Merge branch 'main' into zepan/new_check

ChatQnA/benchmark/performance/kubernetes/intel/gaudi/README.md

@@ -69,10 +69,6 @@ Results will be displayed in the terminal and saved as CSV file named `1_stats.c
   - Persistent Volume Claim (PVC): This is the recommended approach for production setups. For more details on using PVC, refer to [PVC](https://github.com/opea-project/GenAIInfra/blob/main/helm-charts/README.md#using-persistent-volume).
   - Local Host Path: For simpler testing, ensure that each node involved in the deployment follows the steps above to locally prepare the models. After preparing the models, use `--set global.modelUseHostPath=${MODELDIR}` in the deployment command.
 
-- Add OPEA Helm Repository:
-  ```bash
-  python deploy.py --add-repo
-  ```
 - Label Nodes
   ```base
   python deploy.py --add-label --num-nodes 2
@@ -192,13 +188,9 @@ All the test results will come to the folder `GenAIEval/evals/benchmark/benchmar
 
 ## Teardown
 
-After completing the benchmark, use the following commands to clean up the environment:
+After completing the benchmark, use the following command to clean up the environment:
 
 Remove Node Labels:
-```base
-python deploy.py --delete-label
-```
-Delete the OPEA Helm Repository:
 ```bash
-python deploy.py --delete-repo
+python deploy.py --delete-label
 ```

ChatQnA/benchmark/performance/kubernetes/intel/gaudi/deploy.py

@@ -83,26 +83,6 @@ def clear_labels_from_nodes(label, node_names=None):
             print(f"Label {label_key} not found on node {node_name}, skipping.")
 
 
-def add_helm_repo(repo_name, repo_url):
-    # Add the repo if it does not exist
-    add_command = ["helm", "repo", "add", repo_name, repo_url]
-    try:
-        subprocess.run(add_command, check=True)
-        print(f"Added Helm repo {repo_name} from {repo_url}.")
-    except subprocess.CalledProcessError as e:
-        print(f"Failed to add Helm repo {repo_name}: {e}")
-
-
-def delete_helm_repo(repo_name):
-    """Delete Helm repo if it exists."""
-    command = ["helm", "repo", "remove", repo_name]
-    try:
-        subprocess.run(command, check=True)
-        print(f"Deleted Helm repo {repo_name}.")
-    except subprocess.CalledProcessError:
-        print(f"Failed to delete Helm repo {repo_name}. It may not exist.")
-
-
 def install_helm_release(release_name, chart_name, namespace, values_file, device_type):
     """Deploy a Helm release with a specified name and chart.
 
@@ -132,14 +112,14 @@ def install_helm_release(release_name, chart_name, namespace, values_file, devic
     if device_type == "gaudi":
         print("Device type is gaudi. Pulling Helm chart to get gaudi-values.yaml...")
 
-        # Pull and untar the chart
-        subprocess.run(["helm", "pull", chart_name, "--untar"], check=True)
+        # Combine chart_name with fixed prefix
+        chart_pull_url = f"oci://ghcr.io/opea-project/charts/{chart_name}"
 
-        # Determine the directory name (get the actual chart_name if chart_name is in the format 'repo_name/chart_name', else use chart_name directly)
-        chart_dir_name = chart_name.split("/")[-1] if "/" in chart_name else chart_name
+        # Pull and untar the chart
+        subprocess.run(["helm", "pull", chart_pull_url, "--untar"], check=True)
 
-        # Find the untarred directory (assumes only one directory matches chart_dir_name)
-        untar_dirs = glob.glob(f"{chart_dir_name}*")
+        # Find the untarred directory
+        untar_dirs = glob.glob(f"{chart_name}*")
         if untar_dirs:
             untar_dir = untar_dirs[0]
             hw_values_file = os.path.join(untar_dir, "gaudi-values.yaml")
@@ -210,20 +190,14 @@ def main():
     parser.add_argument(
         "--chart-name",
         type=str,
-        default="opea/chatqna",
-        help="The chart name to deploy, composed of repo name and chart name (default: opea/chatqna).",
+        default="chatqna",
+        help="The chart name to deploy, composed of repo name and chart name (default: chatqna).",
     )
     parser.add_argument("--namespace", default="default", help="Kubernetes namespace (default: default).")
     parser.add_argument("--hf-token", help="Hugging Face API token.")
     parser.add_argument(
         "--model-dir", help="Model directory, mounted as volumes for service access to pre-downloaded models"
     )
-    parser.add_argument("--repo-name", default="opea", help="Helm repo name to add/delete (default: opea).")
-    parser.add_argument(
-        "--repo-url",
-        default="https://opea-project.github.io/GenAIInfra",
-        help="Helm repository URL (default: https://opea-project.github.io/GenAIInfra).",
