diff --git a/.buildkite/nightly-benchmarks/README.md b/.buildkite/nightly-benchmarks/README.md
new file mode 100644
index 0000000000000..6a18be947be99
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/README.md
@@ -0,0 +1,98 @@
+# vLLM benchmark suite
+
+## Introduction
+
+This directory contains the performance benchmarking CI for vllm.
+The goal is to help developers know the impact of their PRs on the performance of vllm.
+
+This benchmark will be *triggered* upon:
+- A PR being merged into vllm.
+- Every commit for those PRs with `perf-benchmarks` label.
+
+**Benchmarking Coverage**: latency, throughput and fix-qps serving on A100 (the support for more GPUs is comming later), with different models.
+
+**Benchmarking Duration**: about 1hr.
+
+## Configuring the workload for the quick benchmark
+
+The workload of the quick benchmark contains two parts: latency tests in `latency-tests.json`, throughput tests in `throughput-tests.json` and serving tests in `serving-tests.json`.
+
+### Latency test
+
+Here is an example of one test inside `latency-tests.json`:
+
+```json
+[
+    ...
+    {
+        "test_name": "latency_llama8B_tp1",
+        "parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "tensor_parallel_size": 1,
+            "load_format": "dummy",
+            "num_iters_warmup": 5,
+            "num_iters": 15
+        }
+    },
+    ...
+]
+```
+
+In this example:
+-  The `test_name` attributes is a unique identifier for the test. In `latency-tests.json`, it must start with `latency_`.
+-  The `parameters` attribute control the command line arguments to be used for `benchmark_latency.py`. Note that please use underline `_` instead of the dash `-` when specifying the command line arguments, and `run-benchmarks-suite.sh` will convert the underline to dash when feeding the arguments to `benchmark_latency.py`. For example, the corresponding command line arguments for `benchmark_latency.py` will be `--model meta-llama/Meta-Llama-3-8B --tensor-parallel-size 1 --load-format dummy --num-iters-warmup 5 --num-iters 15`
+
+Note that the performance numbers are highly sensitive to the value of the parameters. Please make sure the parameters are set correctly.
+
+WARNING: The benchmarking script will save json results by itself, so please do not configure `--output-json` parameter in the json file.
+
+
+### Throughput test
+The tests are specified in `throughput-tests.json`. The syntax is similar to `latency-tests.json`, except for that the parameters will be fed forward to `benchmark_throughput.py`.
+
+The number of this test is also stable -- a slight change on the value of this number might vary the performance numbers by a lot.
+
+### Serving test
+We test the throughput by using `benchmark_serving.py` with request rate = inf to cover the online serving overhead. The corresponding parameters are in `serving-tests.json`, and here is an example:
+
+```
+[
+    ...
+    {
+        "test_name": "serving_llama8B_tp1_sharegpt",
+        "qps_list": [1, 4, 16, "inf"],
+        "server_parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "tensor_parallel_size": 1,
+            "swap_space": 16,
+            "disable_log_stats": "",
+            "disable_log_requests": "",
+            "load_format": "dummy"
+        },
+        "client_parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "backend": "vllm",
+            "dataset_name": "sharegpt",
+            "dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200
+        }
+    },
+    ...
+]
+```
+
+Inside this example:
+- The `test_name` attribute is also a unique identifier for the test. It must start with `serving_`.
+- The `server-parameters` includes the command line arguments for vLLM server.
+- The `client-parameters` includes the command line arguments for `benchmark_serving.py`.
+- The `qps_list` controls the list of qps for test. It will be used to configure the `--request-rate` parameter in `benchmark_serving.py`
+
+The number of this test is less stable compared to the delay and latency benchmarks (due to randomized sharegpt dataset sampling inside `benchmark_serving.py`), but a large change on this number (e.g. 5% change) still vary the output greatly.
+
+WARNING: The benchmarking script will save json results by itself, so please do not configure `--save-results` or other results-saving-related parameters in `serving-tests.json`.
+
+## Visualizing the results
+The `convert-results-json-to-markdown.py` helps you put the benchmarking results inside a markdown table.
+You can find the result presented as a table inside the `buildkite/performance-benchmark` job page.
+If you do not see the table, please wait till the benchmark finish running.
+The JSON file is also attached within each buildkite job for further analysis.
