Fix vllm model cache directory (#1642)

Signed-off-by: Wang, Kai Lawrence <kai.lawrence.wang@intel.com>

ChatQnA/docker_compose/intel/cpu/xeon/README.md

@@ -219,7 +219,7 @@ For users in China who are unable to download models directly from Huggingface,
    export HF_ENDPOINT="https://hf-mirror.com"
    model_name="meta-llama/Meta-Llama-3-8B-Instruct"
    # Start vLLM LLM Service
-   docker run -p 8008:80 -v ./data:/data --name vllm-service -e HF_ENDPOINT=$HF_ENDPOINT -e http_proxy=$http_proxy -e https_proxy=$https_proxy --shm-size 128g opea/vllm:latest --model $model_name --host 0.0.0.0 --port 80
+   docker run -p 8008:80 -v ./data:/root/.cache/huggingface/hub --name vllm-service -e HF_ENDPOINT=$HF_ENDPOINT -e http_proxy=$http_proxy -e https_proxy=$https_proxy --shm-size 128g opea/vllm:latest --model $model_name --host 0.0.0.0 --port 80
    # Start TGI LLM Service
    docker run -p 8008:80 -v ./data:/data --name tgi-service -e HF_ENDPOINT=$HF_ENDPOINT -e http_proxy=$http_proxy -e https_proxy=$https_proxy --shm-size 1g ghcr.io/huggingface/text-generation-inference:2.4.0-intel-cpu --model-id $model_name
    ```
@@ -236,7 +236,7 @@ For users in China who are unable to download models directly from Huggingface,
      export HF_TOKEN=${your_hf_token}
      export model_path="/path/to/model"
      # Start vLLM LLM Service
-     docker run -p 8008:80 -v $model_path:/data --name vllm-service --shm-size 128g opea/vllm:latest --model /data --host 0.0.0.0 --port 80
+     docker run -p 8008:80 -v $model_path:/root/.cache/huggingface/hub --name vllm-service --shm-size 128g opea/vllm:latest --model /root/.cache/huggingface/hub --host 0.0.0.0 --port 80
      # Start TGI LLM Service
      docker run -p 8008:80 -v $model_path:/data --name tgi-service --shm-size 1g ghcr.io/huggingface/text-generation-inference:2.4.0-intel-cpu --model-id /data
      ```

ChatQnA/docker_compose/intel/cpu/xeon/README_pinecone.md

@@ -201,7 +201,7 @@ For users in China who are unable to download models directly from Huggingface,
    export HF_TOKEN=${your_hf_token}
    export HF_ENDPOINT="https://hf-mirror.com"
    model_name="meta-llama/Meta-Llama-3-8B-Instruct"
-   docker run -p 8008:80 -v ./data:/data --name vllm-service -e HF_ENDPOINT=$HF_ENDPOINT -e http_proxy=$http_proxy -e https_proxy=$https_proxy --shm-size 128g opea/vllm:latest --model $model_name --host 0.0.0.0 --port 80
+   docker run -p 8008:80 -v ./data:/root/.cache/huggingface/hub --name vllm-service -e HF_ENDPOINT=$HF_ENDPOINT -e http_proxy=$http_proxy -e https_proxy=$https_proxy --shm-size 128g opea/vllm:latest --model $model_name --host 0.0.0.0 --port 80
    ```
 
 2. Offline
@@ -215,7 +215,7 @@ For users in China who are unable to download models directly from Huggingface,
      ```bash
      export HF_TOKEN=${your_hf_token}
      export model_path="/path/to/model"
-     docker run -p 8008:80 -v $model_path:/data --name vllm-service --shm-size 128g opea/vllm:latest --model /data --host 0.0.0.0 --port 80
+     docker run -p 8008:80 -v $model_path:/root/.cache/huggingface/hub --name vllm-service --shm-size 128g opea/vllm:latest --model /root/.cache/huggingface/hub --host 0.0.0.0 --port 80
      ```
 
 ### Setup Environment Variables

ChatQnA/docker_compose/intel/cpu/xeon/compose.yaml

@@ -80,7 +80,7 @@ services:
     ports:
       - "9009:80"
     volumes:
-      - "${MODEL_CACHE:-./data}:/data"
+      - "${MODEL_CACHE:-./data}:/root/.cache/huggingface/hub"
     shm_size: 128g
     environment:
       no_proxy: ${no_proxy}

ChatQnA/docker_compose/intel/cpu/xeon/compose_milvus.yaml

@@ -144,7 +144,7 @@ services:
     ports:
       - "9009:80"
     volumes:
-      - "./data:/data"
+      - "${MODEL_CACHE:-./data}:/root/.cache/huggingface/hub"
     shm_size: 128g
     environment:
       no_proxy: ${no_proxy}

ChatQnA/docker_compose/intel/cpu/xeon/compose_qdrant.yaml

@@ -80,7 +80,7 @@ services:
     ports:
       - "6042:80"
     volumes:
-      - "${MODEL_CACHE:-./data}:/data"
+      - "${MODEL_CACHE:-./data}:/root/.cache/huggingface/hub"
     shm_size: 128g
     environment:
       no_proxy: ${no_proxy}

ChatQnA/docker_compose/intel/cpu/xeon/compose_without_rerank.yaml

@@ -64,7 +64,7 @@ services:
     ports:
       - "9009:80"
     volumes:
-      - "${MODEL_CACHE:-./data}:/data"
+      - "${MODEL_CACHE:-./data}:/root/.cache/huggingface/hub"
     shm_size: 128g
     environment:
       no_proxy: ${no_proxy}