From 53c67d302af253ca0306d8db6e3dcac406929dd5 Mon Sep 17 00:00:00 2001 From: ishaan-jaff Date: Mon, 11 Mar 2024 11:26:21 -0700 Subject: [PATCH 1/2] (fix) fix default dockerfile startup --- proxy_server_config.yaml | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/proxy_server_config.yaml b/proxy_server_config.yaml index 83bcc0626fc..a527d7ae90f 100644 --- a/proxy_server_config.yaml +++ b/proxy_server_config.yaml @@ -4,13 +4,13 @@ model_list: model: azure/chatgpt-v-2 api_base: https://openai-gpt-4-test-v-1.openai.azure.com/ api_version: "2023-05-15" - api_key: os.environ/AZURE_API_KEY # The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault + api_key: sk-defaultKey # use `os.environ/AZURE_API_KEY` for production. The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault - model_name: gpt-4 litellm_params: model: azure/chatgpt-v-2 api_base: https://openai-gpt-4-test-v-1.openai.azure.com/ api_version: "2023-05-15" - api_key: os.environ/AZURE_API_KEY # The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault + api_key: sk-defaultKey # use `os.environ/AZURE_API_KEY` for production. The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault - model_name: sagemaker-completion-model litellm_params: model: sagemaker/berri-benchmarking-Llama-2-70b-chat-hf-4 @@ -18,7 +18,7 @@ model_list: - model_name: text-embedding-ada-002 litellm_params: model: azure/azure-embedding-model - api_key: os.environ/AZURE_API_KEY + api_key: sk-defaultKey # use `os.environ/AZURE_API_KEY` for production. The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault api_base: https://openai-gpt-4-test-v-1.openai.azure.com/ api_version: "2023-05-15" model_info: @@ -26,13 +26,10 @@ model_list: base_model: text-embedding-ada-002 - model_name: dall-e-2 litellm_params: - model: azure/ + model: azure/dall-e-2 api_version: 2023-06-01-preview api_base: https://openai-gpt-4-test-v-1.openai.azure.com/ - api_key: os.environ/AZURE_API_KEY - - model_name: openai-dall-e-3 - litellm_params: - model: dall-e-3 + api_key: sk-defaultKey # use `os.environ/AZURE_API_KEY` for production. The `os.environ/` prefix tells litellm to read this from the env. See https://docs.litellm.ai/docs/simple_proxy#load-api-keys-from-vault litellm_settings: drop_params: True From ab48edca788f557556e04e9cec2cf7ba07a8e9e7 Mon Sep 17 00:00:00 2001 From: ishaan-jaff Date: Mon, 11 Mar 2024 11:39:06 -0700 Subject: [PATCH 2/2] (fix) default config users use on startup --- proxy_server_config.yaml | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/proxy_server_config.yaml b/proxy_server_config.yaml index a527d7ae90f..5f4875a7864 100644 --- a/proxy_server_config.yaml +++ b/proxy_server_config.yaml @@ -1,4 +1,6 @@ model_list: + # NOTE: This is the default config users use with Dockerfile. + # DO not expect users to pass os.environ/<> vars here, this will lead to proxy startup failing for them if they don't have the expected env vars - model_name: gpt-3.5-turbo litellm_params: model: azure/chatgpt-v-2 @@ -37,7 +39,7 @@ litellm_settings: budget_duration: 30d num_retries: 5 request_timeout: 600 -general_settings: +general_settings: master_key: sk-1234 # [OPTIONAL] Only use this if you to require all calls to contain this key (Authorization: Bearer sk-1234) proxy_budget_rescheduler_min_time: 60 proxy_budget_rescheduler_max_time: 64