diff --git a/docs/toolbox.generated/Cluster.capture_servicemonitor_metrics.rst b/docs/toolbox.generated/Cluster.capture_servicemonitor_metrics.rst
new file mode 100644
index 0000000000..e856bc8e54
--- /dev/null
+++ b/docs/toolbox.generated/Cluster.capture_servicemonitor_metrics.rst
@@ -0,0 +1,48 @@
+:orphan:
+
+..
+    _Auto-generated file, do not edit manually ...
+    _Toolbox generate command: repo generate_toolbox_rst_documentation
+    _ Source component: Cluster.capture_servicemonitor_metrics
+
+
+cluster capture_servicemonitor_metrics
+======================================
+
+Captures ServiceMonitor or PodMonitor YAML and status for a given service
+
+Captures the ServiceMonitor/PodMonitor configuration and status information for
+a specific service in a namespace, including related service/pod and
+endpoints information for troubleshooting monitoring setup.
+
+
+Parameters
+----------
+
+
+``service_name``  
+
+* Name of the service to capture ServiceMonitor/PodMonitor metrics for
+
+
+``namespace``  
+
+* Namespace where the service and ServiceMonitor/PodMonitor are located (empty string auto-detects current namespace)
+
+
+``capture_frequency``  
+
+* How often to capture metrics in seconds (default: 15)
+
+* default value: ``60``
+
+
+``is_podmonitor``  
+
+* Whether to use PodMonitor instead of ServiceMonitor (default: False)
+
+
+``finalize``  
+
+* Whether to finalize (capture logs and delete) an existing pod instead of creating new one (default: False)
+
diff --git a/docs/toolbox.generated/Llmd.run_guidellm_benchmark.rst b/docs/toolbox.generated/Llmd.run_guidellm_benchmark.rst
index 6b6e08e843..70eaf7bc94 100644
--- a/docs/toolbox.generated/Llmd.run_guidellm_benchmark.rst
+++ b/docs/toolbox.generated/Llmd.run_guidellm_benchmark.rst
@@ -39,14 +39,14 @@ Parameters
 
 * Container image for the benchmark
 
-* default value: ``ghcr.io/vllm-project/guidellm``
+* default value: ``ghcr.io/albertoperdomo2/guidellm``
 
 
 ``version``  
 
 * Version tag for the benchmark image
 
-* default value: ``pr-590``
+* default value: ``nightly``
 
 
 ``timeout``  
diff --git a/docs/toolbox.generated/index.rst b/docs/toolbox.generated/index.rst
index 0668c6950c..acd27da3f7 100644
--- a/docs/toolbox.generated/index.rst
+++ b/docs/toolbox.generated/index.rst
@@ -14,6 +14,7 @@ Toolbox Documentation
                 
 * :doc:`build_push_image <Cluster.build_push_image>`	 Build and publish an image to quay using either a Dockerfile or git repo.
 * :doc:`capture_environment <Cluster.capture_environment>`	 Captures the cluster environment
+* :doc:`capture_servicemonitor_metrics <Cluster.capture_servicemonitor_metrics>`	 Captures ServiceMonitor or PodMonitor YAML and status for a given service
 * :doc:`create_htpasswd_adminuser <Cluster.create_htpasswd_adminuser>`	 Create an htpasswd admin user.
 * :doc:`create_osd <Cluster.create_osd>`	 Create an OpenShift Dedicated cluster.
 * :doc:`deploy_operator <Cluster.deploy_operator>`	 Deploy an operator from OperatorHub catalog entry.
diff --git a/projects/cluster/toolbox/cluster.py b/projects/cluster/toolbox/cluster.py
index ee8c777dce..170123e2f8 100644
--- a/projects/cluster/toolbox/cluster.py
+++ b/projects/cluster/toolbox/cluster.py
@@ -547,3 +547,23 @@ def enable_userworkload_monitoring(self, namespaces: list = []):
         """
 
