Add deep research agent

DeepResearchAgent/Dockerfile

@@ -0,0 +1,20 @@
+# Copyright (C) 2025 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+ARG IMAGE_REPO=opea
+ARG BASE_TAG=latest
+FROM $IMAGE_REPO/comps-base:$BASE_TAG
+
+COPY ./deep_researcher.yaml $HOME/deep_researcher.yaml
+COPY ./utils.py $HOME/utils.py
+COPY ./requirements.txt $HOME/requirements.txt
+COPY ./research_agent.py $HOME/research_agent.py
+
+USER root
+ARG uvpip='uv pip install --system --no-cache-dir'
+RUN pip install --no-cache-dir --upgrade pip setuptools uv && \
+    $uvpip -r requirements.txt
+
+USER user
+
+ENTRYPOINT ["python", "research_agent.py"]

DeepResearchAgent/README.md

@@ -0,0 +1,40 @@
+## Deep Research Agent Application
+
+Deep Research Agents are a new class of autonomous AI systems designed to perform complex, multi-step research tasks that typically require human-level reasoning, planning, and synthesis.
+
+## Setup Deployment Environment
+
+```
+# get your TAVILY_API_KEY from https://app.tavily.com/
+export TAVILY_API_KEY=""
+# get your HuggingFace Access Token from https://huggingface.co/docs/transformers.js/en/guides/private#step-1-generating-a-user-access-token
+export HF_TOKEN=""
+
+# set proxy if needed by your machine
+export http_proxy=""
+export https_proxy=""
+export no_proxy=""
+
+
+source ./set_env.sh
+```
+
+## Deploy the Services Using Docker Compose
+
+To deploy the Deep Research Agent services, execute the docker compose up command with the appropriate arguments. For a default deployment, execute:
+
+```
+docker compose -f docker_compose/intel/hpu/gaudi/compose.yaml up -d
+
+```
+
+## Validate Microservice
+
+```shell
+curl http://${host_ip}:8022/v1/deep_research_agent \
+    -X POST \
+    -d '{"question":"What is Deep Learning?"}' \
+    -H 'Content-Type: application/json'
+```
+
+## Benchmarks

DeepResearchAgent/benchmark/accuracy/README.md

@@ -0,0 +1 @@
+## Deep Research Agent Benchmarks

DeepResearchAgent/deep_researcher.yaml

@@ -0,0 +1,11 @@
+# Copyright (C) 2025 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+agent:
+  type: langchain_deep_researcher
+  search_api: "tavily"
+  planner_provider: "openai"
+  planner_model: "meta-llama/Llama-3.3-70B-Instruct"
+  writer_provider: "openai"
+  writer_model: "meta-llama/Llama-3.3-70B-Instruct"
+  max_search_depth: 2

DeepResearchAgent/docker_compose/intel/hpu/gaudi/compose.yaml

@@ -0,0 +1,57 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+
+x-common-environment:
+  &common-env
+  no_proxy: ${no_proxy}
+  http_proxy: ${http_proxy}
+  https_proxy: ${https_proxy}
+
+x-common-agent-environment:
+  &common-agent-env
+  <<: *common-env
+  HF_TOKEN: ${HF_TOKEN}
+  model: ${LLM_MODEL_ID}
+  TAVILY_API_KEY: ${TAVILY_API_KEY}
+  OPENAI_API_KEY: ${OPENAI_API_KEY}
+  OPENAI_BASE_URL: ${OPENAI_BASE_URL}
+
+services:
+
+  vllm-service:
+    image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
+    container_name: vllm-gaudi-server
+    ports:
+      - "8000:8000"
+    volumes:
+      - ${HF_CACHE_DIR:-./data}:/data
+    environment:
+      <<: *common-env
+      HF_TOKEN: ${HF_TOKEN}
+      HF_HOME: ./data
+      HABANA_VISIBLE_DEVICES: all
+      OMPI_MCA_btl_vader_single_copy_mechanism: none
+      LLM_MODEL_ID: ${LLM_MODEL_ID}
+      VLLM_TORCH_PROFILER_DIR: "/mnt"
+      VLLM_SKIP_WARMUP: true
+      PT_HPU_ENABLE_LAZY_COLLECTIVES: true
+    healthcheck:
+      test: ["CMD-SHELL", "curl -f http://$HOST_IP:8000/health || exit 1"]
+      interval: 10s
+      timeout: 10s
+      retries: 100
+    runtime: habana
+    cap_add:
+      - SYS_NICE
+    ipc: host
+    command: --model ${LLM_MODEL_ID} --tensor-parallel-size ${NUM_CARDS} --host 0.0.0.0 --port 8000 --max-seq-len-to-capture $MAX_LEN
+
+  deep-research-agent-server:
+    image: ${REGISTRY:-opea}/deep-research-agent:${TAG:-latest}
+    container_name: deep-research-agent-server
+    ports:
+      - "8022:8022"
+    ipc: host
+    environment:
+      <<: *common-agent-env