\ No newline at end of file
diff --git a/.buildkite/nightly-benchmarks/benchmark-pipeline.yaml b/.buildkite/nightly-benchmarks/benchmark-pipeline.yaml
new file mode 100644
index 0000000000000..8f12748b68f39
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/benchmark-pipeline.yaml
@@ -0,0 +1,61 @@
+steps:
+  - label: "Wait for container to be ready"
+    agents:
+      queue: A100
+    plugins:
+    - kubernetes:
+        podSpec:
+          containers:
+          - image: badouralix/curl-jq
+            command:
+            - sh
+            - .buildkite/nightly-benchmarks/scripts/wait-for-image.sh
+  - wait
+  - label: "A100 Benchmark"
+    agents:
+      queue: A100
+    plugins:
+    - kubernetes:
+        podSpec:
+          containers:
+          - image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+            command:
+            - bash .buildkite/nightly-benchmarks/run-benchmarks-suite.sh
+            resources:
+              limits:
+                nvidia.com/gpu: 8
+            volumeMounts:
+            - name: devshm
+              mountPath: /dev/shm
+            env:
+            - name: VLLM_USAGE_SOURCE
+              value: ci-test
+            - name: HF_TOKEN
+              valueFrom:
+                secretKeyRef:
+                  name: hf-token-secret
+                  key: token
+          nodeSelector:
+            nvidia.com/gpu.product: NVIDIA-A100-SXM4-80GB
+          volumes:
+          - name: devshm
+            emptyDir:
+              medium: Memory
+  # - label: "H100: NVIDIA SMI"
+  #   agents:
+  #     queue: H100
+  #   plugins:
+  #   - docker#v5.11.0:
+  #       image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT
+  #       command:
+  #       - bash
+  #       - .buildkite/nightly-benchmarks/run-benchmarks-suite.sh
+  #       mount-buildkite-agent: true
+  #       propagate-environment: true
+  #       propagate-uid-gid: false
+  #       ipc: host
+  #       gpus: all
+  #       environment:
+  #       - VLLM_USAGE_SOURCE
+  #       - HF_TOKEN
+
diff --git a/.buildkite/nightly-benchmarks/kickoff-pipeline.sh b/.buildkite/nightly-benchmarks/kickoff-pipeline.sh
index d3bf3b72980a6..15d411febcee1 100755
--- a/.buildkite/nightly-benchmarks/kickoff-pipeline.sh
+++ b/.buildkite/nightly-benchmarks/kickoff-pipeline.sh
@@ -1,5 +1,6 @@
 #!/usr/bin/env bash
 
+# NOTE(simon): this script runs inside a buildkite agent with CPU only access.
 set -euo pipefail
 
 # Install system packages
@@ -23,4 +24,4 @@ if [ "$BUILDKITE_PULL_REQUEST" != "false" ]; then
 fi
 
 # Upload sample.yaml
-buildkite-agent pipeline upload .buildkite/nightly-benchmarks/sample.yaml
+buildkite-agent pipeline upload .buildkite/nightly-benchmarks/benchmark-pipeline.yaml
diff --git a/.buildkite/nightly-benchmarks/latency-tests.json b/.buildkite/nightly-benchmarks/latency-tests.json
new file mode 100644
index 0000000000000..294a8c439c3ae
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/latency-tests.json
@@ -0,0 +1,32 @@
+[
+    {
+        "test_name": "latency_llama8B_tp1",
+        "parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "tensor_parallel_size": 1,
+            "load_format": "dummy",
+            "num_iters_warmup": 5,
+            "num_iters": 15
+        }
+    },
+    {
+        "test_name": "latency_llama70B_tp4",
+        "parameters": {
+            "model": "meta-llama/Meta-Llama-3-70B-Instruct",
+            "tensor_parallel_size": 4,
+            "load_format": "dummy",
+            "num-iters-warmup": 5,
+            "num-iters": 15
+        }
+    },
+    {
+        "test_name": "latency_mixtral8x7B_tp2",
+        "parameters": {
+            "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+            "tensor_parallel_size": 2,
+            "load_format": "dummy",
+            "num-iters-warmup": 5,
+            "num-iters": 15
+        }
+    }
+]
diff --git a/.buildkite/nightly-benchmarks/run-benchmarks-suite.sh b/.buildkite/nightly-benchmarks/run-benchmarks-suite.sh
new file mode 100644
index 0000000000000..6cff6917f8ad5
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/run-benchmarks-suite.sh
@@ -0,0 +1,358 @@
+#!/bin/bash
+
+# This script should be run inside the CI process
+# This script assumes that we are already inside the vllm/ directory
+# Benchmarking results will be available inside vllm/benchmarks/results/
+
+# Do not set -e, as the mixtral 8x22B model tends to crash occasionally
+# and we still want to see other benchmarking results even when mixtral crashes.
+set -o pipefail
+
+check_gpus() {
+  # check the number of GPUs and GPU type.
+  declare -g gpu_count=$(nvidia-smi --list-gpus | wc -l)
+  if [[ $gpu_count -gt 0 ]]; then
+    echo "GPU found."
+  else
+    echo "Need at least 1 GPU to run benchmarking."
+    exit 1
+  fi
+  declare -g gpu_type=$(echo $(nvidia-smi --query-gpu=name --format=csv,noheader) | awk '{print $2}')
+  echo "GPU type is $gpu_type"
+}
+
+check_hf_token() {
+  # check if HF_TOKEN is available and valid
+  if [[ -z "$HF_TOKEN" ]]; then
+    echo "Error: HF_TOKEN is not set."
+    exit 1
+  elif [[ ! "$HF_TOKEN" =~ ^hf_ ]]; then
+    echo "Error: HF_TOKEN does not start with 'hf_'."
+    exit 1
+  else
+    echo "HF_TOKEN is set and valid."
