main
py 25 lines 966 Bytes
Raw
1 # SPDX-License-Identifier: Apache-2.0
2 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
3 # SPDX-FileCopyrightText: Songlin Yang, Yu Zhang
4 #
5 # This file contains code copied from the flash-linear-attention project.
6 # The original source code was licensed under the MIT license and included
7 # the following copyright notice:
8 # Copyright (c) 2023-2025, Songlin Yang, Yu Zhang
9 from .chunk import chunk_gated_delta_rule
10 from .fused_gdn_prefill_post_conv import fused_post_conv_prep
11 from .fused_recurrent import (
12 fused_recurrent_gated_delta_rule,
13 fused_recurrent_gated_delta_rule_packed_decode,
14 )
15 from .fused_sigmoid_gating import fused_sigmoid_gating_delta_rule_update
16 from .layernorm_guard import RMSNormGated
17
18 __all__ = [
19 "RMSNormGated",
20 "chunk_gated_delta_rule",
21 "fused_recurrent_gated_delta_rule",
22 "fused_recurrent_gated_delta_rule_packed_decode",
23 "fused_post_conv_prep",
24 "fused_sigmoid_gating_delta_rule_update",
25 ]