首页
/ Playwright CI 实战指南：在持续集成环境中稳定运行浏览器自动化测试

Playwright CI 实战指南：在持续集成环境中稳定运行浏览器自动化测试

2026-09-06 21:34:07作者：裴麒琰
playwright
Playwright is a framework for Web Testing and Automation. It allows testing Chromium, Firefox and WebKit with a single API.
项目地址：https://gitcode.com/GitHub_Trending/pl/playwright

本文基于 Playwright 官方文档 docs/src/ci.md 编写，系统讲解如何把 Playwright 测试接入各类 CI 平台（GitHub Actions、Azure Pipelines、CircleCI、Jenkins、GitLab CI 等），覆盖浏览器依赖安装、workers 与 globalTimeout 两大关键配置、各平台完整工作流配置、分片（sharding）并行策略，以及浏览器启动故障排查与有头模式运行等实战技巧。读完本文，你可以直接复制可用的 CI 配置，并结合 Playwright 源码理解每个配置项背后的实现机制。

三步走：让 CI Agent 跑起 Playwright 测试

Playwright 测试完全可以在 CI 环境中执行，官方为常见 CI 提供商提供了示例配置。整体流程分为三步：

第 1 步：确保 CI Agent 能运行浏览器。在 Linux agent 上使用官方 Docker 镜像（见 docs/src/docker.md），或使用 Playwright CLI 安装系统依赖（--with-deps，该选项在 CLI 中的定义可参考 packages/playwright-core/src/cli/program.ts）。

第 2 步：安装 Playwright。各语言生态的安装命令如下：

# Install NPM packages
npm ci

# Install Playwright browsers and dependencies
npx playwright install --with-deps

pip install playwright
playwright install --with-deps

mvn exec:java -e -D exec.mainClass=com.microsoft.playwright.CLI -D exec.args="install --with-deps"

dotnet build
pwsh bin/Debug/netX/playwright.ps1 install --with-deps

第 3 步：运行测试：

npx playwright test

pytest

mvn test

dotnet test

关键配置一：CI 环境中的 workers 设置

官方建议在 CI 环境中把 workers 设置为 1，优先保证稳定性与可复现性。顺序执行确保每个测试独占全部系统资源，避免潜在冲突。如果你拥有性能强劲的自托管 CI 系统，也可以启用并行；对于更宽的并行度，考虑分片（sharding）——把测试分发到多台 CI 机器上。

import { defineConfig, devices } from '@playwright/test';

export default defineConfig({
  // Opt out of parallel tests on CI.
  workers: process.env.CI ? 1 : undefined,
});

源码佐证：从 packages/playwright/src/common/config.ts 可以看到，当配置中没有显式指定 workers 时，默认值是 '50%'；而 resolveWorkers 函数 会把百分比解析为「CPU 核数 × 百分比」并至少为 1。这意味着在 CI 容器（如 GitHub Actions 的 2 核 runner）上不设置 workers 时，Playwright 默认就会按核数并行——这正是官方建议显式设为 1 的原因：核心数很少的 CI 机器上盲目并行容易造成资源争抢与偶发超时。

关键配置二：始终设置 globalTimeout

在 CI 中务必设置 global timeout。默认情况下，一次测试运行没有上限时长：一旦某个测试套件挂起、或随着测试规模增长超出了 CI 提供商的 job 时限，运行会在中途被 runner 杀掉，从而无法生成测试报告。

设置了 globalTimeout 后，Playwright 会自己停止整个运行：

import { defineConfig } from '@playwright/test';

export default defineConfig({
  // Fail the run after an hour, so that the reporters still produce a report.
  globalTimeout: 60 * 60 * 1000,
});

源码佐证：配置解析位于 packages/playwright/src/common/config.ts，globalTimeout 的解析优先级为「CLI 覆盖值 → 用户配置值 → 默认值 0」，其中 0 表示不限时——这与文档中「默认没有上限」的说明完全一致。执行侧的 runTasks 函数 只有在 globalTimeout 大于 0 时才会计算 deadline；超时后 TaskRunner 会中断后续任务并向 reporter 上报 Timed out waiting Xs for the ... to run 错误，保证 reporter 有机会收尾输出完整报告。