         return RunAnsibleRole(locals())
+
+    @AnsibleRole("cluster_capture_servicemonitor_metrics")
+    @AnsibleMappedParams
+    def capture_servicemonitor_metrics(self, service_name, namespace="", capture_frequency=60, is_podmonitor=False, finalize=False):
+        """
+        Captures ServiceMonitor or PodMonitor YAML and status for a given service
+
+        Captures the ServiceMonitor/PodMonitor configuration and status information for
+        a specific service in a namespace, including related service/pod and
+        endpoints information for troubleshooting monitoring setup.
+
+        Args:
+            service_name: Name of the service to capture ServiceMonitor/PodMonitor metrics for
+            namespace: Namespace where the service and ServiceMonitor/PodMonitor are located (empty string auto-detects current namespace)
+            capture_frequency: How often to capture metrics in seconds (default: 15)
+            is_podmonitor: Whether to use PodMonitor instead of ServiceMonitor (default: False)
+            finalize: Whether to finalize (capture logs and delete) an existing pod instead of creating new one (default: False)
+        """
+
+        return RunAnsibleRole(locals())
diff --git a/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/defaults/main/config.yml b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/defaults/main/config.yml
new file mode 100644
index 0000000000..cf6de0a76f
--- /dev/null
+++ b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/defaults/main/config.yml
@@ -0,0 +1,24 @@
+# Auto-generated file, do not edit manually ...
+# Toolbox generate command: repo generate_ansible_default_settings
+# Source component: Cluster.capture_servicemonitor_metrics
+
+# Parameters
+# Name of the service to capture ServiceMonitor/PodMonitor metrics for
+# Mandatory value
+cluster_capture_servicemonitor_metrics_service_name:
+
+# Namespace where the service and ServiceMonitor/PodMonitor are located (empty string auto-detects current namespace)
+cluster_capture_servicemonitor_metrics_namespace:
+
+# How often to capture metrics in seconds (default: 15)
+cluster_capture_servicemonitor_metrics_capture_frequency: 60
+
+# Whether to use PodMonitor instead of ServiceMonitor (default: False)
+cluster_capture_servicemonitor_metrics_is_podmonitor: false
+
+# Whether to finalize (capture logs and delete) an existing pod instead of creating new one (default: False)
+cluster_capture_servicemonitor_metrics_finalize: false
+
+# Default Ansible variables
+# Default value for ansible_os_family to ensure role remains standalone
+ansible_os_family: Linux
diff --git a/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/main.yml b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/main.yml
new file mode 100644
index 0000000000..fd9d1d0c02
--- /dev/null
+++ b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/main.yml
@@ -0,0 +1,98 @@
+---
+- name: Get current namespace if not specified
+  command: oc project -q
+  register: current_namespace_result
+  when: cluster_capture_servicemonitor_metrics_namespace == ""
+
+- name: Set the target namespace
+  set_fact:
+    target_namespace: "{{ current_namespace_result.stdout if cluster_capture_servicemonitor_metrics_namespace == '' else cluster_capture_servicemonitor_metrics_namespace }}"
+
+- name: Create capture directory
+  file:
+    path: "{{ artifact_extra_logs_dir }}/artifacts"
+    state: directory
+    mode: '0755'
+
+- name: "[Finalize mode] capture logs and delete existing pod"
+  when: cluster_capture_servicemonitor_metrics_finalize
+  block:
+  - name: Check if pod exists for finalization
+    shell: |
+      oc get pod topsail-metrics-capture-{{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} --no-headers -o name
+    register: pod_exists_check
+
+  - name: Capture pod logs
+    shell: |
+      oc logs topsail-metrics-capture-{{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} > "{{ artifact_extra_logs_dir }}/artifacts/metrics_capture_logs.txt"
+
+- name: Delete metrics capture pod
+  shell: |
+    oc delete pod topsail-metrics-capture-{{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }}  --grace-period=0 --ignore-not-found
+
+- name: "[Finalize mode] capture logs and delete existing pod"
+  when: cluster_capture_servicemonitor_metrics_finalize
+  meta: end_play
+
+# Normal deployment mode: discover resources and create pod
+- name: Include ServiceMonitor tasks
+  include_tasks: servicemonitor.yml
+  when: not cluster_capture_servicemonitor_metrics_is_podmonitor
+
+- name: Include PodMonitor tasks
+  include_tasks: podmonitor.yml
+  when: cluster_capture_servicemonitor_metrics_is_podmonitor
+
+# Ensure auth_secret_name is always defined with proper structure (fallback for edge cases)
+- name: Set default auth secret name if not defined
+  set_fact:
