738bf2b720
### What this PR does / why we need it?
support qwen3-next full_decode_only mode.
bs=1, max_token=1024
| branch| tps| e2e time|
| --- | --- | --- |
|piecewise |3.06 | 8.15 |
|fulldecodeonly | 7.2 | 3.47 |
- vLLM version: v0.11.0
- vLLM main:
83f478bb19
Signed-off-by: wangxiaoxin-sherie <wangxiaoxin7@huawei.com>
Co-authored-by: wangxiaoxin-sherie <wangxiaoxin7@huawei.com>
57 lines
2.1 KiB
Python
57 lines
2.1 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/tests/basic_correctness/test_basic_correctness.py
|
|
#
|
|
"""Compare the short outputs of HF and vLLM when using greedy sampling.
|
|
|
|
Run `pytest tests/e2e/multicard/test_qwen3_next.py`.
|
|
"""
|
|
|
|
from tests.e2e.conftest import VllmRunner
|
|
|
|
|
|
def test_models_distributed_Qwen3_NEXT_TP4():
|
|
example_prompts = [
|
|
"Hello, my name is",
|
|
] * 4
|
|
max_tokens = 5
|
|
with VllmRunner("Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
tensor_parallel_size=4,
|
|
max_model_len=4096,
|
|
gpu_memory_utilization=0.8,
|
|
distributed_executor_backend="mp",
|
|
enforce_eager=True) as vllm_model:
|
|
vllm_model.generate_greedy(example_prompts, max_tokens)
|
|
|
|
|
|
def test_models_distributed_Qwen3_NEXT_TP4_FULL_DECODE_ONLY():
|
|
example_prompts = [
|
|
"Hello, my name is",
|
|
] * 4
|
|
max_tokens = 5
|
|
with VllmRunner("Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
tensor_parallel_size=4,
|
|
max_model_len=4096,
|
|
gpu_memory_utilization=0.8,
|
|
distributed_executor_backend="mp",
|
|
enforce_eager=False,
|
|
compilation_config={
|
|
"cudagraph_mode": "FULL_DECODE_ONLY",
|
|
"cudagraph_capture_sizes": [1, 8, 24, 48, 60]
|
|
}) as vllm_model:
|
|
vllm_model.generate_greedy(example_prompts, max_tokens)
|