DeepResearchAgent/docker_compose/intel/set_env.sh

@@ -0,0 +1,45 @@
+#!/usr/bin/env bash
+
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+# Navigate to the parent directory and source the environment
+pushd "../../" > /dev/null
+source .set_env.sh
+popd > /dev/null
+
+# Function to check if a variable is set
+check_var() {
+    local var_name="$1"
+    local var_value="${!var_name}"
+    if [ -z "${var_value}" ]; then
+        echo "Error: ${var_name} is not set. Please set ${var_name}."
+        return 1  # Return an error code but do not exit the script
+    fi
+}
+
+# Check critical variables
+check_var "HF_TOKEN"
+export ip_address=$(hostname -I | awk '{print $1}')
+
+# VLLM configuration
+export VLLM_PORT="${VLLM_PORT:-8000}"
+export VLLM_VOLUME="${VLLM_VOLUME:-/data2/huggingface}"
+export VLLM_IMAGE="${VLLM_IMAGE:-opea/vllm-gaudi:latest}"
+export LLM_MODEL_ID="${LLM_MODEL_ID:-meta-llama/Llama-3.3-70B-Instruct}"
+export MAX_LEN="${MAX_LEN:-131072}"
+export NUM_CARDS="${NUM_CARDS:-4}"
+export HF_CACHE_DIR="${HF_CACHE_DIR:-"./data"}"
+export OPENAI_BASE_URL="http://${ip_address}:8000/v1"
+export OPENAI_API_KEY="empty"
+export no_proxy=${no_proxy}
+export http_proxy=${http_proxy}
+export https_proxy=${https_proxy}
+
+
+# Hugging Face API token
+export HF_TOKEN="${HF_TOKEN}"
+
+# API keys
+check_var "TAVILY_API_KEY"
+export TAVILY_API_KEY="${TAVILY_API_KEY}"

DeepResearchAgent/docker_image_build/build.yaml

@@ -0,0 +1,13 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+services:
+  deep_research_agent:
+    build:
+      args:
+        http_proxy: ${http_proxy}
+        https_proxy: ${https_proxy}
+        no_proxy: ${no_proxy}
+      context: ../
+      dockerfile: ./Dockerfile
+    image: ${REGISTRY:-opea}/deep-research-agent:${TAG:-latest}

DeepResearchAgent/requirements.txt

@@ -0,0 +1 @@
+open_deep_research # https://github.com/langchain-ai/open_deep_research

DeepResearchAgent/research_agent.py

@@ -0,0 +1,39 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+import argparse
+import json
+import os
+import re
+from typing import List, Union
+
+from comps import opea_microservices, register_microservice
+from comps.cores.telemetry.opea_telemetry import opea_telemetry
+from pydantic import BaseModel
+from utils import *
+
+agent = create_agent("./deep_researcher.yaml")
+
+
+class SimpleRequest(BaseModel):
+    question: Union[str, List[str]]
+
+
+@register_microservice(
+    name="opea_service@deep_research_agent",
+    endpoint="/v1/deep_research_agent",
+    host="0.0.0.0",
+    port=8022,
+)
+@opea_telemetry
+async def run(request: SimpleRequest):
+
+    question = f"Question: {request.question}"
+
+    result = await agent(question)
+
+    return result
+
+
+if __name__ == "__main__":
+    opea_microservices["opea_service@deep_research_agent"].start()

