[Doc][CI/Build] Update docs and tests to use `vllm serve` (#6431)

docs/source/getting_started/quickstart.rst

@@ -73,16 +73,13 @@ Start the server:
 
 .. code-block:: console
 
-    $ python -m vllm.entrypoints.openai.api_server \
-    $     --model facebook/opt-125m
+    $ vllm serve facebook/opt-125m
 
 By default, the server uses a predefined chat template stored in the tokenizer. You can override this template by using the ``--chat-template`` argument:
 
 .. code-block:: console
 
-   $ python -m vllm.entrypoints.openai.api_server \
-   $     --model facebook/opt-125m \
-   $     --chat-template ./examples/template_chatml.jinja
+    $ vllm serve facebook/opt-125m --chat-template ./examples/template_chatml.jinja
 
 This server can be queried in the same format as OpenAI API. For example, list the models:
 

docs/source/models/adding_model.rst

@@ -114,7 +114,7 @@ Just add the following lines in your code:
     from your_code import YourModelForCausalLM
     ModelRegistry.register_model("YourModelForCausalLM", YourModelForCausalLM)
 
-If you are running api server with `python -m vllm.entrypoints.openai.api_server args`, you can wrap the entrypoint with the following code:
+If you are running api server with :code:`vllm serve <args>`, you can wrap the entrypoint with the following code:
 
 .. code-block:: python
 
@@ -124,4 +124,4 @@ If you are running api server with `python -m vllm.entrypoints.openai.api_server
     import runpy
     runpy.run_module('vllm.entrypoints.openai.api_server', run_name='__main__')
 
-Save the above code in a file and run it with `python your_file.py args`.
+Save the above code in a file and run it with :code:`python your_file.py <args>`.

docs/source/models/engine_args.rst

@@ -8,7 +8,7 @@ Below, you can find an explanation of every engine argument for vLLM:
 .. argparse::
     :module: vllm.engine.arg_utils
     :func: _engine_args_parser
-    :prog: -m vllm.entrypoints.openai.api_server
+    :prog: vllm serve
     :nodefaultconst:
 
 Async Engine Arguments
@@ -19,5 +19,5 @@ Below are the additional arguments related to the asynchronous engine:
 .. argparse::
     :module: vllm.engine.arg_utils
     :func: _async_engine_args_parser
-    :prog: -m vllm.entrypoints.openai.api_server
+    :prog: vllm serve
     :nodefaultconst:
\ No newline at end of file

docs/source/models/lora.rst

@@ -61,8 +61,7 @@ LoRA adapted models can also be served with the Open-AI compatible vLLM server.
 
 .. code-block:: bash
 
-    python -m vllm.entrypoints.openai.api_server \
-        --model meta-llama/Llama-2-7b-hf \
+    vllm serve meta-llama/Llama-2-7b-hf \
         --enable-lora \
         --lora-modules sql-lora=$HOME/.cache/huggingface/hub/models--yard1--llama-2-7b-sql-lora-test/snapshots/0dfa347e8877a4d4ed19ee56c140fa518470028c/
 

docs/source/models/vlm.rst

@@ -94,9 +94,7 @@ Below is an example on how to launch the same ``llava-hf/llava-1.5-7b-hf`` with
 
 .. code-block:: bash
 
-    python -m vllm.entrypoints.openai.api_server \
-        --model llava-hf/llava-1.5-7b-hf \
-        --chat-template template_llava.jinja
+    vllm serve llava-hf/llava-1.5-7b-hf --chat-template template_llava.jinja
 
 .. important::
     We have removed all vision language related CLI args in the ``0.5.1`` release. **This is a breaking change**, so please update your code to follow

docs/source/serving/deploying_with_dstack.rst

@@ -40,7 +40,7 @@ Next, to provision a VM instance with LLM of your choice(`NousResearch/Llama-2-7
         gpu: 24GB
     commands:
         - pip install vllm
-        - python -m vllm.entrypoints.openai.api_server --model $MODEL --port 8000
+        - vllm serve $MODEL --port 8000
     model:
         format: openai
         type: chat

docs/source/serving/distributed_serving.rst

@@ -35,16 +35,14 @@ To run multi-GPU serving, pass in the :code:`--tensor-parallel-size` argument wh
 
 .. code-block:: console
 
-    $ python -m vllm.entrypoints.openai.api_server \
-    $     --model facebook/opt-13b \
+    $ vllm serve facebook/opt-13b \
     $     --tensor-parallel-size 4
 
 You can also additionally specify :code:`--pipeline-parallel-size` to enable pipeline parallelism. For example, to run API server on 8 GPUs with pipeline parallelism and tensor parallelism:
 