+    auth_secret_name: "{% if auth_secret_name_cmd is defined %}{{ auth_secret_name_cmd.stdout }}{% endif %}"
+
+# Common tasks for deployment mode
+- name: Read metrics URL
+  shell: cat "{{ artifact_extra_logs_dir }}/artifacts/metrics_url.txt"
+  register: metrics_url_content
+
+- name: Check if auth secret exists
+  shell: |
+    oc get secret {{ auth_secret_name }} -n {{ target_namespace }} --no-headers -o name
+  register: auth_secret_exists
+  when: auth_secret_name != ""
+
+- name: Save authentication info
+  shell: |
+    {% if cluster_capture_servicemonitor_metrics_is_podmonitor %}
+    echo "PodMonitor: {{ cluster_capture_servicemonitor_metrics_service_name }}" > "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+    {% else %}
+    echo "ServiceMonitor: {{ cluster_capture_servicemonitor_metrics_service_name }}" > "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+    {% endif %}
+    echo "Auth secret name: {{ auth_secret_name | default('none') }}" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+    if [ -n "{{ auth_secret_name }}" ]; then
+      {% if auth_secret_name != '' and auth_secret_exists.rc == 0 %}
+      echo "Secret exists: yes" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+      echo "Secret will be mounted at: /var/run/secrets/auth/token" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+      {% else %}
+      echo "Secret exists: no" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+      echo "WARNING: Secret not found!" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+      {% endif %}
+    else
+      echo "No authentication required" >> "{{ artifact_extra_logs_dir }}/artifacts/auth_info.txt"
+    fi
+
+- name: Create metrics capture pod manifest
+  template:
+    src: metrics_capture_pod.yaml.j2
+    dest: "{{ artifact_extra_logs_dir }}/artifacts/metrics_capture_pod.yaml"
+    mode: '0644'
+  vars:
+    metrics_url: "{{ metrics_url_content.stdout }}"
+    auth_secret_name: "{{ auth_secret_name | default('') }}"
+    capture_frequency: "{{ cluster_capture_servicemonitor_metrics_capture_frequency }}"
+
+- name: Create metrics capture pod
+  shell: |
+    oc create -f "{{ artifact_extra_logs_dir }}/artifacts/metrics_capture_pod.yaml"
+
+- name: Wait for pod to start
+  shell: |
+    oc wait --for=condition=Ready pod/topsail-metrics-capture-{{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} --timeout=60s
diff --git a/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/podmonitor.yml b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/podmonitor.yml
new file mode 100644
index 0000000000..98d9936ea6
--- /dev/null
+++ b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/podmonitor.yml
@@ -0,0 +1,75 @@
+---
+# PodMonitor-specific tasks
+
+- name: Capture PodMonitor YAML
+  shell: |
+    oc get podmonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -oyaml > "{{ artifact_extra_logs_dir }}/artifacts/podmonitor.yaml"
+
+- name: Get PodMonitor status
+  shell: |
+    oc get podmonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} > "{{ artifact_extra_logs_dir }}/artifacts/podmonitor.status"
+
+- name: Capture pod targets by PodMonitor selector
+  shell: |
+    oc get pod -l "app.kubernetes.io/component in (llminferenceservice-workload,llminferenceservice-workload-prefill,llminferenceservice-workload-worker,llminferenceservice-workload-leader,llminferenceservice-workload-leader-prefill,llminferenceservice-workload-worker-prefill),app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} > "{{ artifact_extra_logs_dir }}/artifacts/target_pods.status"
+
+- name: Capture pod targets by PodMonitor selector YAML
+  shell: |
+    oc get pod -l "app.kubernetes.io/component in (llminferenceservice-workload,llminferenceservice-workload-prefill,llminferenceservice-workload-worker,llminferenceservice-workload-leader,llminferenceservice-workload-leader-prefill,llminferenceservice-workload-worker-prefill),app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} -oyaml > "{{ artifact_extra_logs_dir }}/artifacts/target_pods.yaml"
+
+- name: Get all target pod IPs and names
+  shell: |
+    oc get pod -l "app.kubernetes.io/component in (llminferenceservice-workload,llminferenceservice-workload-prefill,llminferenceservice-workload-worker,llminferenceservice-workload-leader,llminferenceservice-workload-leader-prefill,llminferenceservice-workload-worker-prefill),app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} --no-headers -o custom-columns=":metadata.name,:status.podIP"
+  register: target_pods_info
+
+- name: Extract scheme from PodMonitor
+  shell: |
+    oc get podmonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -o jsonpath='{.spec.podMetricsEndpoints[0].scheme}' 2>/dev/null || echo "http"
+  register: metrics_scheme
+
+- name: Extract target port from PodMonitor
+  shell: |