此外，下文给出的示例均不设置 job 级超时（如 GitHub Actions 的 timeout-minutes）。如果你确需添加，请让它显著高于 globalTimeout，确保总是由 Playwright 先停止。

CI 平台配置：GitHub Actions

Playwright 的命令行工具可以在 CI 中安装所有操作系统依赖。

push/pull_request 触发（JavaScript）

测试将在 main/master 分支的 push 或 pull request 时运行。工作流会安装全部依赖、安装 Playwright 浏览器，然后运行测试，并上传 HTML 报告作为 artifact：

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v6
    - uses: actions/setup-node@v6
      with:
        node-version: lts/*
    - name: Install dependencies
      run: npm ci
    - name: Install Playwright Browsers
      run: npx playwright install --with-deps
    - name: Run Playwright tests
      run: npx playwright test
    - uses: actions/upload-artifact@v5
      if: ${{ !cancelled() }}
      with:
        name: playwright-report
        path: playwright-report/
        retention-days: 30

push/pull_request 触发（Python / Java / .NET）

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v6
    - name: Set up Python
      uses: actions/setup-python@v6
      with:
        python-version: '3.13'
    - name: Install dependencies
      run: |
        python -m pip install --upgrade pip
        pip install -r requirements.txt
    - name: Ensure browsers are installed
      run: python -m playwright install --with-deps
    - name: Run your tests
      run: pytest --tracing=retain-on-failure
    - uses: actions/upload-artifact@v5
      if: ${{ !cancelled() }}
      with:
        name: playwright-traces
        path: test-results/

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v6
    - uses: actions/setup-java@v5
      with:
        distribution: 'temurin'
        java-version: '25'
    - name: Build & Install
      run: mvn -B install -D skipTests --no-transfer-progress
    - name: Ensure browsers are installed
      run: mvn exec:java -e -D exec.mainClass=com.microsoft.playwright.CLI -D exec.args="install --with-deps"
    - name: Run tests
      run: mvn test

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v6
    - name: Setup dotnet
      uses: actions/setup-dotnet@v5
      with:
        dotnet-version: 8.0.x
    - name: Build & Install
      run: dotnet build
    - name: Ensure browsers are installed
      run: pwsh bin/Debug/net8.0/playwright.ps1 install --with-deps
    - name: Run your tests
      run: dotnet test

分片执行（Sharded）

GitHub Actions 支持跨多个 job 分片测试，可参考官方 sharding 文档了解分片原理、GitHub Actions 示例（如何把测试分发到多台机器运行）以及如何合并 HTML 报告。

容器化运行（Via Containers）

通过 jobs.<job_id>.container 选项让 job 运行在容器内，好处是：不污染宿主环境依赖；为截图/视觉回归测试在不同操作系统间提供一致的环境。

JavaScript 版：

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  playwright:
    name: 'Playwright Tests'
    runs-on: ubuntu-latest
    container:
      image: mcr.microsoft.com/playwright:v%%VERSION%%-noble
      options: --user 1001
    steps:
      - uses: actions/checkout@v6
      - uses: actions/setup-node@v6
        with:
          node-version: lts/*
      - name: Install dependencies
        run: npm ci
      - name: Run your tests
        run: npx playwright test

Python / Java / .NET 版本分别使用 mcr.microsoft.com/playwright/python:v%%VERSION%%-noble、mcr.microsoft.com/playwright/java:v%%VERSION%%-noble、mcr.microsoft.com/playwright/dotnet:v%%VERSION%%-noble 镜像，容器内已完成浏览器与系统依赖预装，因此步骤中不需要再执行 install --with-deps，只需安装依赖包并直接运行测试即可（%%VERSION%% 为文档构建时替换为具体 Playwright 版本的占位符）。

部署后触发（On deployment）

以下工作流在 GitHub Deployment 进入 success 状态后启动测试。Vercel 等 PaaS 服务支持这一模式，你可以在它们部署出的环境上运行端到端测试：

name: Playwright Tests
on:
  deployment_status:
jobs:
  test:
    runs-on: ubuntu-latest
    if: github.event.deployment_status.state == 'success'
    steps:
    - uses: actions/checkout@v6
    - uses: actions/setup-node@v6
      with:
        node-version: lts/*
    - name: Install dependencies
      run: npm ci
    - name: Install Playwright
      run: npx playwright install --with-deps
    - name: Run Playwright tests
      run: npx playwright test
      env:
        PLAYWRIGHT_TEST_BASE_URL: ${{ github.event.deployment_status.target_url }}

