue, 25 Feb 2025 10:44:10 -0600 Subject: [PATCH 1/6] Adding Azure Phi-4 --- litellm/model_prices_and_context_window_backup.json | 13 +++++++++++++ model_prices_and_context_window.json | 13 +++++++++++++ 2 files changed, 26 insertions(+) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index a11930cc7f..442d5fa776 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -1732,6 +1732,19 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-405b-instruct-offer?tab=PlansAndPrice", "supports_tool_choice": true }, + "azure_ai/Phi-4": { + "max_tokens": 4096, + "max_input_tokens": 128000, + "max_output_tokens": 4096, + "input_cost_per_token": 0.000000125, + "output_cost_per_token": 0.0000005, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_vision": false, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495", + "supports_function_calling": true, + "supports_tool_choice": true + }, "azure_ai/Phi-3.5-mini-instruct": { "max_tokens": 4096, "max_input_tokens": 128000, diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index a11930cc7f..442d5fa776 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -1732,6 +1732,19 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-405b-instruct-offer?tab=PlansAndPrice", "supports_tool_choice": true }, + "azure_ai/Phi-4": { + "max_tokens": 4096, + "max_input_tokens": 128000, + "max_output_tokens": 4096, + "input_cost_per_token": 0.000000125, + "output_cost_per_token": 0.0000005, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_vision": false, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495", + "supports_function_calling": true, + "supports_tool_choice": true + }, "azure_ai/Phi-3.5-mini-instruct": { "max_tokens": 4096, "max_input_tokens": 128000, From 021f0c98f319ea599f641bedd59d12fc3a279087 Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Mon, 10 Mar 2025 20:44:01 -0500 Subject: [PATCH 2/6] Adding vertex_ai/imagen-3.0-generate-002 --- model_prices_and_context_window.json | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 442d5fa776..6389aa991b 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -4383,6 +4383,12 @@ "mode": "image_generation", "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" }, + "vertex_ai/imagen-3.0-generate-002": { + "output_cost_per_image": 0.04, + "litellm_provider": "vertex_ai-image-models", + "mode": "image_generation", + "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" + }, "vertex_ai/imagen-3.0-generate-001": { "output_cost_per_image": 0.04, "litellm_provider": "vertex_ai-image-models", From e38eab8bf6fc2ff0866b28fa51d8e422bb800e3e Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Mon, 10 Mar 2025 21:02:50 -0500 Subject: [PATCH 3/6] Adding/Updating Azure Phi-4 model variants Note: Azure has not yet published any cost for the newer variants (mini and multimodal) There is conflicting information about context size and model capabilities in different documents. --- model_prices_and_context_window.json | 33 +++++++++++++++++++++++----- 1 file changed, 27 insertions(+), 6 deletions(-) diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 6389aa991b..04db405dd3 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -1732,18 +1732,39 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-405b-instruct-offer?tab=PlansAndPrice", "supports_tool_choice": true }, - "azure_ai/Phi-4": { + "azure_ai/Phi-4-mini-instruct": { "max_tokens": 4096, - "max_input_tokens": 128000, + "max_input_tokens": 131072, "max_output_tokens": 4096, + "input_cost_per_token": 0, + "output_cost_per_token": 0, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_function_calling": true, + "source": "https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/models-featured#microsoft" + }, + "azure_ai/Phi-4-multimodal-instruct": { + "max_tokens": 4096, + "max_input_tokens": 131072, + "max_output_tokens": 4096, + "input_cost_per_token": 0, + "output_cost_per_token": 0, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_vision": true, + "source": "https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/models-featured#microsoft" + }, + "azure_ai/Phi-4": { + "max_tokens": 16384, + "max_input_tokens": 16384, + "max_output_tokens": 16384, "input_cost_per_token": 0.000000125, "output_cost_per_token": 0.0000005, "litellm_provider": "azure_ai", "mode": "chat", - "supports_vision": false, - "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495", - "supports_function_calling": true, - "supports_tool_choice": true + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495" }, "azure_ai/Phi-3.5-mini-instruct": { "max_tokens": 4096, From 2b6a41834fc9bd6f9e861695a38435afc0680af5 Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Mon, 10 Mar 2025 21:12:34 -0500 Subject: [PATCH 4/6] Add Azure DeepSeek V3 / Update DeepSeek R1 pricing --- model_prices_and_context_window.json | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 04db405dd3..be90ee3268 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -1587,13 +1587,23 @@ "max_tokens": 8192, "max_input_tokens": 128000, "max_output_tokens": 8192, - "input_cost_per_token": 0.0, - "input_cost_per_token_cache_hit": 0.0, - "output_cost_per_token": 0.0, + "input_cost_per_token": 0.00000135, + "output_cost_per_token": 