Files
xc-llm-ascend/.github/workflows/vllm_ascend_test_nightly_a3.yaml
wangxiyuan a1f142b7ad Drop 0.11.0 support (#4377)
There is a lot hack code for v0.11.0, which makes the code hard to
upgrade to newer vLLM version. Since v0.11.0 will release soon. Let's
drop v0.11.0 support first. Then we'll upgrade to v0.11.2 soon.


- vLLM version: v0.11.0
- vLLM main:
2918c1b49c

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2025-11-24 17:08:20 +08:00

144 lines
5.6 KiB
YAML

#
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
#
# 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.
#
# This workflow related to the resources atlas 800 A3
# **Please note**: current A3 resource pool's maximum allowed concurrency is 5*16 NPUs
# We will limit the concurrency of jobs on A3 to avoid the risk of insufficient resources
name: 'ascend test / nightly-a3'
on:
schedule:
# Run test at 24:00 Beijing time (UTC+8)
- cron: "0 16 * * *"
workflow_dispatch:
pull_request:
branches:
- 'main'
# Bash shells do not use ~/.profile or ~/.bashrc so these shells need to be explicitly
# declared as "shell: bash -el {0}" on steps that need to be properly activated.
# It's used to activate ascend-toolkit environment variables.
defaults:
run:
shell: bash -el {0}
concurrency:
group: ascend-nightly-${{ github.ref }}-a3
cancel-in-progress: true
jobs:
multi-node-tests:
name: multi-node
if: always() && (github.event_name == 'schedule' || github.event_name == 'workflow_dispatch')
strategy:
fail-fast: false
max-parallel: 1
matrix:
test_config:
- name: multi-node-deepseek-pd
config_file_path: DeepSeek-V3.yaml
size: 2
- name: multi-node-qwen3-dp
config_file_path: Qwen3-235B-A3B.yaml
size: 2
- name: multi-node-dpsk-4node-pd
config_file_path: DeepSeek-R1-W8A8.yaml
size: 4
- name: multi-node-qwenw8a8-2node
config_file_path: Qwen3-235B-W8A8.yaml
size: 2
- name: multi-node-glm-2node
config_file_path: GLM-4_5.yaml
size: 2
- name: multi-node-dpsk3.2-exp-2node
config_file_path: DeepSeek-V3_2-Exp-bf16.yaml
size: 2
- name: multi-node-deepseek-r1-w8a8-eplb
config_file_path: DeepSeek-R1-W8A8-EPLB.yaml
size: 4
- name: multi-node-qwenw8a8-2node-eplb
config_file_path: Qwen3-235B-W8A8-EPLB.yaml
size: 2
uses: ./.github/workflows/_e2e_nightly_multi_node.yaml
with:
soc_version: a3
runner: linux-aarch64-a3-0
image: 'swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/vllm-ascend:nightly-a3'
replicas: 1
size: ${{ matrix.test_config.size }}
config_file_path: ${{ matrix.test_config.config_file_path }}
secrets:
KUBECONFIG_B64: ${{ secrets.KUBECONFIG_B64 }}
single-node-tests:
name: single-node
if: always() && (github.event_name == 'schedule' || github.event_name == 'workflow_dispatch')
needs: multi-node-tests
strategy:
fail-fast: false
matrix:
test_config:
- name: qwen3-32b-in8-a3
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/models/test_qwen3_32b_int8.py
- name: qwen3-32b-int8-a3-feature-stack3
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/features/test_qwen3_32b_int8_a3_feature_stack3.py
- name: qwen3-235b-a22b-w8a8-eplb
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/models/test_qwen3_235b_a22b_w8a8_eplb.py
- name: deepseek-r1-w8a8-eplb
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/models/test_deepseek_r1_w8a8_eplb.py
- name: deepseek-r1-w8a8-mtpx
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/features/test_mtpx_deepseek_r1_0528_w8a8.py
- name: qwen2-5-vl-7b
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/models/test_qwen2_5_vl_7b.py
- name: qwen2-5-vl-32b
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/models/test_qwen2_5_vl_32b.py
- name: qwen3-32b-int8-prefix-cache
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/features/test_prefix_cache_qwen3_32b_int8.py
- name: deepseek-r1-0528-w8a8
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/models/test_deepseek_r1_0528_w8a8.py
- name: deepseek-r1-0528-w8a8-prefix-cache
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/features/test_prefix_cache_deepseek_r1_0528_w8a8.py
- name: qwq-32b-a3
os: linux-aarch64-a3-4
tests: tests/e2e/nightly/models/test_qwq_32b.py
- name: qwen3-30b-w8a8
os: linux-aarch64-a3-2
tests: tests/e2e/nightly/models/test_qwen3_30b_w8a8.py
- name: qwen3-235b-w8a8
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/models/test_qwen3_235b_w8a8.py
- name: deepseek3_2-exp-w8a8
os: linux-aarch64-a3-16
tests: tests/e2e/nightly/models/test_deepseek_v3_2_exp_w8a8.py
uses: ./.github/workflows/_e2e_nightly_single_node.yaml
with:
vllm: 2918c1b49c88c29783c86f78d2c4221cb9622379
runner: ${{ matrix.test_config.os }}
image: 'swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/vllm-ascend:nightly-a3'
tests: ${{ matrix.test_config.tests }}
name: ${{ matrix.test_config.name }}