Python、Java、.NET 版本的写法同理：监听 deployment_status 事件、在 success 时运行，并把 PLAYWRIGHT_TEST_BASE_URL 环境变量指向 ${{ github.event.deployment_status.target_url }}（不同测试 runner 对环境变量名的约定可能不同，需自行适配）。

Fail-Fast：--only-changed 预检

大型测试套件执行耗时很长。通过 --only-changed 参数先做一轮预检，可以优先执行最可能失败的测试文件，在 Pull Request 阶段获得更快的反馈、并略微降低 CI 消耗。--only-changed 通过分析测试套件的依赖图来检测受改动影响的测试文件——这是一种启发式方法，可能漏掉测试，因此预检之后必须始终跑完整测试套件。

name: Playwright Tests
on:
  push:
    branches: [ main, master ]
  pull_request:
    branches: [ main, master ]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v6
      with:
        # Force a non-shallow checkout, so that we can reference $GITHUB_BASE_REF.
        fetch-depth: 0
    - uses: actions/setup-node@v6
      with:
        node-version: lts/*
    - name: Install dependencies
      run: npm ci
    - name: Install Playwright Browsers
      run: npx playwright install --with-deps
    - name: Run changed Playwright tests
      run: npx playwright test --only-changed=origin/$GITHUB_BASE_REF
      if: github.event_name == 'pull_request'
    - name: Run Playwright tests
      run: npx playwright test
    - uses: actions/upload-artifact@v5
      if: ${{ !cancelled() }}
      with:
        name: playwright-report
        path: playwright-report/
        retention-days: 30

源码佐证：CLI 参数定义于 packages/playwright/src/program.ts，其帮助文本说明该参数「只运行 HEAD 与 ref 之间有变更的测试文件，默认对比所有未提交的改动，仅支持 Git」。实际检测逻辑在 packages/playwright/src/runner/vcs.ts 的 detectChangedTestFiles 中：它执行 git diff <base> --name-only 与 git ls-files --others --exclude-standard 收集变更文件，再交给 cc.affectedTestFiles 换算为受影响的测试文件。值得注意的是该函数会显式检测 shallow clone（git rev-parse --is-shallow-repository）——这正是上方 YAML 中必须设置 fetch-depth: 0 的原因，否则浅克隆仓库无法解析基线引用，会直接抛出错误。

CI 平台配置：Docker

官方提供了预构建 Docker 镜像，可直接使用，也可作为参考来改造你自己的 Docker 定义。建议遵循 Recommended Docker Configuration 以获得最佳性能。

从 docs/src/docker.md 可以看到三条关键推荐：

  1. 使用 --init 标志，避免 PID=1 进程的特殊处理（僵尸进程的常见原因）；
  2. 使用 Chromium 时建议使用 --ipc=host，否则 Chromium 可能内存不足而崩溃；
  3. 若本地开发时 Chromium 启动出现奇怪错误，可尝试 docker run --cap-add=SYS_ADMIN。

CI 平台配置：Azure Pipelines

Windows 或 macOS agent 无需额外配置，安装 Playwright 后直接运行测试即可。Linux agent 可选用官方 Docker 容器运行容器化 job，或使用命令行工具安装所有必要依赖。

JavaScript：

trigger:
- main

pool:
  vmImage: ubuntu-latest

steps:
- task: UseNode@1
  inputs:
    version: '22'
  displayName: 'Install Node.js'
- script: npm ci
  displayName: 'npm ci'
- script: npx playwright install --with-deps
  displayName: 'Install Playwright browsers'
- script: npx playwright test
  displayName: 'Run Playwright tests'
  env:
    CI: 'true'

Python：

trigger:
- main

pool:
  vmImage: ubuntu-latest

steps:
- task: UsePythonVersion@0
  inputs:
    versionSpec: '3.13'
  displayName: 'Use Python'
- script: |
    python -m pip install --upgrade pip
    pip install -r requirements.txt
  displayName: 'Install dependencies'
- script: playwright install --with-deps
  displayName: 'Install Playwright browsers'
- script: pytest
  displayName: 'Run Playwright tests'