-    )
     parser.add_argument("--user-values", help="Path to a user-specified values.yaml file.")
     parser.add_argument(
         "--create-values-only", action="store_true", help="Only create the values.yaml file without deploying."
@@ -244,8 +218,6 @@ def main():
         action="store_true",
         help="Modify resources for services and change extraCmdArgs when creating values.yaml.",
     )
-    parser.add_argument("--add-repo", action="store_true", help="Add the Helm repo specified by --repo-url.")
-    parser.add_argument("--delete-repo", action="store_true", help="Delete the Helm repo specified by --repo-name.")
     parser.add_argument(
         "--device-type",
         type=str,
@@ -264,14 +236,6 @@ def main():
         else:
             args.num_nodes = num_node_names
 
-    # Helm repository management
-    if args.add_repo:
-        add_helm_repo(args.repo_name, args.repo_url)
-        return
-    elif args.delete_repo:
-        delete_helm_repo(args.repo_name)
-        return
-
     # Node labeling management
     if args.add_label:
         add_labels_to_nodes(args.num_nodes, args.label, args.node_names)

ChatQnA/docker_compose/amd/gpu/rocm/README.md

@@ -290,7 +290,7 @@ docker compose up -d
    Try the command below to check whether the TGI service is ready.
 
    ```bash
-   docker logs ${CONTAINER_ID} | grep Connected
+   docker logs chatqna-tgi-server | grep Connected
    ```
 
    If the service is ready, you will get the response like below.

ChatQnA/docker_compose/intel/hpu/gaudi/README.md

@@ -314,7 +314,7 @@ For validation details, please refer to [how-to-validate_service](./how_to_valid
    Try the command below to check whether the LLM serving is ready.
 
    ```bash
-   docker logs tgi-service | grep Connected
+   docker logs tgi-gaudi-server | grep Connected
    ```
 
    If the service is ready, you will get the response like below.
@@ -327,15 +327,15 @@ For validation details, please refer to [how-to-validate_service](./how_to_valid
 
    ```bash
    # TGI service
-   curl http://${host_ip}:9009/v1/chat/completions \
+   curl http://${host_ip}:8005/v1/chat/completions \
      -X POST \
      -d '{"model": ${LLM_MODEL_ID}, "messages": [{"role": "user", "content": "What is Deep Learning?"}], "max_tokens":17}' \
      -H 'Content-Type: application/json'
    ```
 
    ```bash
    # vLLM Service
-   curl http://${host_ip}:9009/v1/chat/completions \
+   curl http://${host_ip}:8007/v1/chat/completions \
      -H "Content-Type: application/json" \
      -d '{"model": ${LLM_MODEL_ID}, "messages": [{"role": "user", "content": "What is Deep Learning?"}]}'
    ```

ChatQnA/docker_compose/nvidia/gpu/README.md

@@ -273,7 +273,7 @@ docker compose up -d
    Try the command below to check whether the TGI service is ready.
 
    ```bash
-   docker logs ${CONTAINER_ID} | grep Connected
+   docker logs tgi-server | grep Connected
    ```
 
    If the service is ready, you will get the response like below.
@@ -285,7 +285,7 @@ docker compose up -d
    Then try the `cURL` command below to validate TGI.
 
    ```bash
-   curl http://${host_ip}:9009/v1/chat/completions \
+   curl http://${host_ip}:8008/v1/chat/completions \
      -X POST \
      -d '{"model": "Intel/neural-chat-7b-v3-3", "messages": [{"role": "user", "content": "What is Deep Learning?"}], "max_tokens":17}' \
      -H 'Content-Type: application/json'

EdgeCraftRAG/Dockerfile.server

@@ -23,6 +23,11 @@ RUN useradd -m -s /bin/bash user && \
     mkdir -p /home/user && \
     chown -R user /home/user/ 
 
+RUN mkdir /templates && \
+    chown -R user /templates
+COPY ./edgecraftrag/prompt_template/default_prompt.txt /templates/
+RUN chown -R user /templates/default_prompt.txt
+
 COPY ./edgecraftrag /home/user/edgecraftrag
 
 RUN mkdir -p /home/user/gradio_cache 

EdgeCraftRAG/README.md

@@ -32,14 +32,14 @@ Please follow this link [vLLM with OpenVINO](https://github.com/opea-project/Gen
 
 ### Start Edge Craft RAG Services with Docker Compose
 
-If you want to enable vLLM with OpenVINO service, please finish the steps in [Launch vLLM with OpenVINO service](#optional-launch-vllm-with-openvino-service) first.
-
 ```bash
 cd GenAIExamples/EdgeCraftRAG/docker_compose/intel/gpu/arc
 
 export MODEL_PATH="your model path for all your models"
 export DOC_PATH="your doc path for uploading a dir of files"
 export GRADIO_PATH="your gradio cache path for transferring files"
+# If you have a specific prompt template, please uncomment the following line
+# export PROMPT_PATH="your prompt path for prompt templates"
 
