[Doc] Sync CLI guide with actual help modes and launch subcommand (#40326)

Signed-off-by: Rui Wang <raygorous@gmail.com>
Co-authored-by: Rui Wang <raygorous@gmail.com>

docs/cli/.nav.yml

@@ -6,3 +6,5 @@ nav:
   - run-batch.md
   - vllm bench:
     - bench/**/*.md
+  - vllm launch:
+    - launch/**/*.md

docs/cli/README.md

@@ -9,7 +9,7 @@ vllm --help
 Available Commands:
 
 ```bash
-vllm {chat,complete,serve,bench,collect-env,run-batch}
+vllm {chat,complete,serve,launch,bench,collect-env,run-batch}
 ```
 
 ## serve
@@ -37,24 +37,36 @@ vllm serve meta-llama/Llama-2-7b-hf --uds /tmp/vllm.sock
 Check with --help for more options:
 
 ```bash
-# To list all groups
-vllm serve --help=listgroup
+# To list all flags
+vllm serve --help=all
 
-# To view a argument group
+# To view an argument group
 vllm serve --help=ModelConfig
 
 # To view a single argument
 vllm serve --help=max-num-seqs
 
-# To search by keyword
+# To search by keyword or flag name
 vllm serve --help=max
-
-# To view full help with pager (less/more)
-vllm serve --help=page
 ```
 
 See [vllm serve](./serve.md) for the full reference of all available arguments.
 
+## launch
+
+Launch individual vLLM components.
+
+```bash
+# Launch the rendering server component
+vllm launch render meta-llama/Llama-3.2-1B-Instruct
+
+# Inspect all available flags for the render component
+vllm launch render --help=all
+```
+
+See [vllm launch render](./launch/render.md) for the current launch
+component reference.
+
 ## chat
 
 Generate chat completions via the running API server.

docs/cli/launch/render.md

@@ -0,0 +1,22 @@
+# vllm launch render
+
+## Overview
+
+`vllm launch render` starts a GPU-less rendering server for preprocessing and
+postprocessing only.
+
+```bash
+vllm launch render meta-llama/Llama-3.2-1B-Instruct --port 8100
+```
+
+This command reuses the standard serving parser, so model, frontend,
+networking, and related CLI options follow the same conventions as
+[`vllm serve`](../serve.md).
+
+## JSON CLI Arguments
+
+--8<-- "docs/cli/json_tip.inc.md"
+
+## Arguments
+
+--8<-- "docs/generated/argparse/launch_render.inc.md"

docs/mkdocs/hooks/generate_argparse.py

@@ -147,6 +147,7 @@ AsyncEngineArgs = auto_mock("vllm.engine.arg_utils", "AsyncEngineArgs")
 EngineArgs = auto_mock("vllm.engine.arg_utils", "EngineArgs")
 ChatCommand = auto_mock("vllm.entrypoints.cli.openai", "ChatCommand")
 CompleteCommand = auto_mock("vllm.entrypoints.cli.openai", "CompleteCommand")
+RenderSubcommand = auto_mock("vllm.entrypoints.cli.launch", "RenderSubcommand")
 openai_cli_args = auto_mock("vllm.entrypoints.openai", "cli_args")
 openai_run_batch = auto_mock("vllm.entrypoints.openai", "run_batch")
 
@@ -260,6 +261,7 @@ def on_startup(command: Literal["build", "gh-deploy", "serve"], dirty: bool):
         "serve": create_parser(openai_cli_args.make_arg_parser),
         "chat": create_parser(ChatCommand.add_cli_args),
         "complete": create_parser(CompleteCommand.add_cli_args),
+        "launch_render": create_parser(RenderSubcommand.add_cli_args),
         "run-batch": create_parser(openai_run_batch.make_arg_parser),
         # Benchmark CLI
         "bench_latency": create_parser(bench_latency.add_cli_args),