+  fi
+}
+
+json2args() {
+  # transforms the JSON string to command line args, and '_' is replaced to '-'
+  # example:
+  # input: { "model": "meta-llama/Llama-2-7b-chat-hf", "tensor_parallel_size": 1 }
+  # output: --model meta-llama/Llama-2-7b-chat-hf --tensor-parallel-size 1
+  local json_string=$1
+  local args=$(
+    echo "$json_string" | jq -r '
+      to_entries |
+      map("--" + (.key | gsub("_"; "-")) + " " + (.value | tostring)) |
+      join(" ")
+    '
+  )
+  echo "$args"
+}
+
+wait_for_server() {
+  # wait for vllm server to start
+  # return 1 if vllm server crashes
+  timeout 1200 bash -c '
+    until curl localhost:8000/v1/completions; do
+      sleep 1
+    done' && return 0 || return 1
+}
+
+kill_gpu_processes() {
+  # kill all processes on GPU.
+  pids=$(nvidia-smi --query-compute-apps=pid --format=csv,noheader)
+  if [ -z "$pids" ]; then
+      echo "No GPU processes found."
+  else
+      for pid in $pids; do
+          kill -9 "$pid"
+          echo "Killed process with PID: $pid"
+      done
+
+      echo "All GPU processes have been killed."
+  fi
+
+  # waiting for GPU processes to be fully killed
+  sleep 10
+
+  # remove vllm config file
+  rm -rf ~/.config/vllm
+
+  # Print the GPU memory usage
+  # so that we know if all GPU processes are killed.
+  gpu_memory_usage=$(nvidia-smi --query-gpu=memory.used --format=csv,noheader,nounits -i 0)
+  # The memory usage should be 0 MB.
+  echo "GPU 0 Memory Usage: $gpu_memory_usage MB"
+}
+
+upload_to_buildkite() {
+  # upload the benchmarking results to buildkite
+
+  # if the agent binary is not found, skip uploading the results, exit 0
+  if [ ! -f /workspace/buildkite-agent ]; then
+    echo "buildkite-agent binary not found. Skip uploading the results."
+    return 0
+  fi
+  /workspace/buildkite-agent annotate --style "info" --context "benchmark-results" < $RESULTS_FOLDER/benchmark_results.md
+  /workspace/buildkite-agent artifact upload "$RESULTS_FOLDER/*"
+}
+
+run_latency_tests() {
+  # run latency tests using `benchmark_latency.py`
+  # $1: a json file specifying latency test cases
+
+  local latency_test_file
+  latency_test_file=$1
+
+  # Iterate over latency tests
+  jq -c '.[]' "$latency_test_file" | while read -r params; do
+    # get the test name, and append the GPU type back to it.
+    test_name=$(echo "$params" | jq -r '.test_name')
+    if [[ ! "$test_name" =~ ^latency_ ]]; then
+      echo "In latency-test.json, test_name must start with \"latency_\"."
+      exit 1
+    fi
+
+    # if TEST_SELECTOR is set, only run the test cases that match the selector
+    if [[ -n "$TEST_SELECTOR" ]] && [[ ! "$test_name" =~ $TEST_SELECTOR ]]; then
+      echo "Skip test case $test_name."
+      continue
+    fi
+
+    # get arguments
+    latency_params=$(echo "$params" | jq -r '.parameters')
+    latency_args=$(json2args "$latency_params")
+
+    # check if there is enough GPU to run the test
+    tp=$(echo "$latency_params" | jq -r '.tensor_parallel_size')
+    if [[ $gpu_count -lt $tp ]]; then
+      echo "Required tensor-parallel-size $tp but only $gpu_count GPU found. Skip testcase $testname."
+      continue
+    fi
+
+    latency_command="python3 benchmark_latency.py \
+      --output-json $RESULTS_FOLDER/${test_name}.json \
+      $latency_args"
+
+    echo "Running test case $test_name"
+    echo "Latency command: $latency_command"
+
+    # recoding benchmarking command ang GPU command
+    jq_output=$(jq -n \
+      --arg latency "$latency_command" \
+      --arg gpu "$gpu_type" \
+      '{
+        latency_command: $latency,
+        gpu_type: $gpu
+      }')
+    echo "$jq_output" > "$RESULTS_FOLDER/$test_name.commands"
+
+    # run the benchmark
+    eval "$latency_command"
+
+    kill_gpu_processes
+
+  done
+}
+
+
+run_throughput_tests() {
+  # run throughput tests using `benchmark_throughput.py`
+  # $1: a json file specifying throughput test cases
+
+  local throughput_test_file
+  throughput_test_file=$1
+
+  # Iterate over throughput tests
+  jq -c '.[]' "$throughput_test_file" | while read -r params; do
+    # get the test name, and append the GPU type back to it.
+    test_name=$(echo "$params" | jq -r '.test_name')
+    if [[ ! "$test_name" =~ ^throughput_ ]]; then
+      echo "In throughput-test.json, test_name must start with \"throughput_\"."
