xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-13 13:55:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Misc] Remove deprecated args in v0.10 (#21349)

Browse Source 
Signed-off-by: Kebe <mail@kebe7jun.com>
This commit is contained in:
Kebe 2025-07-22 20:26:39 +08:00 committed by GitHub
parent 32142b3c62
commit bc8a8ce5ec
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 0 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
examples/offline_inference/neuron_speculation.py
View File

@ -37,7 +37,6 @@ def initialize_llm():
max_num_seqs=4, max_num_seqs=4,
max_model_len=2048, max_model_len=2048,
block_size=2048, block_size=2048,
use_v2_block_manager=True,
device="neuron", device="neuron",
tensor_parallel_size=32, tensor_parallel_size=32,
) )

1
tests/neuron/2_core/test_mistral.py
View File

@ -9,7 +9,6 @@ def test_mistral():
tensor_parallel_size=2, tensor_parallel_size=2,
max_num_seqs=4, max_num_seqs=4,
max_model_len=128, max_model_len=128,
use_v2_block_manager=True,
override_neuron_config={ override_neuron_config={
"sequence_parallel_enabled": False, "sequence_parallel_enabled": False,
"skip_warmup": True "skip_warmup": True

2
tests/neuron/2_core/test_multi_lora.py
View File

@ -14,7 +14,6 @@ def test_llama_single_lora():
tensor_parallel_size=2, tensor_parallel_size=2,
max_num_seqs=4, max_num_seqs=4,
max_model_len=512, max_model_len=512,
use_v2_block_manager=True,
override_neuron_config={ override_neuron_config={
"sequence_parallel_enabled": False, "sequence_parallel_enabled": False,
"skip_warmup": True, "skip_warmup": True,
@ -57,7 +56,6 @@ def test_llama_multiple_lora():
tensor_parallel_size=2, tensor_parallel_size=2,
max_num_seqs=4, max_num_seqs=4,
max_model_len=512, max_model_len=512,
use_v2_block_manager=True,
override_neuron_config={ override_neuron_config={
"sequence_parallel_enabled": "sequence_parallel_enabled":
False, False,

21
vllm/engine/arg_utils.py
View File

@ -313,7 +313,6 @@ class EngineArgs:
CacheConfig.prefix_caching_hash_algo CacheConfig.prefix_caching_hash_algo
disable_sliding_window: bool = ModelConfig.disable_sliding_window disable_sliding_window: bool = ModelConfig.disable_sliding_window
disable_cascade_attn: bool = ModelConfig.disable_cascade_attn disable_cascade_attn: bool = ModelConfig.disable_cascade_attn
use_v2_block_manager: bool = True
swap_space: float = CacheConfig.swap_space swap_space: float = CacheConfig.swap_space
cpu_offload_gb: float = CacheConfig.cpu_offload_gb cpu_offload_gb: float = CacheConfig.cpu_offload_gb
gpu_memory_utilization: float = CacheConfig.gpu_memory_utilization gpu_memory_utilization: float = CacheConfig.gpu_memory_utilization
@ -364,7 +363,6 @@ class EngineArgs:
max_prompt_adapter_token: int = \ max_prompt_adapter_token: int = \
PromptAdapterConfig.max_prompt_adapter_token PromptAdapterConfig.max_prompt_adapter_token
device: Device = DeviceConfig.device
num_scheduler_steps: int = SchedulerConfig.num_scheduler_steps num_scheduler_steps: int = SchedulerConfig.num_scheduler_steps
multi_step_stream_outputs: bool = SchedulerConfig.multi_step_stream_outputs multi_step_stream_outputs: bool = SchedulerConfig.multi_step_stream_outputs
ray_workers_use_nsight: bool = ParallelConfig.ray_workers_use_nsight ray_workers_use_nsight: bool = ParallelConfig.ray_workers_use_nsight
@ -745,16 +743,6 @@ class EngineArgs:
"--max-prompt-adapter-token", "--max-prompt-adapter-token",
**prompt_adapter_kwargs["max_prompt_adapter_token"]) **prompt_adapter_kwargs["max_prompt_adapter_token"])
# Device arguments
device_kwargs = get_kwargs(DeviceConfig)
device_group = parser.add_argument_group(
title="DeviceConfig",
description=DeviceConfig.__doc__,
)
device_group.add_argument("--device",
**device_kwargs["device"],
deprecated=True)
# Speculative arguments # Speculative arguments
speculative_group = parser.add_argument_group( speculative_group = parser.add_argument_group(
title="SpeculativeConfig", title="SpeculativeConfig",
@ -856,15 +844,6 @@ class EngineArgs:
**vllm_kwargs["additional_config"]) **vllm_kwargs["additional_config"])
# Other arguments # Other arguments
parser.add_argument('--use-v2-block-manager',
action='store_true',
default=True,
deprecated=True,
help='[DEPRECATED] block manager v1 has been '
'removed and SelfAttnBlockSpaceManager (i.e. '
'block manager v2) is now the default. '
'Setting this flag to True or False'
' has no effect on vLLM behavior.')
parser.add_argument('--disable-log-stats', parser.add_argument('--disable-log-stats',
action='store_true', action='store_true',
help='Disable logging statistics.') help='Disable logging statistics.')