Java：

trigger:
- main

pool:
  vmImage: ubuntu-latest

steps:
- task: JavaToolInstaller@1
  inputs:
    versionSpec: '25'
    jdkArchitectureOption: 'x64'
    jdkSourceOption: AzureStorage
- script: mvn -B install -D skipTests --no-transfer-progress
  displayName: 'Build and install'
- script: mvn exec:java -e -D exec.mainClass=com.microsoft.playwright.CLI -D exec.args="install --with-deps"
  displayName: 'Install Playwright browsers'
- script: mvn test
  displayName: 'Run tests'

.NET：

trigger:
- main

pool:
  vmImage: ubuntu-latest

steps:
- task: UseDotNet@2
  inputs:
    packageType: sdk
    version: '8.0.x'
  displayName: 'Use .NET SDK'
- script: dotnet build --configuration Release
  displayName: 'Build'
- script: pwsh bin/Release/net8.0/playwright.ps1 install --with-deps
  displayName: 'Install Playwright browsers'
- script: dotnet test --configuration Release
  displayName: 'Run tests'

上传 playwright-report 并集成测试结果

以下配置让 pipeline 在任一 Playwright 测试失败时整体失败；同时通过 PublishTestResults 任务把测试结果集成进 Azure DevOps：

trigger:
- main

pool:
  vmImage: ubuntu-latest

steps:
- task: UseNode@1
  inputs:
    version: '22'
  displayName: 'Install Node.js'

- script: npm ci
  displayName: 'npm ci'
- script: npx playwright install --with-deps
  displayName: 'Install Playwright browsers'
- script: npx playwright test
  displayName: 'Run Playwright tests'
  env:
    CI: 'true'
- task: PublishTestResults@2
  displayName: 'Publish test results'
  inputs:
    searchFolder: 'test-results'
    testResultsFormat: 'JUnit'
    testResultsFiles: 'e2e-junit-results.xml'
    mergeTestResults: true
    failTaskOnFailedTests: true
    testRunTitle: 'My End-To-End Tests'
  condition: succeededOrFailed()
- task: PublishPipelineArtifact@1
  inputs:
    targetPath: playwright-report
    artifact: playwright-report
    publishLocation: 'pipeline'
  condition: succeededOrFailed()

注意：JUnit reporter 需要在 playwright.config.ts 中相应配置：

import { defineConfig } from '@playwright/test';

export default defineConfig({
  reporter: [['junit', { outputFile: 'test-results/e2e-junit-results.xml' }]],
});

分片执行（Azure Pipelines, sharded）

trigger:
- main

pool:
  vmImage: ubuntu-latest

strategy:
  matrix:
    chromium-1:
      project: chromium
      shard: 1/3
    chromium-2:
      project: chromium
      shard: 2/3
    chromium-3:
      project: chromium
      shard: 3/3
    firefox-1:
      project: firefox
      shard: 1/3
    firefox-2:
      project: firefox
      shard: 2/3
    firefox-3:
      project: firefox
      shard: 3/3
    webkit-1:
      project: webkit
      shard: 1/3
    webkit-2:
      project: webkit
      shard: 2/3
    webkit-3:
      project: webkit
      shard: 3/3
steps:
- task: UseNode@1
  inputs:
    version: '22'
  displayName: 'Install Node.js'

- script: npm ci
  displayName: 'npm ci'
- script: npx playwright install --with-deps
  displayName: 'Install Playwright browsers'
- script: npx playwright test --project=$(project) --shard=$(shard)
  displayName: 'Run Playwright tests'
  env:
    CI: 'true'

容器化执行（containerized）

在 job 层面直接指定 container（按语言选择对应镜像），容器内已预装浏览器，无需 install --with-deps：

trigger:
- main

pool:
  vmImage: ubuntu-latest
container: mcr.microsoft.com/playwright:v%%VERSION%%-noble

steps:
- task: UseNode@1
  inputs:
    version: '22'
  displayName: 'Install Node.js'

- script: npm ci
  displayName: 'npm ci'
- script: npx playwright test
  displayName: 'Run Playwright tests'
  env:
    CI: 'true'