+      exit 1
+    fi
+
+    # if TEST_SELECTOR is set, only run the test cases that match the selector
+    if [[ -n "$TEST_SELECTOR" ]] && [[ ! "$test_name" =~ $TEST_SELECTOR ]]; then
+      echo "Skip test case $test_name."
+      continue
+    fi
+
+    # get arguments
+    throughput_params=$(echo "$params" | jq -r '.parameters')
+    throughput_args=$(json2args "$throughput_params")
+
+    # check if there is enough GPU to run the test
+    tp=$(echo $throughput_params | jq -r '.tensor_parallel_size')
+    if [[ $gpu_count -lt $tp ]]; then
+      echo "Required tensor-parallel-size $tp but only $gpu_count GPU found. Skip testcase $testname."
+      continue
+    fi
+
+    throughput_command="python3 benchmark_throughput.py \
+      --output-json $RESULTS_FOLDER/${test_name}.json \
+      $throughput_args"
+
+    echo "Running test case $test_name"
+    echo "Throughput command: $throughput_command"
+    # recoding benchmarking command ang GPU command
+    jq_output=$(jq -n \
+      --arg command "$throughput_command" \
+      --arg gpu "$gpu_type" \
+      '{
+        throughput_command: $command,
+        gpu_type: $gpu
+      }')
+    echo "$jq_output" > "$RESULTS_FOLDER/$test_name.commands"
+
+    # run the benchmark
+    eval "$throughput_command"
+
+    kill_gpu_processes
+
+  done
+}
+
+run_serving_tests() {
+  # run serving tests using `benchmark_serving.py`
+  # $1: a json file specifying serving test cases
+
+  local serving_test_file
+  serving_test_file=$1
+
+  # Iterate over serving tests
+  jq -c '.[]' "$serving_test_file" | while read -r params; do
+    # get the test name, and append the GPU type back to it.
+    test_name=$(echo "$params" | jq -r '.test_name')
+    if [[ ! "$test_name" =~ ^serving_ ]]; then
+      echo "In serving-test.json, test_name must start with \"serving_\"."
+      exit 1
+    fi
+
+    # if TEST_SELECTOR is set, only run the test cases that match the selector
+    if [[ -n "$TEST_SELECTOR" ]] && [[ ! "$test_name" =~ $TEST_SELECTOR ]]; then
+      echo "Skip test case $test_name."
+      continue
+    fi
+
+
+    # get client and server arguments
+    server_params=$(echo "$params" | jq -r '.server_parameters')
+    client_params=$(echo "$params" | jq -r '.client_parameters')
+    server_args=$(json2args "$server_params")
+    client_args=$(json2args "$client_params")
+    qps_list=$(echo "$params" | jq -r '.qps_list')
+    qps_list=$(echo "$qps_list" | jq -r '.[] | @sh')
+    echo "Running over qps list $qps_list"
+
+    # check if there is enough GPU to run the test
+    tp=$(echo "$server_params" | jq -r '.tensor_parallel_size')
+    if [[ $gpu_count -lt $tp ]]; then
+      echo "Required tensor-parallel-size $tp but only $gpu_count GPU found. Skip testcase $testname."
+      continue
+    fi
+
+    # check if server model and client model is aligned
+    server_model=$(echo "$server_params" | jq -r '.model')
+    client_model=$(echo "$client_params" | jq -r '.model')
+    if [[ $server_model != "$client_model" ]]; then
+      echo "Server model and client model must be the same. Skip testcase $testname."
+      continue
+    fi
+
+    server_command="python3 \
+      -m vllm.entrypoints.openai.api_server \
+      $server_args"
+
+    # run the server
+    echo "Running test case $test_name"
+    echo "Server command: $server_command"
+    eval "$server_command" &
+
+    # wait until the server is alive
+    wait_for_server
+    if [ $? -eq 0 ]; then
+      echo ""
+      echo "vllm server is up and running."
+    else
+      echo ""
+      echo "vllm failed to start within the timeout period."
