eff3e5fc6f
### What this PR does / why we need it?
1. Refactor the file `mtp_proposer.py`, splits torchair related codes
into `mtp_torchair_proposer.py`
2. According to https://github.com/vllm-project/vllm/pull/24539,
implements padded speculative decoding as described in
https://github.com/vllm-project/vllm/issues/21984.
### Does this PR introduce _any_ user-facing change?
User can use `disable_padded_drafter_batch` to disable/enable padded
speculation, default is `False`.
offline example:
```
speculative_config={"method": "deepseek_mtp", "num_speculative_tokens": 1, "disable_padded_drafter_batch": False}
```
### How was this patch tested?
- [x] egaer with pad/unpad:
- [x] aclgraph with pad/unpad
- [x] torchair with pad/unpad
performance test of deepseek-r1 with tp16、dp1
aclgraph with pad ITL: 168ms
aclgraph with unpad ITL: 169ms
original: 178ms
- vLLM version: v0.11.0rc3
- vLLM main:
83f478bb19
---------
Signed-off-by: xuyexiong <xuyexiong@huawei.com>
41 lines
1.7 KiB
Python
41 lines
1.7 KiB
Python
#
|
|
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
|
|
# Copyright 2023 The vLLM team.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
# This file is a part of the vllm-ascend project.
|
|
# Adapted from vllm-project/vllm/vllm/worker/gpu_model_runner.py
|
|
#
|
|
from vllm_ascend.spec_decode.eagle_proposer import EagleProposer
|
|
from vllm_ascend.spec_decode.mtp_proposer import MtpProposer
|
|
from vllm_ascend.spec_decode.ngram_proposer import NgramProposer
|
|
from vllm_ascend.torchair.torchair_mtp_proposer import TorchairMtpProposer
|
|
|
|
|
|
def get_spec_decode_method(method,
|
|
vllm_config,
|
|
device,
|
|
runner,
|
|
is_torchair_graph=False):
|
|
if method == "ngram":
|
|
return NgramProposer(vllm_config, device, runner)
|
|
elif method in ["eagle", "eagle3"]:
|
|
return EagleProposer(vllm_config, device, runner)
|
|
elif method == 'deepseek_mtp':
|
|
if is_torchair_graph:
|
|
return TorchairMtpProposer(vllm_config, device, runner)
|
|
return MtpProposer(vllm_config, device, runner)
|
|
else:
|
|
raise ValueError("Unknown speculative decoding method: "
|
|
f"{method}")
|