 .. code-block:: console
 
-    $ python -m vllm.entrypoints.openai.api_server \
-    $     --model gpt2 \
+    $ vllm serve gpt2 \
     $     --tensor-parallel-size 4 \
     $     --pipeline-parallel-size 2 \
     $     --distributed-executor-backend ray

docs/source/serving/openai_compatible_server.md

@@ -4,7 +4,7 @@ vLLM provides an HTTP server that implements OpenAI's [Completions](https://plat
 
 You can start the server using Python, or using [Docker](deploying_with_docker.rst):
 ```bash
-python -m vllm.entrypoints.openai.api_server --model NousResearch/Meta-Llama-3-8B-Instruct --dtype auto --api-key token-abc123
+vllm serve NousResearch/Meta-Llama-3-8B-Instruct --dtype auto --api-key token-abc123
 ```
 
 To call the server, you can use the official OpenAI Python client library, or any other HTTP client.
@@ -97,9 +97,7 @@ template, or the template in string form. Without a chat template, the server wi
 and all chat requests will error.
 
 ```bash
-python -m vllm.entrypoints.openai.api_server \
-  --model ... \
-  --chat-template ./path-to-chat-template.jinja
+vllm serve <model> --chat-template ./path-to-chat-template.jinja
 ```
 
 vLLM community provides a set of chat templates for popular models. You can find them in the examples
@@ -110,7 +108,7 @@ directory [here](https://github.com/vllm-project/vllm/tree/main/examples/)
 ```{argparse}
 :module: vllm.entrypoints.openai.cli_args
 :func: create_parser_for_docs
-:prog: -m vllm.entrypoints.openai.api_server
+:prog: vllm serve
 ```
 
 ## Tool calling in the chat completion API

examples/api_client.py

-"""Example Python client for vllm.entrypoints.api_server
+"""Example Python client for `vllm.entrypoints.api_server`
 NOTE: The API server is used only for demonstration and simple performance
 benchmarks. It is not intended for production use.
-For production use, we recommend vllm.entrypoints.openai.api_server
-and the OpenAI client API
+For production use, we recommend `vllm serve` and the OpenAI client API.
 """
 
 import argparse

examples/logging_configuration.md

@@ -95,9 +95,7 @@ to the path of the custom logging configuration JSON file:
 
 ```bash
 VLLM_LOGGING_CONFIG_PATH=/path/to/logging_config.json \
-    python3 -m vllm.entrypoints.openai.api_server \
-    --max-model-len 2048 \
-    --model mistralai/Mistral-7B-v0.1
+    vllm serve mistralai/Mistral-7B-v0.1 --max-model-len 2048
 ```
 
 
@@ -152,9 +150,7 @@ to the path of the custom logging configuration JSON file:
 
 ```bash
 VLLM_LOGGING_CONFIG_PATH=/path/to/logging_config.json \
-    python3 -m vllm.entrypoints.openai.api_server \
-    --max-model-len 2048 \
-    --model mistralai/Mistral-7B-v0.1
+    vllm serve mistralai/Mistral-7B-v0.1 --max-model-len 2048
 ```
 
 
@@ -167,9 +163,7 @@ loggers.
 
 ```bash
 VLLM_CONFIGURE_LOGGING=0 \
-    python3 -m vllm.entrypoints.openai.api_server \
-    --max-model-len 2048 \
-    --model mistralai/Mistral-7B-v0.1
+    vllm serve mistralai/Mistral-7B-v0.1 --max-model-len 2048
 ```
 
 

examples/openai_vision_api_client.py

 """An example showing how to use vLLM to serve VLMs.
 
 Launch the vLLM server with the following command:
-python -m vllm.entrypoints.openai.api_server \
-    --model llava-hf/llava-1.5-7b-hf \
-    --chat-template template_llava.jinja
+vllm serve llava-hf/llava-1.5-7b-hf --chat-template template_llava.jinja
 """
 import base64
 

examples/production_monitoring/Otel.md

@@ -36,7 +36,7 @@
     ```
     export OTEL_SERVICE_NAME="vllm-server"
     export OTEL_EXPORTER_OTLP_TRACES_INSECURE=true
-    python -m vllm.entrypoints.openai.api_server --model="facebook/opt-125m" --otlp-traces-endpoint="$OTEL_EXPORTER_OTLP_TRACES_ENDPOINT"
+    vllm serve facebook/opt-125m --otlp-traces-endpoint="$OTEL_EXPORTER_OTLP_TRACES_ENDPOINT"
     ```
 