 # Make sure all 3 folders have 1000:1000 permission, otherwise
 # chown 1000:1000 ${MODEL_PATH} ${DOC_PATH} ${GRADIO_PATH}
@@ -70,49 +70,25 @@ optimum-cli export openvino -m BAAI/bge-small-en-v1.5 ${MODEL_PATH}/BAAI/bge-sma
 optimum-cli export openvino -m BAAI/bge-reranker-large ${MODEL_PATH}/BAAI/bge-reranker-large --task sentence-similarity
 optimum-cli export openvino -m Qwen/Qwen2-7B-Instruct ${MODEL_PATH}/Qwen/Qwen2-7B-Instruct/INT4_compressed_weights --weight-format int4
 
-docker compose up -d
+```
+
+#### Launch services with local inference
 
+```bash
+docker compose -f compose.yaml up -d
 ```
 
-#### (Optional) Launch vLLM with OpenVINO service
+#### Launch services with vLLM + OpenVINO inference service
 
-1. Set up Environment Variables
+Set up Additional Environment Variables and start with compose_vllm.yaml
 
 ```bash
 export LLM_MODEL=#your model id
 export VLLM_SERVICE_PORT=8008
 export vLLM_ENDPOINT="http://${HOST_IP}:${VLLM_SERVICE_PORT}"
 export HUGGINGFACEHUB_API_TOKEN=#your HF token
-```
-
-2. Uncomment below code in 'GenAIExamples/EdgeCraftRAG/docker_compose/intel/gpu/arc/compose.yaml'
 