+    oc get podmonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -o jsonpath='{.spec.podMetricsEndpoints[0].targetPort}' 2>/dev/null || echo "9090"
+  register: metrics_port
+
+- name: Build metrics URLs for all matching pods
+  shell: |
+    set -o pipefail;
+
+    SCHEME="{{ metrics_scheme.stdout | trim | default('http') }}"
+    PORT="{{ metrics_port.stdout | trim | default('9090') }}"
+
+    # Count total pods and initialize files
+    TOTAL_PODS=$(echo "{{ target_pods_info.stdout }}" | wc -l)
+
+    # Initialize files
+    echo "" > "{{ artifact_extra_logs_dir }}/artifacts/metrics_url.txt"
+    echo "PodMonitor target pods (found: $TOTAL_PODS):" > "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Scheme: $SCHEME" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Port: $PORT" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+
+    # Build URL list for each pod
+    while IFS= read -r line; do
+      if [ -n "$line" ]; then
+        POD_NAME=$(echo "$line" | awk '{print $1}')
+        POD_IP=$(echo "$line" | awk '{print $2}')
+        if [ -n "$POD_IP" ] && [ "$POD_IP" != "<none>" ]; then
+          URL="$SCHEME://$POD_IP:$PORT/metrics"
+          echo "$URL" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_url.txt"
+          echo "Pod: $POD_NAME ($POD_IP) -> $URL" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+        else
+          echo "Pod: $POD_NAME (no IP available)" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+        fi
+      fi
+    done <<< "{{ target_pods_info.stdout }}"
+
+    # Count valid URLs and add summary
+    VALID_URLS=$(grep -c "^http" "{{ artifact_extra_logs_dir }}/artifacts/metrics_url.txt" || echo "0")
+    echo "" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Total metrics URLs: $VALID_URLS" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+
+- name: Extract authorization secret name from PodMonitor
+  shell: |
+    oc get podmonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -o jsonpath='{.spec.podMetricsEndpoints[0].authorization.credentials.name}' 2>/dev/null || echo ""
+  register: auth_secret_name
diff --git a/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/servicemonitor.yml b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/servicemonitor.yml
new file mode 100644
index 0000000000..7e4dc51620
--- /dev/null
+++ b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/tasks/servicemonitor.yml
@@ -0,0 +1,76 @@
+---
+# ServiceMonitor-specific tasks
+
+- name: Capture ServiceMonitor YAML
+  shell: |
+    oc get servicemonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -oyaml > "{{ artifact_extra_logs_dir }}/artifacts/servicemonitor.yaml"
+
+- name: Get ServiceMonitor status
+  shell: |
+    oc get servicemonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} > "{{ artifact_extra_logs_dir }}/artifacts/servicemonitor.status"
+
+- name: Capture service target by ServiceMonitor selector
+  shell: |
+    oc get service -l "app.kubernetes.io/component=llminferenceservice-router-scheduler,app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} > "{{ artifact_extra_logs_dir }}/artifacts/target_service.status"
+
+- name: Capture service target by ServiceMonitor selector YAML
+  shell: |
+    oc get service -l "app.kubernetes.io/component=llminferenceservice-router-scheduler,app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} -oyaml > "{{ artifact_extra_logs_dir }}/artifacts/target_service.yaml"
+
+- name: Get target service name
+  shell: |
+    set -o pipefail;
+    oc get service -l "app.kubernetes.io/component=llminferenceservice-router-scheduler,app.kubernetes.io/part-of=llminferenceservice" -n {{ target_namespace }} --no-headers -o custom-columns=":metadata.name" | head -1
+  register: target_service_name
+
+- name: Extract port name from ServiceMonitor
+  shell: |
+    oc get servicemonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -o jsonpath='{.spec.endpoints[0].port}' 2>/dev/null || echo "metrics"
+  register: metrics_port_name
+
+- name: Get port number from Service by name
+  shell: |
+    SERVICE_NAME='{{ target_service_name.stdout | trim }}'
+    PORT_NAME='{{ metrics_port_name.stdout | trim | default("metrics") }}'
+    oc get service "$SERVICE_NAME" -n {{ target_namespace }} -o jsonpath="{.spec.ports[?(@.name=='$PORT_NAME')].port}" 2>/dev/null || echo ""
+  register: named_port_result
+
+- name: Get first port as fallback
+  shell: |
+    SERVICE_NAME='{{ target_service_name.stdout | trim }}'
+    oc get service "$SERVICE_NAME" -n {{ target_namespace }} -o jsonpath='{.spec.ports[0].port}' 2>/dev/null || echo "9090"
+  register: first_port_result
+  when: named_port_result.stdout == ""
+
+- name: Set final port number
+  set_fact:
+    final_port: "{{ named_port_result.stdout if named_port_result.stdout != '' else first_port_result.stdout | default('9090') }}"
+
+- name: Determine scheme from port
+  set_fact:
+    final_scheme: >-
