xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-07-11 02:07:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

cleaner

Browse Source 
Signed-off-by: Reagan <reaganjlee@gmail.com>
This commit is contained in:
Reagan 2025-12-18 21:32:18 -08:00
parent ae0c59e57e
commit c531de0640
4 changed files with 5 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/mkdocs/hooks/generate_argparse.py
View File

@ -92,7 +92,7 @@ def auto_mock(module_name: str, attr: str, max_mocks: int = 100):
bench_latency = auto_mock("vllm.benchmarks", "latency") bench_latency = auto_mock("vllm.benchmarks", "latency")
bench_multimodal_processor = auto_mock("vllm.benchmarks", "multimodal_processor") bench_mm_processor = auto_mock("vllm.benchmarks", "mm_processor")
bench_serve = auto_mock("vllm.benchmarks", "serve") bench_serve = auto_mock("vllm.benchmarks", "serve")
bench_sweep_plot = auto_mock("vllm.benchmarks.sweep.plot", "SweepPlotArgs") bench_sweep_plot = auto_mock("vllm.benchmarks.sweep.plot", "SweepPlotArgs")
bench_sweep_plot_pareto = auto_mock( bench_sweep_plot_pareto = auto_mock(
@ -223,8 +223,8 @@ def on_startup(command: Literal["build", "gh-deploy", "serve"], dirty: bool):
"run-batch": create_parser(openai_run_batch.make_arg_parser), "run-batch": create_parser(openai_run_batch.make_arg_parser),
# Benchmark CLI # Benchmark CLI
"bench_latency": create_parser(bench_latency.add_cli_args), "bench_latency": create_parser(bench_latency.add_cli_args),
"bench_multimodal_processor": create_parser( "bench_mm_processor": create_parser(
bench_multimodal_processor.add_cli_args bench_mm_processor.add_cli_args
), ),
"bench_serve": create_parser(bench_serve.add_cli_args), "bench_serve": create_parser(bench_serve.add_cli_args),
"bench_sweep_plot": create_parser(bench_sweep_plot.add_cli_args), "bench_sweep_plot": create_parser(bench_sweep_plot.add_cli_args),

2
vllm/config/observability.py
View File

@ -67,7 +67,7 @@ class ObservabilityConfig:
enable_mm_processor_stats: bool = False enable_mm_processor_stats: bool = False
"""Enable collection of timing statistics for multimodal processor operations. """Enable collection of timing statistics for multimodal processor operations.
This is for internal use only (e.g., benchmarks) and is not exposed as a CLI This is for internal use only (e.g., benchmarks) and is not exposed as a CLI
argument. Defaults to `False` (disabled).""" argument."""
@cached_property @cached_property
def collect_model_forward_time(self) -> bool: def collect_model_forward_time(self) -> bool:

3
vllm/engine/arg_utils.py
View File

@ -1041,9 +1041,6 @@ class EngineArgs:
"--enable-layerwise-nvtx-tracing", "--enable-layerwise-nvtx-tracing",
**observability_kwargs["enable_layerwise_nvtx_tracing"], **observability_kwargs["enable_layerwise_nvtx_tracing"],
) )
# Note: --enable-mm-processor-stats is intentionally not exposed as a CLI
# argument. It can be set programmatically via parser.set_defaults() for
# internal use (e.g., benchmarks), but is not part of the public API.
# Scheduler arguments # Scheduler arguments
scheduler_kwargs = get_kwargs(SchedulerConfig) scheduler_kwargs = get_kwargs(SchedulerConfig)

2
vllm/entrypoints/cli/benchmark/mm_processor.py
View File

@ -2,7 +2,7 @@
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import argparse import argparse
from vllm.benchmarks.multimodal_processor import add_cli_args, main from vllm.benchmarks.mm_processor import add_cli_args, main
from vllm.entrypoints.cli.benchmark.base import BenchmarkSubcommandBase from vllm.entrypoints.cli.benchmark.base import BenchmarkSubcommandBase