 1. In a new shell, send requests with trace context from a dummy client
@@ -62,7 +62,7 @@ By default, `grpc` is used. To set `http/protobuf` as the protocol, configure th
 ```
 export OTEL_EXPORTER_OTLP_TRACES_PROTOCOL=http/protobuf
 export OTEL_EXPORTER_OTLP_TRACES_ENDPOINT=http://$JAEGER_IP:4318/v1/traces
-python -m vllm.entrypoints.openai.api_server --model="facebook/opt-125m" --otlp-traces-endpoint="$OTEL_EXPORTER_OTLP_TRACES_ENDPOINT"
+vllm serve facebook/opt-125m --otlp-traces-endpoint="$OTEL_EXPORTER_OTLP_TRACES_ENDPOINT"
 ```
 
 ## Instrumentation of FastAPI
@@ -74,7 +74,7 @@ OpenTelemetry allows automatic instrumentation of FastAPI.
 
 1. Run vLLM with `opentelemetry-instrument`
     ```
-    opentelemetry-instrument python -m vllm.entrypoints.openai.api_server --model="facebook/opt-125m" 
+    opentelemetry-instrument vllm serve facebook/opt-125m
     ```
 
 1. Send a request to vLLM and find its trace in Jaeger. It should contain spans from FastAPI.

examples/production_monitoring/README.md

@@ -10,8 +10,7 @@ Install:
 
 Prometheus metric logging is enabled by default in the OpenAI-compatible server. Launch via the entrypoint:
 ```bash
-python3 -m vllm.entrypoints.openai.api_server \
-    --model mistralai/Mistral-7B-v0.1 \
+vllm serve mistralai/Mistral-7B-v0.1 \
     --max-model-len 2048 \
     --disable-log-requests
 ```

tests/async_engine/test_openapi_server_ray.py

@@ -9,17 +9,17 @@ MODEL_NAME = "facebook/opt-125m"
 
 @pytest.fixture(scope="module")
 def server():
-    with RemoteOpenAIServer([
-            "--model",
-            MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "float16",
-            "--max-model-len",
-            "2048",
-            "--enforce-eager",
-            "--engine-use-ray"
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "float16",
+        "--max-model-len",
+        "2048",
+        "--enforce-eager",
+        "--engine-use-ray"
+    ]
+
+    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server
 
 

tests/distributed/test_pipeline_parallel.py

@@ -15,8 +15,6 @@ from ..utils import RemoteOpenAIServer
     ])
 def test_compare_tp(TP_SIZE, PP_SIZE, EAGER_MODE, CHUNKED_PREFILL, MODEL_NAME):
     pp_args = [
-        "--model",
-        MODEL_NAME,
         # use half precision for speed and memory savings in CI environment
         "--dtype",
         "bfloat16",
@@ -34,8 +32,6 @@ def test_compare_tp(TP_SIZE, PP_SIZE, EAGER_MODE, CHUNKED_PREFILL, MODEL_NAME):
     #  schedule all workers in a node other than the head node,
     #  which can cause the test to fail.
     tp_args = [
-        "--model",
-        MODEL_NAME,
         # use half precision for speed and memory savings in CI environment
         "--dtype",
         "bfloat16",
@@ -53,7 +49,7 @@ def test_compare_tp(TP_SIZE, PP_SIZE, EAGER_MODE, CHUNKED_PREFILL, MODEL_NAME):
 
     results = []
     for args in [pp_args, tp_args]:
-        with RemoteOpenAIServer(args) as server:
+        with RemoteOpenAIServer(MODEL_NAME, args) as server:
             client = server.get_client()
 
             # test models list

tests/entrypoints/openai/test_chat.py

@@ -27,27 +27,27 @@ def zephyr_lora_files():
 
 @pytest.fixture(scope="module")
 def server(zephyr_lora_files):
-    with RemoteOpenAIServer([
-            "--model",
-            MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "bfloat16",
-            "--max-model-len",
-            "8192",
-            "--enforce-eager",
-            # lora config below
-            "--enable-lora",
-            "--lora-modules",
-            f"zephyr-lora={zephyr_lora_files}",
-            f"zephyr-lora2={zephyr_lora_files}",
-            "--max-lora-rank",
-            "64",
-            "--max-cpu-loras",
-            "2",
-            "--max-num-seqs",
-            "128",
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "bfloat16",
+        "--max-model-len",
+        "8192",
+        "--enforce-eager",
+        # lora config below
+        "--enable-lora",
+        "--lora-modules",
+        f"zephyr-lora={zephyr_lora_files}",
+        f"zephyr-lora2={zephyr_lora_files}",
+        "--max-lora-rank",
+        "64",
+        "--max-cpu-loras",
+        "2",
+        "--max-num-seqs",
+        "128",
+    ]
+
+    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server
 
