[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
@@ -25,7 +25,7 @@ from typing import Literal
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
@@ -122,7 +122,7 @@ class HrmTextAttention(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
layer_idx_in_stack: int,
stack_kind: Literal["L", "H"],
cache_config: CacheConfig | None = None,
@@ -245,7 +245,7 @@ class HrmTextAttention(nn.Module):
class HrmTextDecoderLayer(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
layer_idx_in_stack: int,
stack_kind: Literal["L", "H"],
cache_config: CacheConfig | None = None,
@@ -310,7 +310,7 @@ class HrmTextStack(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: PreTrainedConfig,
stack_kind: Literal["L", "H"],
cache_config: CacheConfig | None = None,
quant_config: QuantizationConfig | None = None,