+      {{
+        'https' if (
+          final_port in ['443', '8443', '6443'] or
+          (metrics_port_name.stdout | trim | default('metrics')) is match('.*(https|secure|tls).*')
+        ) else 'http'
+      }}
+
+- name: Build metrics URL for ServiceMonitor
+  shell: |
+    SERVICE_NAME='{{ target_service_name.stdout | trim }}'
+    PORT_NAME='{{ metrics_port_name.stdout | trim | default("metrics") }}'
+    PORT_NUMBER='{{ final_port }}'
+    SCHEME='{{ final_scheme }}'
+
+    echo "$SCHEME://$SERVICE_NAME.{{ target_namespace }}.svc:$PORT_NUMBER/metrics" > "{{ artifact_extra_logs_dir }}/artifacts/metrics_url.txt"
+    echo "Service: $SERVICE_NAME" > "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Port name: $PORT_NAME" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Port number: $PORT_NUMBER" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "Scheme: $SCHEME" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+    echo "URL: $SCHEME://$SERVICE_NAME.{{ target_namespace }}.svc:$PORT_NUMBER/metrics" >> "{{ artifact_extra_logs_dir }}/artifacts/metrics_info.txt"
+
+- name: Extract authorization secret name from ServiceMonitor
+  shell: |
+    oc get servicemonitor {{ cluster_capture_servicemonitor_metrics_service_name }} -n {{ target_namespace }} -o jsonpath='{.spec.endpoints[0].authorization.credentials.name}' 2>/dev/null || echo ""
+  register: auth_secret_name_cmd
diff --git a/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/templates/metrics_capture_pod.yaml.j2 b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/templates/metrics_capture_pod.yaml.j2
new file mode 100644
index 0000000000..d6c9b7044f
--- /dev/null
+++ b/projects/cluster/toolbox/cluster_capture_servicemonitor_metrics/templates/metrics_capture_pod.yaml.j2
@@ -0,0 +1,94 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: topsail-metrics-capture-{{ cluster_capture_servicemonitor_metrics_service_name }}
+  namespace: {{ target_namespace }}
+spec:
+  restartPolicy: Never
+  containers:
+  - name: metrics-capture
+    image: registry.access.redhat.com/ubi9/ubi:latest
+    command: ["/bin/bash"]
+    args:
+    - -c
+    - |
+      {% if auth_secret_name -%}
+      echo "Using authorization token from mounted secret: {{ auth_secret_name }}"
+      {% endif %}
+
+      # Function to fetch metrics with proper options
+      fetch_metrics() {
+        local url="$1"
+        local token="$2"
+
+        # Build curl options
+        local curl_opts="-sSf"
+        if [[ "$url" =~ ^https:// ]]; then
+          curl_opts="${curl_opts}k"
+        fi
+
+        # Execute curl with proper auth handling and filter output
+        local curl_cmd="curl $curl_opts"
+        if [ -n "$token" ]; then
+          if $curl_cmd -H "Authorization: Bearer $token" "$url" | grep -v "^# HELP\|^# TYPE\|^go_\|^process_\|^controller_runtime_\|^rest_client_\|^http_request_duration_\|^python_gc_"; then
+            return 0
+          else
+            echo "Failed to fetch metrics from $url"
+            return 1
+          fi
+        else
+          if $curl_cmd "$url" | grep -v "^# HELP\|^# TYPE\|^go_\|^process_\|^controller_runtime_\|^rest_client_\|^http_request_duration_\|^python_gc_"; then
+            return 0
+          else
+            echo "Failed to fetch metrics from $url"
+            return 1
+          fi
+        fi
+      }
+
+      # Get list of URLs to monitor from environment variable
+      echo "Monitoring the following URLs:"
+      echo "$METRICS_URLS"
+      echo ""
+
+      while true; do
+        echo "=== $(date) ==="
+
+        echo "$METRICS_URLS" | while IFS= read -r url; do
+          if [ -n "$url" ] && [ "$url" != " " ]; then
+            echo "Fetching from: $url at $(date +%s)"
+
+            {% if auth_secret_name %}
+            if [ -f /var/run/secrets/auth/token ]; then
+              TOKEN=$(cat /var/run/secrets/auth/token)
+              fetch_metrics "$url" "$TOKEN"
+            else
+              echo "Auth token file not found"
+            fi
+            {% else %}
+            fetch_metrics "$url" ""
+            {% endif %}
+
+            echo "--- End of $url at $(date +%s) ---"
+          fi
+        done
+
+        echo ""
+        sleep {{ capture_frequency }}
+      done
+    env:
+    - name: METRICS_URLS
+      value: |
+{{ metrics_url | indent(8, true) }}
+{% if auth_secret_name %}
+    volumeMounts:
+    - name: auth-token
+      mountPath: /var/run/secrets/auth
+      readOnly: true
+{% endif %}
+{% if auth_secret_name %}
+  volumes:
+  - name: auth-token
+    secret:
+      secretName: {{ auth_secret_name }}
+{% endif %}
diff --git a/projects/llm-d/testing/config.yaml b/projects/llm-d/testing/config.yaml
index a62489de5c..52690df29f 100644
--- a/projects/llm-d/testing/config.yaml
+++ b/projects/llm-d/testing/config.yaml
@@ -48,6 +48,9 @@ ci_presets:
 