 

tests/entrypoints/openai/test_completion.py

@@ -37,36 +37,36 @@ def zephyr_pa_files():
 
 @pytest.fixture(scope="module")
 def server(zephyr_lora_files, zephyr_pa_files):
-    with RemoteOpenAIServer([
-            "--model",
-            MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "bfloat16",
-            "--max-model-len",
-            "8192",
-            "--max-num-seqs",
-            "128",
-            "--enforce-eager",
-            # lora config
-            "--enable-lora",
-            "--lora-modules",
-            f"zephyr-lora={zephyr_lora_files}",
-            f"zephyr-lora2={zephyr_lora_files}",
-            "--max-lora-rank",
-            "64",
-            "--max-cpu-loras",
-            "2",
-            # pa config
-            "--enable-prompt-adapter",
-            "--prompt-adapters",
-            f"zephyr-pa={zephyr_pa_files}",
-            f"zephyr-pa2={zephyr_pa_files}",
-            "--max-prompt-adapters",
-            "2",
-            "--max-prompt-adapter-token",
-            "128",
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "bfloat16",
+        "--max-model-len",
+        "8192",
+        "--max-num-seqs",
+        "128",
+        "--enforce-eager",
+        # lora config
+        "--enable-lora",
+        "--lora-modules",
+        f"zephyr-lora={zephyr_lora_files}",
+        f"zephyr-lora2={zephyr_lora_files}",
+        "--max-lora-rank",
+        "64",
+        "--max-cpu-loras",
+        "2",
+        # pa config
+        "--enable-prompt-adapter",
+        "--prompt-adapters",
+        f"zephyr-pa={zephyr_pa_files}",
+        f"zephyr-pa2={zephyr_pa_files}",
+        "--max-prompt-adapters",
+        "2",
+        "--max-prompt-adapter-token",
+        "128",
+    ]
+
+    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server
 
 

tests/entrypoints/openai/test_embedding.py

@@ -11,17 +11,17 @@ EMBEDDING_MODEL_NAME = "intfloat/e5-mistral-7b-instruct"
 
 @pytest.fixture(scope="module")
 def embedding_server():
-    with RemoteOpenAIServer([
-            "--model",
-            EMBEDDING_MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "bfloat16",
-            "--enforce-eager",
-            "--max-model-len",
-            "8192",
-            "--enforce-eager",
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "bfloat16",
+        "--enforce-eager",
+        "--max-model-len",
+        "8192",
+        "--enforce-eager",
+    ]
+
+    with RemoteOpenAIServer(EMBEDDING_MODEL_NAME, args) as remote_server:
         yield remote_server
 
 

tests/entrypoints/openai/test_models.py

@@ -19,27 +19,27 @@ def zephyr_lora_files():
 
 @pytest.fixture(scope="module")
 def server(zephyr_lora_files):
-    with RemoteOpenAIServer([
-            "--model",
-            MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "bfloat16",
-            "--max-model-len",
-            "8192",
-            "--enforce-eager",
-            # lora config below
-            "--enable-lora",
-            "--lora-modules",
-            f"zephyr-lora={zephyr_lora_files}",
-            f"zephyr-lora2={zephyr_lora_files}",
-            "--max-lora-rank",
-            "64",
-            "--max-cpu-loras",
-            "2",
-            "--max-num-seqs",
-            "128",
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "bfloat16",
+        "--max-model-len",
+        "8192",
+        "--enforce-eager",
+        # lora config below
+        "--enable-lora",
+        "--lora-modules",
+        f"zephyr-lora={zephyr_lora_files}",
+        f"zephyr-lora2={zephyr_lora_files}",
+        "--max-lora-rank",
+        "64",
+        "--max-cpu-loras",
+        "2",
+        "--max-num-seqs",
+        "128",
+    ]
+
+    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server
 
 

tests/entrypoints/openai/test_tokenization.py

@@ -12,18 +12,18 @@ MODEL_NAME = "HuggingFaceH4/zephyr-7b-beta"
 
 @pytest.fixture(scope="module")
 def server():
-    with RemoteOpenAIServer([
-            "--model",
-            MODEL_NAME,
-            # use half precision for speed and memory savings in CI environment
-            "--dtype",
-            "bfloat16",
-            "--max-model-len",
-            "8192",
-            "--enforce-eager",
-            "--max-num-seqs",
-            "128",
-    ]) as remote_server:
+    args = [
+        # use half precision for speed and memory savings in CI environment
+        "--dtype",
+        "bfloat16",
+        "--max-model-len",
+        "8192",
+        "--enforce-eager",
+        "--max-num-seqs",
+        "128",
+    ]
+
+    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server