DeepResearchAgent/tests/test_compose_on_gaudi.sh

@@ -0,0 +1,117 @@
+#!/bin/bash
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+set -xe
+IMAGE_REPO=${IMAGE_REPO:-"opea"}
+IMAGE_TAG=${IMAGE_TAG:-"latest"}
+echo "REGISTRY=IMAGE_REPO=${IMAGE_REPO}"
+echo "TAG=IMAGE_TAG=${IMAGE_TAG}"
+export REGISTRY=${IMAGE_REPO}
+export TAG=${IMAGE_TAG}
+export MODEL_CACHE=${model_cache:-"./data"}
+export HF_TOKEN=${HF_TOKEN}
+export TAVILY_API_KEY=${TAVILY_API_KEY}
+
+WORKPATH=$(dirname "$PWD")
+LOG_PATH="$WORKPATH/tests"
+ip_address=$(hostname -I | awk '{print $1}')
+
+function build_docker_images() {
+    cd $WORKPATH/docker_image_build
+    echo "Build all the images with --no-cache, check docker_image_build.log for details..."
+    docker compose -f build.yaml build --no-cache > ${LOG_PATH}/docker_image_build.log
+
+    docker images && sleep 1s
+}
+
+function start_services() {
+    cd $WORKPATH/docker_compose/intel/hpu/gaudi
+    source set_env.sh
+
+    # Start Docker Containers
+    docker compose -f compose.yaml up -d --quiet-pull > ${LOG_PATH}/start_services_with_compose.log
+    n=0
+    until [[ "$n" -ge 200 ]]; do
+        echo "n=$n"
+        docker logs vllm-gaudi-server > vllm_service_start.log 2>&1
+        if grep -q "Warmup finished" vllm_service_start.log; then
+            break
+        fi
+        sleep 5s
+        n=$((n+1))
+    done
+}
+
+function validate_service() {
+    local URL="$1"
+    local EXPECTED_RESULT="$2"
+    local SERVICE_NAME="$3"
+    local DOCKER_NAME="$4"
+    local INPUT_DATA="$5"
+
+    local HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$URL")
+    if [ "$HTTP_STATUS" -eq 200 ]; then
+        echo "[ $SERVICE_NAME ] HTTP status is 200. Checking content..."
+
+        local CONTENT=$(curl -s -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$URL" | tee ${LOG_PATH}/${SERVICE_NAME}.log)
+
+        if echo "$CONTENT" | grep -q "$EXPECTED_RESULT"; then
+            echo "[ $SERVICE_NAME ] Content is as expected."
+        else
+            echo "[ $SERVICE_NAME ] Content does not match the expected result: $CONTENT"
+            docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+            exit 1
+        fi
+    else
+        echo "[ $SERVICE_NAME ] HTTP status is not 200. Received status was $HTTP_STATUS"
+        docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+        exit 1
+    fi
+    sleep 1s
+}
+
+function validate_microservices() {
+    # Check if the microservices are running correctly.
+
+    validate_service \
+        "${ip_address}:8022/v1/deep_research_agent" \
+	"deep" \
+        "deep-research-agent" \
+        "deep-research-agent-server" \
+        '{"question": "what is the deep learning?"}'
+}
+
+
+function stop_docker() {
+    cd $WORKPATH/docker_compose/intel/hpu/gaudi
+    docker compose -f compose.yaml down
+}
+
+function main() {
+
+    echo "::group::stop_docker"
+    stop_docker
+    echo "::endgroup::"
+
+    echo "::group::build_docker_images"
+    if [[ "$IMAGE_REPO" == "opea" ]]; then build_docker_images; fi
+    echo "::endgroup::"
+
+    echo "::group::start_services"
+    start_services
+    echo "::endgroup::"
+
+    echo "::group::validate_microservices"
+    validate_microservices
+    echo "::endgroup::"
+
+    echo "::group::stop_docker"
+    stop_docker
+    echo "::endgroup::"
+
+    docker system prune -f
+
+}
+
+main