     tests.capture_prom: false
     tests.capture_prom_uwm: false
+    tests.llmd.inference_service.metrics.manual_capture: false
+    tests.llmd.inference_service.gateway.name: gateway-internal
+
     tests.llmd.skip_prepare: true
     prepare.namespace.name: kpouget-dev
     prepare.preload.node_selector_key: gpu.nvidia.com/class
@@ -259,6 +262,10 @@ tests:
       gateway:
         name: gateway-external
       model: llama3-1-8b
+      metrics:
+        manual_capture: false
+        scheduler_servicemonitor_name: kserve-llm-isvc-scheduler
+        vllm_podmonitor_name: kserve-llm-isvc-vllm-engine
 
       # vLLM arguments (always applied)
       vllm_args:
@@ -286,6 +293,11 @@ tests:
       guidellm:
         enabled: true
         name: guidellm-benchmark
+
+        image:
+          name: ghcr.io/vllm-project/guidellm
+          version: pr-590
+
         backend_type: openai_http
         rate_type: concurrent
         max_seconds: 120
@@ -294,6 +306,7 @@ tests:
         data: prompt_tokens=256,output_tokens=128
         rate: 1
         sample_requests: 20
+        request_type: text_completions
 
   capture_prom: true
   capture_prom_uwm: true
diff --git a/projects/llm-d/testing/test_llmd.py b/projects/llm-d/testing/test_llmd.py
index 33ca3aff96..1f9a06d1f5 100644
--- a/projects/llm-d/testing/test_llmd.py
+++ b/projects/llm-d/testing/test_llmd.py
@@ -55,6 +55,9 @@ def test_single_flavor(flavor, flavor_index, total_flavors, namespace):
             # Deploy LLM inference service
             _, _, llmisvc_path = deploy_llm_inference_service(flavor, llmisvc_name, namespace, model_ref)
 