-```bash
-  # vllm-openvino-server:
-  #   container_name: vllm-openvino-server
-  #   image: opea/vllm-arc:latest
-  #   ports:
-  #     - ${VLLM_SERVICE_PORT:-8008}:80
-  #   environment:
-  #     HTTPS_PROXY: ${https_proxy}
-  #     HTTP_PROXY: ${https_proxy}
-  #     VLLM_OPENVINO_DEVICE: GPU
-  #     HF_ENDPOINT: ${HF_ENDPOINT}
-  #     HF_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
-  #   volumes:
-  #     - /dev/dri/by-path:/dev/dri/by-path
-  #     - $HOME/.cache/huggingface:/root/.cache/huggingface
-  #   devices:
-  #     - /dev/dri
-  #   entrypoint: /bin/bash -c "\
-  #     cd / && \
-  #     export VLLM_CPU_KVCACHE_SPACE=50 && \
-  #     export VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS=ON && \
-  #     python3 -m vllm.entrypoints.openai.api_server \
-  #       --model '${LLM_MODEL}' \
-  #       --max_model_len=1024 \
-  #       --host 0.0.0.0 \
-  #       --port 80"
+docker compose -f compose_vllm.yaml up -d
 ```
 
 ### ChatQnA with LLM Example (Command Line)

EdgeCraftRAG/docker_compose/intel/gpu/arc/compose.yaml

@@ -16,6 +16,7 @@ services:
       - ${DOC_PATH:-${PWD}}:/home/user/docs
       - ${GRADIO_PATH:-${PWD}}:/home/user/gradio_cache
       - ${HF_CACHE:-${HOME}/.cache}:/home/user/.cache
+      - ${PROMPT_PATH:-${PWD}}:/templates/custom
     ports:
       - ${PIPELINE_SERVICE_PORT:-16010}:${PIPELINE_SERVICE_PORT:-16010}
     devices:

EdgeCraftRAG/docker_compose/intel/gpu/arc/compose_vllm.yaml

@@ -16,6 +16,7 @@ services:
       - ${DOC_PATH:-${PWD}}:/home/user/docs
       - ${GRADIO_PATH:-${PWD}}:/home/user/gradio_cache
       - ${HF_CACHE:-${HOME}/.cache}:/home/user/.cache
+      - ${PROMPT_PATH:-${PWD}}:/templates/custom
     ports:
       - ${PIPELINE_SERVICE_PORT:-16010}:${PIPELINE_SERVICE_PORT:-16010}
     devices:

EdgeCraftRAG/edgecraftrag/components/generator.py

@@ -26,12 +26,13 @@ def __init__(self, llm_model, prompt_template, inference_type, **kwargs):
             ("\n\n", "\n"),
             ("\t\n", "\n"),
         )
-        template = prompt_template
-        self.prompt = (
-            DocumentedContextRagPromptTemplate.from_file(template)
-            if os.path.isfile(template)
-            else DocumentedContextRagPromptTemplate.from_template(template)
-        )
+        safe_root = "/templates"
+        template = os.path.normpath(os.path.join(safe_root, prompt_template))
+        if not template.startswith(safe_root):
+            raise ValueError("Invalid template path")
+        if not os.path.exists(template):
+            raise ValueError("Template file not exists")
+        self.prompt = DocumentedContextRagPromptTemplate.from_file(template)
         self.llm = llm_model
         if isinstance(llm_model, str):
             self.model_id = llm_model

EdgeCraftRAG/tests/configs/test_pipeline_local_llm.json

@@ -37,7 +37,7 @@
       "device": "auto",
       "weight": "INT4"
     },
-    "prompt_path": "./edgecraftrag/prompt_template/default_prompt.txt",
+    "prompt_path": "./default_prompt.txt",
     "inference_type": "local"
   },
   "active": "True"

EdgeCraftRAG/tests/configs/test_pipeline_vllm.json

@@ -37,7 +37,7 @@
       "device": "auto",
       "weight": "INT4"
     },
-    "prompt_path": "./edgecraftrag/prompt_template/default_prompt.txt",
+    "prompt_path": "./default_prompt.txt",
     "inference_type": "vllm"
   },
   "active": "True"

EdgeCraftRAG/tests/test_compose_vllm_on_arc.sh

@@ -31,8 +31,7 @@ vLLM_ENDPOINT="http://${HOST_IP}:${VLLM_SERVICE_PORT}"
 function build_docker_images() {
     cd $WORKPATH/docker_image_build
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."
-    service_list="server ui ecrag"
-    docker compose -f build.yaml build ${service_list} --no-cache > ${LOG_PATH}/docker_image_build.log
+    docker compose -f build.yaml build --no-cache > ${LOG_PATH}/docker_image_build.log
 
     echo "Build vllm_openvino image from GenAIComps..."
     cd $WORKPATH && git clone https://github.com/opea-project/GenAIComps.git && cd GenAIComps && git checkout "${opea_branch:-"main"}"

EdgeCraftRAG/tests/test_pipeline_local_llm.json

@@ -37,7 +37,7 @@
       "device": "auto",
       "weight": "INT4"
     },
-    "prompt_path": "./edgecraftrag/prompt_template/default_prompt.txt",
+    "prompt_path": "./default_prompt.txt",
     "inference_type": "local"
   },
   "active": "True"

EdgeCraftRAG/ui/gradio/default.yaml

@@ -29,7 +29,7 @@ postprocessor: "reranker"
 
 # Generator
 generator: "chatqna"
-prompt_path: "./edgecraftrag/prompt_template/default_prompt.txt"
+prompt_path: "./default_prompt.txt"
 
 # Models
 embedding_model_id: "BAAI/bge-small-en-v1.5"

EdgeCraftRAG/ui/gradio/ecrag_client.py

@@ -78,7 +78,7 @@ def create_update_pipeline(
         ],
         generator=api_schema.GeneratorIn(
             # TODO: remove hardcoding
-            prompt_path="./edgecraftrag/prompt_template/default_prompt.txt",
+            prompt_path="./default_prompt.txt",
             model=api_schema.ModelIn(model_id=llm_id, model_path=llm_path, device=llm_device, weight=llm_weights),
             inference_type=llm_infertype,
         ),

Translation/kubernetes/intel/cpu/xeon/manifest/translation.yaml

@@ -10,7 +10,7 @@ metadata:
     app.kubernetes.io/instance: translation
     app.kubernetes.io/version: "2.1.0"
 data:
-  LLM_MODEL_ID: "haoranxu/ALMA-13B"
+  MODEL_ID: "haoranxu/ALMA-13B"
   PORT: "2080"
   HF_TOKEN: "insert-your-huggingface-token-here"
   http_proxy: ""

Translation/kubernetes/intel/hpu/gaudi/manifest/translation.yaml

@@ -10,7 +10,7 @@ metadata:
     app.kubernetes.io/instance: translation
     app.kubernetes.io/version: "2.1.0"
 data:
-  LLM_MODEL_ID: "haoranxu/ALMA-13B"
+  MODEL_ID: "haoranxu/ALMA-13B"
   PORT: "2080"
   HF_TOKEN: "insert-your-huggingface-token-here"
   http_proxy: ""