fix: update base image uri (#189)

Author: yanbasic

src/emd/cfn/shared/ecs_cluster.yaml

@@ -10,7 +10,7 @@ Parameters:
   APIRouterImageURI:
     Type: String
     Description: The URI of OpenAI compatbile API router image. If provided, the router will be deployed.
-    Default: "public.ecr.aws/aws-gcr-solutions/dmaa/api-router:latest"
+    Default: "quay.io/dmaa/api-router:latest"
   UseSpot:
     Type: String
     Description: Use Fargate Spot capacity?

src/pipeline/backend/huggingface/embedding/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
+FROM quay.io/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
 
 # Create a new stage based on vllm-base
 FROM huggingface-base AS sagemaker-serving

src/pipeline/backend/huggingface/llm/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
+FROM quay.io/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
 
 # Create a new stage based on vllm-base
 FROM huggingface-base AS sagemaker-serving

src/pipeline/backend/huggingface/llm/Dockerfile_baichuan

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
+FROM quay.io/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
 
 # Create a new stage based on vllm-base
 FROM huggingface-base AS sagemaker-serving

src/pipeline/backend/huggingface/rerank/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
+FROM quay.io/dmaa/huggingface/transformers-pytorch-gpu:{{VERSION}} AS huggingface-base
 
 # Create a new stage based on vllm-base
 FROM huggingface-base AS sagemaker-serving

src/pipeline/backend/llama_cpp/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/llama-cpp:{{VERSION}}
+FROM quay.io/dmaa/llama-cpp:{{VERSION}}
 
 # Ensure the serve script has executable permissions
 # RUN chmod +x /usr/bin/serve

src/pipeline/backend/lmdeploy/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/openmmlab/lmdeploy:{{VERSION}} AS lmdeploy-base
+FROM quay.io/dmaa/openmmlab/lmdeploy:{{VERSION}} AS lmdeploy-base
 
 # Create a new stage based on vllm-base
 FROM lmdeploy-base AS sagemaker-serving

src/pipeline/backend/lmdeploy/Dockerfile_internvl2

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/openmmlab/lmdeploy:{{VERSION}} AS lmdeploy-base
+FROM quay.io/dmaa/openmmlab/lmdeploy:{{VERSION}} AS lmdeploy-base
 
 # Create a new stage based on vllm-base
 FROM lmdeploy-base AS sagemaker-serving

src/pipeline/backend/ollama/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/ollama:{{VERSION}} AS ollama-base
+FROM quay.io/dmaa/ollama:{{VERSION}} AS ollama-base
 
 # Ensure the serve script has executable permissions
 # RUN chmod +x /usr/bin/serve

src/pipeline/backend/tgi/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/aws-gcr-solutions/dmaa/huggingface/text-generation-inference:{{VERSION}} AS tgi-base
+FROM quay.io/dmaa/huggingface/text-generation-inference:{{VERSION}} AS tgi-base
 
 # Create a new stage based on vllm-base
 FROM tgi-base AS sagemaker-serving