+            # Start metrics capture after deployment
+            start_metrics_capture(flavor)
+
             models = config.project.get_config(f"models")
             model_name = models[model_ref]["name"]
             logging.info(f"Using model: {model_name} (from config reference: {model_ref})")
@@ -75,6 +78,12 @@ def test_single_flavor(flavor, flavor_index, total_flavors, namespace):
             flavor_failed = e
 
         finally:
+            # Stop metrics capture after testing (success or failure)
+            try:
+                stop_metrics_capture(flavor)
+            except Exception as metrics_e:
+                logging.exception(f"Failed to stop metrics capture: {metrics_e}")
+
             # Always capture LLM inference service state (success or failure)
             logging.info("Capturing LLM inference service state for debugging")
             try:
@@ -892,6 +901,8 @@ def run_guidellm_benchmark(endpoint_url, llmisvc_name, namespace):
     data = config.project.get_config("tests.llmd.benchmarks.guidellm.data")
     sample_requests = config.project.get_config("tests.llmd.benchmarks.guidellm.sample_requests")
 
+    request_type = config.project.get_config("tests.llmd.benchmarks.guidellm.request_type")
+
     failed = False
 
     # Handle rate as list/tuple - iterate over each rate value
@@ -949,6 +960,9 @@ def evaluate_expression(match):
             if rate_type:
                 guidellm_args.append(f"--rate-type={rate_type}")
 
+            if request_type:
+                guidellm_args.append(f"--request-type={request_type}")
+
             # Add rate parameter
             guidellm_args.append(f"--rate={rate_value}")
 
@@ -1007,6 +1021,84 @@ def capture_llm_inference_service_state(llmisvc_name, namespace):
         logging.error(f"Failed to capture LLM inference service state: {e}")
 