Python / Java / .NET 分别使用 mcr.microsoft.com/playwright/python:…、mcr.microsoft.com/playwright/java:…、mcr.microsoft.com/playwright/dotnet:… 镜像，步骤简化为「安装依赖 → 运行测试」即可。

CI 平台配置：CircleCI

在 CircleCI 上运行 Playwright 与 GitHub Actions 非常相似。在 config 的 agent 定义中加入 docker: 指定官方预构建镜像即可：

executors:
  pw-noble-development:
    docker:
      - image: mcr.microsoft.com/playwright:v%%VERSION%%-noble

Python / Java / .NET 同理，把镜像换成 mcr.microsoft.com/playwright/{python,java,dotnet}:v%%VERSION%%-noble。

资源档位注意：使用 docker agent 定义时，你需要显式指定 Playwright 运行的 resource class 为 medium 档位。Playwright 的默认行为是把 workers 数设置为检测到的 CPU 核数（medium 档位为 2）。把 workers 数设置得高于核数会导致不必要的超时和失败——这与前文 resolveWorkers 按核数解析 '50%' 默认值的机制相呼应。

CircleCI 分片

CircleCI 的 CIRCLE_NODE_INDEX 从 0 开始计数，因此需要覆盖默认并行环境变量：给 CIRCLE_NODE_INDEX 加 1 后再传给 --shard：

    playwright-job-name:
      executor: pw-noble-development
      parallelism: 4
      steps:
        - run: SHARD="$((${CIRCLE_NODE_INDEX}+1))"; npx playwright test --shard=${SHARD}/${CIRCLE_NODE_TOTAL}

CI 平台配置：Jenkins

Jenkins 的 pipeline 支持 Docker agent，使用 Playwright Docker 镜像即可运行测试：

pipeline {
   agent { docker { image 'mcr.microsoft.com/playwright:v%%VERSION%%-noble' } }
   stages {
      stage('e2e-tests') {
         steps {
            sh 'npm ci'
            sh 'npx playwright test'
         }
      }
   }
}

Python 版本在容器内执行 pip install -r requirements.txt + pytest；Java 版本执行 mvn -B install -D skipTests --no-transfer-progress + mvn test；.NET 版本执行 dotnet build + dotnet test，镜像分别为 mcr.microsoft.com/playwright/{python,java,dotnet}:v%%VERSION%%-noble。

CI 平台配置：Bitbucket Pipelines、GitLab CI、Google Cloud Build 与 Drone

Bitbucket Pipelines 支持将公共 Docker 镜像作为构建环境，直接使用官方镜像：

image: mcr.microsoft.com/playwright:v%%VERSION%%-noble

GitLab CI 同样使用官方公共 Docker 镜像：

stages:
  - test

tests:
  stage: test
  image: mcr.microsoft.com/playwright:v%%VERSION%%-noble
  script:
  ...

各语言镜像选择规则与其他平台一致（python / java / dotnet 变体）。

GitLab 分片：GitLab CI 支持用 parallel 关键字把 job 拆分为多个并行小 job（命名为 job_name 1/N … job_name N/N）：

stages:
  - test

tests:
  stage: test
  image: mcr.microsoft.com/playwright:v%%VERSION%%-noble
  parallel: 7
  script:
    - npm ci
    - npx playwright test --shard=$CI_NODE_INDEX/$CI_NODE_TOTAL

GitLab 还支持 parallel:matrix：单个 job 在一条 pipeline 内以不同变量值运行多次。下例中 2 个 PROJECT 值 × 10 个 SHARD 值 = 共 20 个 job：

stages:
  - test

tests:
  stage: test
  image: mcr.microsoft.com/playwright:v%%VERSION%%-noble
  parallel:
    matrix:
      - PROJECT: ['chromium', 'webkit']
        SHARD: ['1/10', '2/10', '3/10', '4/10', '5/10', '6/10', '7/10', '8/10', '9/10', '10/10']
  script:
    - npm ci
    - npx playwright test --project=$PROJECT --shard=$SHARD

Google Cloud Build：

steps:
- name: mcr.microsoft.com/playwright:v%%VERSION%%-noble
  script:
  ...
  env:
  - 'CI=true'

Drone：

kind: pipeline
name: default
type: docker

steps:
  - name: test
    image: mcr.microsoft.com/playwright:v%%VERSION%%-noble
    commands:
      - npx playwright test

