frozenleaves/vllm
1
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2025-10-20 14:53:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add deprecation warning for lora_extra_vocab_size (#23635)

Browse Source 
Signed-off-by: Jinheng Li <ahengljh@gmail.com>
This commit is contained in:
Jinheng
2025-08-28 13:34:29 +08:00
committed by GitHub
parent f48a9af892
commit c8851a4723
1 changed files with 8 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/config/__init__.py
View File

@ -2439,8 +2439,8 @@ class LoRAConfig:
lora_dtype: Union[torch.dtype, LoRADType] = "auto"
"""Data type for LoRA. If auto, will default to base model dtype."""
lora_extra_vocab_size: int = 256
"""Maximum size of extra vocabulary that can be present in a LoRA adapter
(added to the base model vocabulary)."""
"""(Deprecated) Maximum size of extra vocabulary that can be present in a
LoRA adapter. Will be removed in v0.12.0."""
lora_vocab_padding_size: ClassVar[int] = current_platform\
.get_lora_vocab_padding_size()
@ -2482,6 +2482,12 @@ class LoRAConfig:
return hash_str
def __post_init__(self):
# Deprecation warning for lora_extra_vocab_size
logger.warning(
"`lora_extra_vocab_size` is deprecated and will be removed "
"in v0.12.0. Additional vocabulary support for "
"LoRA adapters is being phased out.")
# Setting the maximum rank to 512 should be able to satisfy the vast
# majority of applications.
possible_max_ranks = (8, 16, 32, 64, 128, 256, 320, 512)