[Chore] Use Transformers v5 names and drop redundant processor use_fast (#58550)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Harry Mellor
2026-09-24 14:14:10 +00:00
committed by GitHub
co-authored by Claude Opus 5.5
parent 484c211fe8
commit d881812500
218 changed files with 869 additions and 813 deletions
+3 -3
View File
@@ -20,7 +20,7 @@ from itertools import islice
import torch
from torch import nn
from transformers import PretrainedConfig
from transformers import PreTrainedConfig
from vllm.compilation.decorators import support_torch_compile
from vllm.config import CacheConfig, VllmConfig, get_current_vllm_config
@@ -65,7 +65,7 @@ from .utils import (
class ExaoneMoe(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
swiglu_limit: float | None = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",
@@ -289,7 +289,7 @@ class ExaoneMoeAttention(nn.Module):
class ExaoneMoeDecoderLayer(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,
is_mtp: bool = False,