[Bugfix] Reject tokenizer-less Qwen VL processor init (#54886)

Signed-off-by: Luyi Xiao <81631911+luyixiao95@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Luyi Xiao
2026-09-03 21:25:35 -07:00
committed by GitHub
co-authored by mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent da55494a12
commit 8a728663c1
3 changed files with 14 additions and 0 deletions
+10
View File
@@ -8,6 +8,7 @@ from enum import Enum
from functools import lru_cache
from typing import (
TYPE_CHECKING,
ClassVar,
Generic,
NamedTuple,
Protocol,
@@ -1033,6 +1034,8 @@ class BaseMultiModalProcessor(ABC, Generic[_I]):
Not to be confused with `transformers.ProcessorMixin`.
"""
requires_tokenizer: ClassVar[bool] = True
def __init__(
self,
info: _I,
@@ -1042,6 +1045,13 @@ class BaseMultiModalProcessor(ABC, Generic[_I]):
) -> None:
super().__init__()
if self.requires_tokenizer and info.ctx.tokenizer is None:
raise ValueError(
f"{type(self).__name__} requires a tokenizer and cannot be "
"initialized when `skip_tokenizer_init=True`. Disable "
"`--skip-tokenizer-init`."
)
self.info = info
self.dummy_inputs = dummy_inputs
self.cache = cache