不建议缓存浏览器二进制文件

官方不建议缓存浏览器二进制：恢复缓存所花的时间与直接下载二进制的时间相当，而且在 Linux 上还必须安装操作系统依赖——这部分是不可缓存的。

如果你仍然希望在 CI 运行之间缓存浏览器二进制，请缓存这些目录，并以 Playwright 版本的哈希作为缓存 key。

调试浏览器启动失败

Playwright 支持 DEBUG 环境变量在执行期间输出调试日志。排查 Error: Failed to launch browser 错误时，把它设为 pw:browser 会非常有帮助：

DEBUG=pw:browser npx playwright test

DEBUG=pw:browser pytest

DEBUG=pw:browser mvn test

DEBUG=pw:browser dotnet test

在 CI 中以有头模式（headed）运行测试

Playwright 默认以 headless 模式启动浏览器。Linux agent 上运行有头模式需要安装 Xvfb——官方 Docker 镜像 与 GitHub Action 已预装 Xvfb。在有 Xvfb 的环境中以有头模式运行浏览器，只需在命令前加 xvfb-run：

xvfb-run npx playwright test

xvfb-run pytest

xvfb-run mvn test

xvfb-run dotnet test

小结：CI 配置要点速查

要点 建议 依据
浏览器依赖 npx playwright install --with-deps 或官方 Docker 镜像 docs/src/ci.md、docs/src/docker.md
workers CI 中设为 1，自托管强机器可并行、跨机用 shard docs/src/test-sharding.md、resolveWorkers
globalTimeout 始终设置（如 1 小时），且 job 级超时须高于它 runTasks
报告 容器化/镜像环境保证跨 OS 视觉一致；上传 playwright-report 各平台配置示例
大套件 --only-changed 预检 + 全量兜底；注意 fetch-depth: 0 detectChangedTestFiles
排查 DEBUG=pw:browser；有头模式用 xvfb-run docs/src/ci.md

以上所有配置均可在当前仓库的 docs/src/ci.md 中查阅原始版本，配合 docs/src/docker.md 的镜像说明与 docs/src/test-sharding.md 的分片详解，即可搭建完整的 Playwright CI 流水线。

playwright
Playwright is a framework for Web Testing and Automation. It allows testing Chromium, Firefox and WebKit with a single API.
项目地址：https://gitcode.com/GitHub_Trending/pl/playwright
登录后查看全文
热门项目推荐
相关项目推荐

项目优选

收起
kernelkernel
deepin linux kernel
C
33
18
pytorchpytorch
作为 Ascend for PyTorch 社区的核心组件，TorchNPU 是昇腾专为 PyTorch 打造的深度学习适配插件，使 PyTorch 框架能够直接调用昇腾 NPU，为开发者提供昇腾 AI 处理器的超强算力。
Python
855
1.34 K
kernelkernel
openEuler内核是openEuler操作系统的核心，既是系统性能与稳定性的基石，也是连接处理器、设备与服务的桥梁。
C
528
589
leetcodeleetcode
🔥LeetCode solutions in any programming language | 多种编程语言实现 LeetCode、《剑指 Offer（第 2 版）》、《程序员面试金典（第 6 版）》题解
Markdown
77
23
ops-transformerops-transformer
本项目是CANN提供的transformer类大模型算子库，实现网络在NPU上加速计算。
C++
1.13 K
2.73 K
docsdocs
暂无描述
Markdown
895
5.79 K
jiuwenswarmjiuwenswarm
JiuwenSwarm 是一款基于openJiuwen开发的智能AI Agent，它能够将大语言模型的强大能力，通过你日常使用的各类通讯应用，直接延伸至你的指尖。
Python
3.56 K
1.01 K
cann-learning-hubcann-learning-hub
CANN 学习中心仓，支持在线互动运行、边学边练，提供教程、示例与优化方案，一站式助力昇腾开发者快速上手。
Jupyter Notebook
998
511
ops-nnops-nn
本项目是CANN提供的神经网络类计算算子库，实现网络在NPU上加速计算。
C++
908
1.83 K
openGauss-serveropenGauss-server
openGauss kernel ~ openGauss is an open source relational database management system
C++
213
313