[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
+4 -4
View File
@@ -30,7 +30,7 @@ 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
@@ -104,7 +104,7 @@ class ExaoneGatedMLP(nn.Module):
class ExaoneAttention(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
hidden_size: int,
num_heads: int,
num_kv_heads: int,
@@ -191,7 +191,7 @@ class ExaoneAttention(nn.Module):
class ExaoneBlockAttention(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
hidden_size: int,
num_heads: int,
num_kv_heads: int,
@@ -228,7 +228,7 @@ class ExaoneBlockAttention(nn.Module):
class ExaoneDecoderLayer(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,
prefix: str = "",