+    fi
+
+    # iterate over different QPS
+    for qps in $qps_list; do
+      # remove the surrounding single quote from qps
+      if [[ "$qps" == *"inf"* ]]; then
+        echo "qps was $qps"
+        qps="inf"
+        echo "now qps is $qps"
+      fi
+
+      new_test_name=$test_name"_qps_"$qps
+
+      client_command="python3 benchmark_serving.py \
+        --save-result \
+        --result-dir $RESULTS_FOLDER \
+        --result-filename ${new_test_name}.json \
+        --request-rate $qps \
+        $client_args"
+
+      echo "Running test case $test_name with qps $qps"
+      echo "Client command: $client_command"
+
+      eval "$client_command"
+
+      # record the benchmarking commands
+      jq_output=$(jq -n \
+        --arg server "$server_command" \
+        --arg client "$client_command" \
+        --arg gpu "$gpu_type" \
+        '{
+          server_command: $server,
+          client_command: $client,
+          gpu_type: $gpu
+        }')
+      echo "$jq_output" > "$RESULTS_FOLDER/${new_test_name}.commands"
+
+    done
+
+    # clean up
+    kill_gpu_processes
+  done
+}
+
+main() {
+  check_gpus
+  check_hf_token
+
+  # dependencies
+  (which wget && which curl) || (apt-get update && apt-get install -y wget curl)
+  (which jq) || (apt-get update && apt-get -y install jq)
+
+  # get the current IP address, required by benchmark_serving.py
+  export VLLM_HOST_IP=$(hostname -I | awk '{print $1}')
+  # turn of the reporting of the status of each request, to clean up the terminal output
+  export VLLM_LOG_LEVEL="WARNING"
+
+  # prepare for benchmarking
+  cd benchmarks || exit 1
+  wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json
+  declare -g RESULTS_FOLDER=results/
+  mkdir -p $RESULTS_FOLDER
+  QUICK_BENCHMARK_ROOT=../.buildkite/nightly-benchmarks/
+
+  # benchmarking
+  run_serving_tests $QUICK_BENCHMARK_ROOT/serving-tests.json
+  run_latency_tests $QUICK_BENCHMARK_ROOT/latency-tests.json
+  run_throughput_tests $QUICK_BENCHMARK_ROOT/throughput-tests.json
+
+
+  # postprocess benchmarking results
+  pip install tabulate pandas
+  python3 $QUICK_BENCHMARK_ROOT/scripts/convert-results-json-to-markdown.py
+
+  upload_to_buildkite
+}
+
+main "$@"
diff --git a/.buildkite/nightly-benchmarks/sample.yaml b/.buildkite/nightly-benchmarks/sample.yaml
deleted file mode 100644
index 50e6e82072186..0000000000000
--- a/.buildkite/nightly-benchmarks/sample.yaml
+++ /dev/null
@@ -1,39 +0,0 @@
-steps:
-  # NOTE(simon): You can create separate blocks for different jobs
-  - label: "A100: NVIDIA SMI"
-    agents:
-      queue: A100
-    plugins:
-    - kubernetes:
-        podSpec:
-          containers:
-          # - image: us-central1-docker.pkg.dev/vllm-405802/vllm-ci-test-repo/vllm-test:$BUILDKITE_COMMIT
-          # TODO(simon): check latest main branch or use the PR image.
-          - image: us-central1-docker.pkg.dev/vllm-405802/vllm-ci-test-repo/vllm-test:45c35f0d58f4508bf43bd6af1d3d0d0ec0c915e6
-            command:
-            - bash -c 'nvidia-smi && nvidia-smi topo -m && pwd && ls'
-            resources:
-              limits:
-                nvidia.com/gpu: 8
-            volumeMounts:
-            - name: devshm
-              mountPath: /dev/shm
-          nodeSelector:
-            nvidia.com/gpu.product: NVIDIA-A100-SXM4-80GB
-          volumes:
-          - name: devshm
-            emptyDir:
-              medium: Memory
-  # TODO(simon): bring H100 online
-  # - label: "H100: NVIDIA SMI"
-  #   agents:
-  #     queue: H100
-  #   plugins:
-  #   - docker#v5.11.0:
-  #       image: us-central1-docker.pkg.dev/vllm-405802/vllm-ci-test-repo/vllm-test:45c35f0d58f4508bf43bd6af1d3d0d0ec0c915e6
-  #       command:
-  #       - bash -c 'nvidia-smi && nvidia-smi topo -m'
-  #       propagate-environment: true
-  #       ipc: host
-  #       gpus: all
-
diff --git a/.buildkite/nightly-benchmarks/scripts/convert-results-json-to-markdown.py b/.buildkite/nightly-benchmarks/scripts/convert-results-json-to-markdown.py
new file mode 100644
index 0000000000000..75cff84347942
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/scripts/convert-results-json-to-markdown.py
@@ -0,0 +1,155 @@
+import json
+from pathlib import Path
+
+import pandas as pd
+from tabulate import tabulate
+
+results_folder = Path("results/")
+
+# latency results and the keys that will be printed into markdown
+latency_results = []
+latency_column_mapping = {
+    "test_name": "Test name",
+    "gpu_type": "GPU",
+    "avg_latency": "Average latency (s)",
+    "P10": "P10 (s)",
+    "P25": "P25 (s)",
+    "P50": "P50 (s)",
+    "P75": "P75 (s)",
+    "P90": "P90 (s)",
+}
+
+# thoughput tests and the keys that will be printed into markdown
+throughput_results = []
+throughput_results_column_mapping = {
+    "test_name": "Test name",
+    "gpu_type": "GPU",
+    "num_requests": "# of req.",
+    "total_num_tokens": "Total # of tokens",
+    "elapsed_time": "Elapsed time (s)",
+    "requests_per_second": "Tput (req/s)",
+    "tokens_per_second": "Tput (tok/s)",
+}
+
+# serving results and the keys that will be printed into markdown