 
+def start_metrics_capture(flavor):
+    """
+    Starts metrics capture for both ServiceMonitor and PodMonitor if enabled
+    """
+    if not config.project.get_config("tests.llmd.inference_service.metrics.manual_capture"):
+        return
+
+    logging.info("Starting metrics capture")
+
+    namespace = config.project.get_config("tests.llmd.namespace")
+    scheduler_name = config.project.get_config("tests.llmd.inference_service.metrics.scheduler_servicemonitor_name")
+    vllm_name = config.project.get_config("tests.llmd.inference_service.metrics.vllm_podmonitor_name")
+
+    # Parse flavor to check if it's simple
+    components = parse_flavor_components(flavor)
+    is_simple_flavor = components['base'] == "simple"
+
+    # Start vLLM PodMonitor capture (always)
+    logging.info(f"Starting PodMonitor metrics capture for {vllm_name}")
+    run.run_toolbox("cluster", "capture_servicemonitor_metrics",
+                    service_name=vllm_name,
+                    namespace=namespace,
+                    is_podmonitor=True,
+                    mute_stdout=True)
+
+    # Start scheduler ServiceMonitor capture (only for non-simple flavors)
+    if not is_simple_flavor:
+        logging.info(f"Starting ServiceMonitor metrics capture for {scheduler_name}")
+        run.run_toolbox("cluster", "capture_servicemonitor_metrics",
+                        service_name=scheduler_name,
+                        namespace=namespace,
+                        mute_stdout=True)
+    else:
+        logging.info("Skipping scheduler metrics capture for simple flavor")
+
+    logging.info("Metrics capture started successfully")
+
+
+def stop_metrics_capture(flavor):
+    """
+    Stops metrics capture for both ServiceMonitor and PodMonitor if enabled
+    """
+    if not config.project.get_config("tests.llmd.inference_service.metrics.manual_capture"):
+        return
+
+    logging.info("Stopping metrics capture")
+
+    namespace = config.project.get_config("tests.llmd.namespace")
+    scheduler_name = config.project.get_config("tests.llmd.inference_service.metrics.scheduler_servicemonitor_name")
+    vllm_name = config.project.get_config("tests.llmd.inference_service.metrics.vllm_podmonitor_name")
+
+    # Parse flavor to check if it's simple
+    components = parse_flavor_components(flavor)
+    is_simple_flavor = components['base'] == "simple"
+
+    # Stop vLLM PodMonitor capture (always)
+    logging.info(f"Stopping PodMonitor metrics capture for {vllm_name}")
+    run.run_toolbox("cluster", "capture_servicemonitor_metrics",
+                    service_name=vllm_name,
+                    namespace=namespace,
+                    is_podmonitor=True,
+                    finalize=True,
+                    mute_stdout=True,
+                    artifact_dir_suffix="_finalize",)
+
+    # Stop scheduler ServiceMonitor capture (only for non-simple flavors)
+    if not is_simple_flavor:
+        logging.info(f"Stopping ServiceMonitor metrics capture for {scheduler_name}")
+        run.run_toolbox("cluster", "capture_servicemonitor_metrics",
+                        service_name=scheduler_name,
+                        namespace=namespace,
+                        finalize=True,
+                        mute_stdout=True,
+                        artifact_dir_suffix="_finalize",)
+
+    logging.info("Metrics capture stopped successfully")
+
+
 def cleanup_llm_inference_resources():
     """
     Clean up all llminferenceservice resources in the namespace before testing.
diff --git a/projects/llm-d/toolbox/llmd.py b/projects/llm-d/toolbox/llmd.py
index 1def5b6dd1..e55c42a3eb 100644
--- a/projects/llm-d/toolbox/llmd.py
+++ b/projects/llm-d/toolbox/llmd.py
@@ -52,7 +52,7 @@ def run_guidellm_benchmark(
             self,
             endpoint_url,
             name="guidellm-benchmark", namespace="",
-            image="ghcr.io/vllm-project/guidellm", version="pr-590",
+            image="ghcr.io/albertoperdomo2/guidellm", version="nightly",
             timeout=900,
             pvc_size="1Gi",
             guidellm_args=[],
diff --git a/projects/llm-d/toolbox/llmd_run_guidellm_benchmark/defaults/main/config.yml b/projects/llm-d/toolbox/llmd_run_guidellm_benchmark/defaults/main/config.yml
index c8fca0e8ab..0594b19207 100644
--- a/projects/llm-d/toolbox/llmd_run_guidellm_benchmark/defaults/main/config.yml
+++ b/projects/llm-d/toolbox/llmd_run_guidellm_benchmark/defaults/main/config.yml
@@ -14,10 +14,10 @@ llmd_run_guidellm_benchmark_name: guidellm-benchmark
 llmd_run_guidellm_benchmark_namespace:
 
 # Container image for the benchmark
-llmd_run_guidellm_benchmark_image: ghcr.io/vllm-project/guidellm
+llmd_run_guidellm_benchmark_image: ghcr.io/albertoperdomo2/guidellm
 
 # Version tag for the benchmark image
-llmd_run_guidellm_benchmark_version: pr-590
+llmd_run_guidellm_benchmark_version: nightly
 
 # Timeout in seconds to wait for job completion
 llmd_run_guidellm_benchmark_timeout: 900
diff --git a/projects/llm-d/visualizations/llmd_inference/store/parsers.py b/projects/llm-d/visualizations/llmd_inference/store/parsers.py
index 644ddf529e..c61a8cd0ee 100644
--- a/projects/llm-d/visualizations/llmd_inference/store/parsers.py
+++ b/projects/llm-d/visualizations/llmd_inference/store/parsers.py
@@ -31,6 +31,8 @@
     # to support multiple benchmark directories (multi-rate scenarios)
     f"{artifact_dirnames.PROMETHEUS_DUMP_DIR}/prometheus.t*",
     f"{artifact_dirnames.PROMETHEUS_UWM_DUMP_DIR}/prometheus.t*",
+    f"{artifact_dirnames.GUIDELLM_BENCHMARK_DIR}/artifacts/results/benchmarks.json",
+    f"{artifact_dirnames.GUIDELLM_BENCHMARK_DIR}/artifacts/guidellm_benchmark_job.logs",
 ]