From c90942e5508cbd4dbdc308debf979860fbed74e2 Mon Sep 17 00:00:00 2001 From: Emerson Gomes Date: Tue, 17 Dec 2024 19:56:38 -0600 Subject: [PATCH] Correct max_tokens on Model DB (#7284) Corrects several instances of `max_tokens` and `max_output_tokens` for Azure AI models --- model_prices_and_context_window.json | 41 ++++++++++++++-------------- 1 file changed, 20 insertions(+), 21 deletions(-) diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 6275c9c4aa..a607bfbc6f 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -1308,7 +1308,7 @@ "mode": "chat" }, "azure_ai/mistral-large-2407": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.000002, @@ -1319,7 +1319,7 @@ "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.mistral-ai-large-2407-offer?tab=Overview" }, "azure_ai/ministral-3b": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000004, @@ -1330,7 +1330,7 @@ "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/000-000.ministral-3b-2410-offer?tab=Overview" }, "azure_ai/Llama-3.2-11B-Vision-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, "max_output_tokens": 2048, "input_cost_per_token": 0.00000037, @@ -1342,19 +1342,18 @@ "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/metagenai.meta-llama-3-2-11b-vision-instruct-offer?tab=Overview" }, "azure_ai/Llama-3.3-70B-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, "max_output_tokens": 2048, "input_cost_per_token": 0.00000071, "output_cost_per_token": 0.00000071, "litellm_provider": "azure_ai", "supports_function_calling": true, - "supports_vision": true, "mode": "chat", "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/metagenai.llama-3-3-70b-instruct-offer?tab=Overview" }, "azure_ai/Llama-3.2-90B-Vision-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, "max_output_tokens": 2048, "input_cost_per_token": 0.00000204, @@ -1366,18 +1365,18 @@ "source": "https://azuremarketplace.microsoft.com/en/marketplace/apps/metagenai.meta-llama-3-2-90b-vision-instruct-offer?tab=Overview" }, "azure_ai/Meta-Llama-3-70B-Instruct": { - "max_tokens": 8192, + "max_tokens": 2048, "max_input_tokens": 8192, - "max_output_tokens": 8192, + "max_output_tokens": 2048, "input_cost_per_token": 0.0000011, "output_cost_per_token": 0.00000037, "litellm_provider": "azure_ai", "mode": "chat" }, "azure_ai/Meta-Llama-3.1-8B-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, - "max_output_tokens": 128000, + "max_output_tokens": 2048, "input_cost_per_token": 0.0000003, "output_cost_per_token": 0.00000061, "litellm_provider": "azure_ai", @@ -1385,9 +1384,9 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-8b-instruct-offer?tab=PlansAndPrice" }, "azure_ai/Meta-Llama-3.1-70B-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, - "max_output_tokens": 128000, + "max_output_tokens": 2048, "input_cost_per_token": 0.00000268, "output_cost_per_token": 0.00000354, "litellm_provider": "azure_ai", @@ -1395,9 +1394,9 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-70b-instruct-offer?tab=PlansAndPrice" }, "azure_ai/Meta-Llama-3.1-405B-Instruct": { - "max_tokens": 128000, + "max_tokens": 2048, "max_input_tokens": 128000, - "max_output_tokens": 128000, + "max_output_tokens": 2048, "input_cost_per_token": 0.00000533, "output_cost_per_token": 0.000016, "litellm_provider": "azure_ai", @@ -1405,7 +1404,7 @@ "source":"https://azuremarketplace.microsoft.com/en-us/marketplace/apps/metagenai.meta-llama-3-1-405b-instruct-offer?tab=PlansAndPrice" }, "azure_ai/Phi-3.5-mini-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000013, @@ -1416,7 +1415,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3.5-vision-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000013, @@ -1427,7 +1426,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3.5-MoE-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000016, @@ -1449,7 +1448,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3-mini-128k-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000013, @@ -1460,7 +1459,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3-small-8k-instruct": { - "max_tokens": 8192, + "max_tokens": 4096, "max_input_tokens": 8192, "max_output_tokens": 4096, "input_cost_per_token": 0.00000015, @@ -1471,7 +1470,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3-small-128k-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000015, @@ -1493,7 +1492,7 @@ "source": "https://azure.microsoft.com/en-us/pricing/details/phi-3/" }, "azure_ai/Phi-3-medium-128k-instruct": { - "max_tokens": 128000, + "max_tokens": 4096, "max_input_tokens": 128000, "max_output_tokens": 4096, "input_cost_per_token": 0.00000017,