+serving_results = []
+serving_column_mapping = {
+    "test_name": "Test name",
+    "gpu_type": "GPU",
+    "completed": "# of req.",
+    "request_throughput": "Tput (req/s)",
+    "input_throughput": "Input Tput (tok/s)",
+    "output_throughput": "Output Tput (tok/s)",
+    "mean_ttft_ms": "Mean TTFT (ms)",
+    # do not say TTFT again to avoid the table getting too wide
+    "median_ttft_ms": "Median",
+    "p99_ttft_ms": "P99",
+    "mean_tpot_ms": "Mean TPOT (ms)",
+    "median_tpot_ms": "Median",
+    "p99_tpot_ms": "P99",
+    "mean_itl_ms": "Mean ITL (ms)",
+    "median_itl_ms": "Median",
+    "p99_itl_ms": "P99",
+}
+
+for test_file in results_folder.glob("*.json"):
+
+    with open(test_file, "r") as f:
+        raw_result = json.loads(f.read())
+
+    if "serving" in str(test_file):
+        # this result is generated via `benchmark_serving.py`
+
+        # attach the benchmarking command to raw_result
+        with open(test_file.with_suffix(".commands"), "r") as f:
+            command = json.loads(f.read())
+        raw_result.update(command)
+
+        # update the test name of this result
+        raw_result.update({"test_name": test_file.stem})
+
+        # add the result to raw_result
+        serving_results.append(raw_result)
+        continue
+
+    elif "latency" in f.name:
+        # this result is generated via `benchmark_latency.py`
+
+        # attach the benchmarking command to raw_result
+        with open(test_file.with_suffix(".commands"), "r") as f:
+            command = json.loads(f.read())
+        raw_result.update(command)
+
+        # update the test name of this result
+        raw_result.update({"test_name": test_file.stem})
+
+        # get different percentiles
+        for perc in [10, 25, 50, 75, 90]:
+            raw_result.update(
+                {f"P{perc}": raw_result["percentiles"][str(perc)]})
+
+        # add the result to raw_result
+        latency_results.append(raw_result)
+        continue
+
+    elif "throughput" in f.name:
+        # this result is generated via `benchmark_throughput.py`
+
+        # attach the benchmarking command to raw_result
+        with open(test_file.with_suffix(".commands"), "r") as f:
+            command = json.loads(f.read())
+        raw_result.update(command)
+
+        # update the test name of this result
+        raw_result.update({"test_name": test_file.stem})
+
+        # add the result to raw_result
+        throughput_results.append(raw_result)
+        continue
+
+    print(f"Skipping {test_file}")
+
+latency_results = pd.DataFrame.from_dict(latency_results)
+serving_results = pd.DataFrame.from_dict(serving_results)
+throughput_results = pd.DataFrame.from_dict(throughput_results)
+
+# remapping the key, for visualization purpose
+if not latency_results.empty:
+    latency_results = latency_results[list(
+        latency_column_mapping.keys())].rename(columns=latency_column_mapping)
+if not serving_results.empty:
+    serving_results = serving_results[list(
+        serving_column_mapping.keys())].rename(columns=serving_column_mapping)
+if not throughput_results.empty:
+    throughput_results = throughput_results[list(
+        throughput_results_column_mapping.keys())].rename(
+            columns=throughput_results_column_mapping)
+
+# get markdown tables
+latency_md_table = tabulate(latency_results,
+                            headers='keys',
+                            tablefmt='pipe',
+                            showindex=False)
+serving_md_table = tabulate(serving_results,
+                            headers='keys',
+                            tablefmt='pipe',
+                            showindex=False)
+throughput_md_table = tabulate(throughput_results,
+                               headers='keys',
+                               tablefmt='pipe',
+                               showindex=False)
+
+# document the result
+with open(results_folder / "benchmark_results.md", "w") as f:
+    if not latency_results.empty:
+        f.write("## Latency tests\n")
+        f.write(latency_md_table)
+        f.write("\n")
+    if not throughput_results.empty:
+        f.write("## Throughput tests\n")
+        f.write(throughput_md_table)
+        f.write("\n")
+    if not serving_results.empty:
+        f.write("## Serving tests\n")
+        f.write(serving_md_table)
+        f.write("\n")
diff --git a/.buildkite/nightly-benchmarks/scripts/wait-for-image.sh b/.buildkite/nightly-benchmarks/scripts/wait-for-image.sh
new file mode 100644
index 0000000000000..c785e6a0da628
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/scripts/wait-for-image.sh
@@ -0,0 +1,17 @@
+#!/bin/sh
+TOKEN=$(curl -s -L "https://public.ecr.aws/token?service=public.ecr.aws&scope=repository:q9t5s3a7/vllm-ci-test-repo:pull" | jq -r .token)
+URL="https://public.ecr.aws/v2/q9t5s3a7/vllm-ci-test-repo/manifests/$BUILDKITE_COMMIT"
+
+retries=0
+while [ $retries -lt 1000 ]; do
+    if [ $(curl -s -L -H "Authorization: Bearer $TOKEN" -o /dev/null -w "%{http_code}" $URL) -eq 200 ]; then
+        exit 0
+    fi
+
+    echo "Waiting for image to be available..."