0.0000054, "litellm_provider": "azure_ai", "mode": "chat", - "supports_prompt_caching": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/deepseek-r1-improved-performance-higher-limits-and-transparent-pricing/4386367" + }, + "azure_ai/deepseek-v3": { + "max_tokens": 8192, + "max_input_tokens": 128000, + "max_output_tokens": 8192, + "input_cost_per_token": 0.00000114, + "output_cost_per_token": 0.00000456, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_tool_choice": true, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/announcing-deepseek-v3-on-azure-ai-foundry-and-github/4390438" }, "azure_ai/jamba-instruct": { "max_tokens": 4096, From ebf20df1b9e951c55575d4ce4fe3e1143c118578 Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Mon, 10 Mar 2025 21:17:29 -0500 Subject: [PATCH 5/6] Add azure_ai/mistral-nemo --- model_prices_and_context_window.json | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index be90ee3268..9bc12953c5 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -1615,6 +1615,17 @@ "mode": "chat", "supports_tool_choice": true }, + "azure_ai/mistral-nemo": { + "max_tokens": 4096, + "max_input_tokens": 131072, + "max_output_tokens": 4096, + "input_cost_per_token": 0.00000015, + "output_cost_per_token": 0.00000015, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_function_calling": true, + "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-nemo-12b-2407?tab=PlansAndPrice" + }, "azure_ai/mistral-large": { "max_tokens": 8191, "max_input_tokens": 32000, From 3d85aa63a7ea0e8deafdaba1ab28bada8252a6f2 Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Mon, 10 Mar 2025 21:19:04 -0500 Subject: [PATCH 6/6] sync model db copy --- ...odel_prices_and_context_window_backup.json | 70 ++++++++++++++++--- 1 file changed, 59 insertions(+), 11 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 442d5fa776..9bc12953c5 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -1587,13 +1587,23 @@ "max_tokens": 8192, "max_input_tokens": 128000, "max_output_tokens": 8192, - "input_cost_per_token": 0.0, - "input_cost_per_token_cache_hit": 0.0, - "output_cost_per_token": 0.0, + "input_cost_per_token": 0.00000135, + "output_cost_per_token": 0.0000054, "litellm_provider": "azure_ai", "mode": "chat", - "supports_prompt_caching": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/deepseek-r1-improved-performance-higher-limits-and-transparent-pricing/4386367" + }, + "azure_ai/deepseek-v3": { + "max_tokens": 8192, + "max_input_tokens": 128000, + "max_output_tokens": 8192, + "input_cost_per_token": 0.00000114, + "output_cost_per_token": 0.00000456, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_tool_choice": true, + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/announcing-deepseek-v3-on-azure-ai-foundry-and-github/4390438" }, "azure_ai/jamba-instruct": { "max_tokens": 4096, @@ -1605,6 +1615,17 @@ "mode": "chat", "supports_tool_choice": true }, + "azure_ai/mistral-nemo": { + "max_tokens": 4096, + "max_input_tokens": 131072, + "max_output_tokens": 4096, + "input_cost_per_token": 0.00000015, + "output_cost_per_token": 0.00000015, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_function_calling": true, + "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-nemo-12b-2407?tab=PlansAndPrice" + }, "azure_ai/mistral-large": { "max_tokens": 8191, "max_input_tokens": 32000, @@ -1732,18 +1753,39 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-405b-instruct-offer?tab=PlansAndPrice", "supports_tool_choice": true }, - "azure_ai/Phi-4": { + "azure_ai/Phi-4-mini-instruct": { "max_tokens": 4096, - "max_input_tokens": 128000, + "max_input_tokens": 131072, "max_output_tokens": 4096, + "input_cost_per_token": 0, + "output_cost_per_token": 0, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_function_calling": true, + "source": "https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/models-featured#microsoft" + }, + "azure_ai/Phi-4-multimodal-instruct": { + "max_tokens": 4096, + "max_input_tokens": 131072, + "max_output_tokens": 4096, + "input_cost_per_token": 0, + "output_cost_per_token": 0, + "litellm_provider": "azure_ai", + "mode": "chat", + "supports_audio_input": true, + "supports_function_calling": true, + "supports_vision": true, + "source": "https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/models-featured#microsoft" + }, + "azure_ai/Phi-4": { + "max_tokens": 16384, + "max_input_tokens": 16384, + "max_output_tokens": 16384, "input_cost_per_token": 0.000000125, "output_cost_per_token": 0.0000005, "litellm_provider": "azure_ai", "mode": "chat", - "supports_vision": false, - "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495", - "supports_function_calling": true, - "supports_tool_choice": true + "source": "https://techcommunity.microsoft.com/blog/machinelearningblog/affordable-innovation-unveiling-the-pricing-of-phi-3-slms-on-models-as-a-service/4156495" }, "azure_ai/Phi-3.5-mini-instruct": { "max_tokens": 4096, @@ -4383,6 +4425,12 @@ "mode": "image_generation", "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" }, + "vertex_ai/imagen-3.0-generate-002": { + "output_cost_per_image": 0.04, + "litellm_provider": "vertex_ai-image-models", + "mode": "image_generation", + "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing" + }, "vertex_ai/imagen-3.0-generate-001": { "output_cost_per_image": 0.04, "litellm_provider": "vertex_ai-image-models",