+
+    retries=$((retries + 1))
+    sleep 5
+done
+
+exit 1
\ No newline at end of file
diff --git a/.buildkite/nightly-benchmarks/serving-tests.json b/.buildkite/nightly-benchmarks/serving-tests.json
new file mode 100644
index 0000000000000..bb674661279d3
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/serving-tests.json
@@ -0,0 +1,59 @@
+[
+    {
+        "test_name": "serving_llama8B_tp1_sharegpt",
+        "qps_list": [1, 4, 16, "inf"],
+        "server_parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "tensor_parallel_size": 1,
+            "swap_space": 16,
+            "disable_log_stats": "",
+            "disable_log_requests": "",
+            "load_format": "dummy"
+        },
+        "client_parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "backend": "vllm",
+            "dataset_name": "sharegpt",
+            "dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200
+        }
+    },
+    {
+        "test_name": "serving_llama70B_tp4_sharegpt",
+        "qps_list": [1, 4, 16, "inf"],
+        "server_parameters": {
+            "model": "meta-llama/Meta-Llama-3-70B-Instruct",
+            "tensor_parallel_size": 4,
+            "swap_space": 16,
+            "disable_log_stats": "",
+            "disable_log_requests": "",
+            "load_format": "dummy"
+        },
+        "client_parameters": {
+            "model": "meta-llama/Meta-Llama-3-70B-Instruct",
+            "backend": "vllm",
+            "dataset_name": "sharegpt",
+            "dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200
+        }
+    },
+    {
+        "test_name": "serving_mixtral8x7B_tp2_sharegpt",
+        "qps_list": [1, 4, 16, "inf"],
+        "server_parameters": {
+            "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+            "tensor_parallel_size": 2,
+            "swap_space": 16,
+            "disable_log_stats": "",
+            "disable_log_requests": "",
+            "load_format": "dummy"
+        },
+        "client_parameters": {
+            "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+            "backend": "vllm",
+            "dataset_name": "sharegpt",
+            "dataset_path": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200
+        }
+    }
+]
diff --git a/.buildkite/nightly-benchmarks/throughput-tests.json b/.buildkite/nightly-benchmarks/throughput-tests.json
new file mode 100644
index 0000000000000..db4f908d79971
--- /dev/null
+++ b/.buildkite/nightly-benchmarks/throughput-tests.json
@@ -0,0 +1,35 @@
+[
+    {
+        "test_name": "throughput_llama8B_tp1",
+        "parameters": {
+            "model": "meta-llama/Meta-Llama-3-8B",
+            "tensor_parallel_size": 1,
+            "load_format": "dummy",
+            "dataset": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200,
+            "backend": "vllm"
+        }
+    },
+    {
+        "test_name": "throughput_llama70B_tp4",
+        "parameters": {
+            "model": "meta-llama/Meta-Llama-3-70B-Instruct",
+            "tensor_parallel_size": 4,
+            "load_format": "dummy",
+            "dataset": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200,
+            "backend": "vllm"
+        }
+    },
+    {
+        "test_name": "throughput_mixtral8x7B_tp2",
+        "parameters": {
+            "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
+            "tensor_parallel_size": 2,
+            "load_format": "dummy",
+            "dataset": "./ShareGPT_V3_unfiltered_cleaned_split.json",
+            "num_prompts": 200,
+            "backend": "vllm"
+        }
+    }
+]
diff --git a/benchmarks/benchmark_latency.py b/benchmarks/benchmark_latency.py
index 17edb7515964a..9937f8333fb7e 100644
--- a/benchmarks/benchmark_latency.py
+++ b/benchmarks/benchmark_latency.py
@@ -10,6 +10,7 @@
 from tqdm import tqdm
 
 from vllm import LLM, SamplingParams
+from vllm.engine.arg_utils import EngineArgs
 from vllm.inputs import PromptStrictInputs
 from vllm.model_executor.layers.quantization import QUANTIZATION_METHODS
 
@@ -37,6 +38,7 @@ def main(args: argparse.Namespace):
               download_dir=args.download_dir,
               block_size=args.block_size,
               gpu_memory_utilization=args.gpu_memory_utilization,
+              load_format=args.load_format,
               distributed_executor_backend=args.distributed_executor_backend)
 
     sampling_params = SamplingParams(
@@ -222,6 +224,29 @@ def run_to_completion(profile_dir: Optional[str] = None):
                         help='the fraction of GPU memory to be used for '
                         'the model executor, which can range from 0 to 1.'
                         'If unspecified, will use the default value of 0.9.')
+    parser.add_argument(
+        '--load-format',
+        type=str,
+        default=EngineArgs.load_format,
+        choices=[
+            'auto', 'pt', 'safetensors', 'npcache', 'dummy', 'tensorizer',
+            'bitsandbytes'
+        ],
+        help='The format of the model weights to load.\n\n'
+        '* "auto" will try to load the weights in the safetensors format '
+        'and fall back to the pytorch bin format if safetensors format '
+        'is not available.\n'
+        '* "pt" will load the weights in the pytorch bin format.\n'
+        '* "safetensors" will load the weights in the safetensors format.\n'
+        '* "npcache" will load the weights in pytorch format and store '
+        'a numpy cache to speed up the loading.\n'
+        '* "dummy" will initialize the weights with random values, '
+        'which is mainly for profiling.\n'
+        '* "tensorizer" will load the weights using tensorizer from '
+        'CoreWeave. See the Tensorize vLLM Model script in the Examples'
+        'section for more information.\n'
+        '* "bitsandbytes" will load the weights using bitsandbytes '
+        'quantization.\n')
     parser.add_argument(
         '--distributed-executor-backend',
         choices=['ray', 'mp'],
diff --git a/benchmarks/benchmark_serving.py b/benchmarks/benchmark_serving.py
index 4112a3272518e..df32b366c4140 100644
--- a/benchmarks/benchmark_serving.py
+++ b/benchmarks/benchmark_serving.py
@@ -499,6 +499,8 @@ def main(args: argparse.Namespace):
         # Save to file
         base_model_id = model_id.split("/")[-1]
         file_name = f"{backend}-{args.request_rate}qps-{base_model_id}-{current_dt}.json"  #noqa
+        if args.result_filename:
+            file_name = args.result_filename
         if args.result_dir:
             file_name = os.path.join(args.result_dir, file_name)
         with open(file_name, "w") as outfile:
@@ -639,6 +641,15 @@ def main(args: argparse.Namespace):
         help="Specify directory to save benchmark json results."
         "If not specified, results are saved in the current directory.",
     )
+    parser.add_argument(
+        "--result-filename",
+        type=str,
+        default=None,
+        help="Specify the filename to save benchmark json results."
+        "If not specified, results will be saved in "
+        "{backend}-{args.request_rate}qps-{base_model_id}-{current_dt}.json"
+        " format.",
+    )
 
     args = parser.parse_args()
     main(args)
diff --git a/benchmarks/benchmark_throughput.py b/benchmarks/benchmark_throughput.py
index 07b2f85410e3c..463d9973d00dd 100644
--- a/benchmarks/benchmark_throughput.py
+++ b/benchmarks/benchmark_throughput.py
@@ -10,6 +10,7 @@
 from transformers import (AutoModelForCausalLM, AutoTokenizer,
                           PreTrainedTokenizerBase)
 
+from vllm.engine.arg_utils import EngineArgs
 from vllm.model_executor.layers.quantization import QUANTIZATION_METHODS
 
 
@@ -81,6 +82,7 @@ def run_vllm(
     distributed_executor_backend: Optional[str],
     gpu_memory_utilization: float = 0.9,
     download_dir: Optional[str] = None,
+    load_format: str = EngineArgs.load_format,
 ) -> float:
     from vllm import LLM, SamplingParams
     llm = LLM(
@@ -102,6 +104,7 @@ def run_vllm(
         enable_chunked_prefill=enable_chunked_prefill,
         max_num_batched_tokens=max_num_batched_tokens,
         distributed_executor_backend=distributed_executor_backend,
+        load_format=load_format,
     )
 
     # Add the requests to the engine.
@@ -228,7 +231,7 @@ def main(args: argparse.Namespace):
             args.quantization_param_path, args.device,
             args.enable_prefix_caching, args.enable_chunked_prefill,
             args.max_num_batched_tokens, args.distributed_executor_backend,
-            args.gpu_memory_utilization, args.download_dir)
+            args.gpu_memory_utilization, args.download_dir, args.load_format)
     elif args.backend == "hf":
         assert args.tensor_parallel_size == 1
         elapsed_time = run_hf(requests, args.model, tokenizer, args.n,
@@ -377,6 +380,29 @@ def main(args: argparse.Namespace):
         help='Backend to use for distributed serving. When more than 1 GPU '
         'is used, will be automatically set to "ray" if installed '
         'or "mp" (multiprocessing) otherwise.')
+    parser.add_argument(
+        '--load-format',
+        type=str,
+        default=EngineArgs.load_format,
+        choices=[
+            'auto', 'pt', 'safetensors', 'npcache', 'dummy', 'tensorizer',
+            'bitsandbytes'
+        ],
+        help='The format of the model weights to load.\n\n'
+        '* "auto" will try to load the weights in the safetensors format '
+        'and fall back to the pytorch bin format if safetensors format '
+        'is not available.\n'
+        '* "pt" will load the weights in the pytorch bin format.\n'
+        '* "safetensors" will load the weights in the safetensors format.\n'
+        '* "npcache" will load the weights in pytorch format and store '
+        'a numpy cache to speed up the loading.\n'
+        '* "dummy" will initialize the weights with random values, '
+        'which is mainly for profiling.\n'
+        '* "tensorizer" will load the weights using tensorizer from '
+        'CoreWeave. See the Tensorize vLLM Model script in the Examples'
+        'section for more information.\n'
+        '* "bitsandbytes" will load the weights using bitsandbytes '
+        'quantization.\n')
     args = parser.parse_args()
     if args.tokenizer is None:
         args.tokenizer = args.model