试用零成本！普通手机畅跑 Gemma 4 模型（原生多模态），安卓与 iPhone 部署实测！

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/13a8fdfa266adb2.webp" alt="未标题 3" width="1920" height="1080" class="alignnone size-full wp-image-23672" decoding="async"</p
<ul data-path-to-node="11"

<h3 data-path-to-node="12"</h3

2. 性能模式与模型下载</h3

进入应用后，我们需要进行基础设置并下载对应的AI模型：</p
<ul data-path-to-node="14"

选择运行模式： 顶部可以选择AI模型的运行模式（快速、平衡、高性能、自定义）。</p

进入模型库： 点击左上方三个横杠菜单，进入 Model Hub。</p

下载Gemma 4： 向下滑动找到 Gemma 4 ECB量化版模型。</p
<ul data-path-to-node="14,2,1"

注：系统会根据手机配置推荐合适的版本。由于我的手机配置不高，它推荐的是1.2GB大小的 Q2_K_S_L 量化版。如果你的安卓机配置较高，可以选择最高2.3GB的版本以获得更好的体验。</i</p

勾选推荐版本，拉到底部点击 Download，耐心等待一分钟左右即可下载完成。<img src="https://www.foxnan.com/wp-content/uploads/2026/06/0fa3682580f4af7.webp" alt="2026 04 06 16 39 57.00 02 25 08.Still014 scaled" width="2560" height="1440" class="alignnone size-full wp-image-23673" decoding="async"</p

3. 载入与测试模型</h3
<ul data-path-to-node="16"

返回首页，进入 Fast 选项卡并拉到底部，打开 自定义模式。</p

创建模型： 名称可以随意填写。</p

选择模型： 选中我们刚刚下载好的Gemma 4 1.2G模型。</p

参数设置： * 上下文长度：根据手机硬件配置自定义（配置低时切勿拉太高）。</p
<ul data-path-to-node="16,3,1"

最高Token输出量：设置为 512。</p

点击保存，自定义模型就配置完成了！</p

实测表现： 我让它在完全离线的状态下帮我编写了一个贪吃蛇小游戏，生成速度非常快，日常对话也完全没有问题。</p

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/8ffa69a65ed405d.webp" alt="2026 04 06 16 39 57.00 02 51 02.Still015 scaled" width="2560" height="1440" class="alignnone size-full wp-image-23674" decoding="async"</h2
<h2 data-path-to-node="19" 第二部分：iOS 苹果端部署</h2
<p data-path-to-node="20"看完安卓，我们再来看看 iPhone 上的表现。</p
<h3 data-path-to-node="21"1. 下载 Locally AI</h3

【点击前往</span】</h3

通过下方链接前往App Store下载名为 Locally AI 的应用程序。这是一款完全免费、主打隐私与安全、支持加载本地离线模型的强大工具。</p
</div

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/eb18ee8dadb04f4.webp" alt="屏幕截图 2026 04 06 211153" width="758" height="650" class="alignnone size-full wp-image-23675 aligncenter" decoding="async"</p
<div
<h3 data-path-to-node="23"</h3
<h3 data-path-to-node="23"2. 下载 iOS 版 Gemma 4</h3
<ul data-path-to-node="24"

打开应用，跳过欢迎页的默认模型推荐。</p

点击上方的 选择模型，在列表中找到支持深入思考和多模态的最新的 Gemma 4。</p

点击下载（文件大小约为 3.61GB，版本为 E2B 量化版）。</p

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/4b7618a558fc47b.webp" alt="屏幕截图 2026 04 06 211237" width="666" height="369" class=" wp-image-23676 aligncenter" decoding="async"</h2
<h2 data-path-to-node="26"</h2
<h2 data-path-to-node="26"极客硬核测试：Gemma 4 到底有多强？</h2

<div class="code-block code-block-default code-block-bc-rgba-211-211-210-0.82 code-block-3 ai-track" data-ai="WzMsMCwiQmxvY2sgMyIsIiIsMV0="
<div class="ai-wait-for-interaction ai-wait-for-interaction-3" data-code="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" data-class="Y29kZS1ibG9jaw=="</div
</div

测试一：多模态视觉识别（拍照识物）</h3

我随手在桌面上放了一些杂物，打开 深入思考模式 拍照发给它，看看它能识别出什么。</p
<blockquote data-path-to-node="30"

我的提问： 你看到了什么？桌面上堆有哪些东西？ Gemma 4回答： > * ✅ 一部深色的智能手机</p
<ul data-path-to-node="30,1"

✅ 一个带花卉图案的偏紫色/粉色手机保护壳</p

✅ 一个亮蓝绿色的小瓶（护肤品/精油）</p

✅ 一个带标签的大罐子（益生菌）</p

✅ 一个黑色小电子配件（声卡录音设备）</p

</blockquote

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/2750fa573e4c69d.webp" alt="2026 04 06 16 39 57.00 04 10 21.Still016 scaled" width="2560" height="1440" class="alignnone size-full wp-image-23677" decoding="async"</p
<p data-path-to-node="31"<b data-path-to-node="31" data-index-in-node="0"翻车环节：</b 唯独桌上的西瓜子它没认出来。我再次特写拍照问它“黑色点点是什么，总共有多少个？” 它推测是干燥的种子（算答对），但数量数成了 9 个（实际是 16 个）。 <i data-path-to-node="31" data-index-in-node="84"吐槽：为了公平起见，我也问了 ChatGPT，虽然 ChatGPT 认出了西瓜子，但也数错了（数成了 17 个）。看来 AI 数数依然是个老大难问题！</i</p
<h3 data-path-to-node="32"测试二：逻辑推理（蒙提霍尔问题）</h3
<blockquote data-path-to-node="33"
<p data-path-to-node="33,0"<b data-path-to-node="33,0" data-index-in-node="0"题目：</b 三扇门（一车两羊），你选定一扇后，主持人打开一扇羊门。问：换门是否有利？ <b data-path-to-node="33,0" data-index-in-node="41"Gemma 4 回答：</b 这是一个经典的蒙提霍尔问题变种。答案是<b data-path-to-node="33,0" data-index-in-node="72"换门更有利</b。坚持原门只有 1/3 的概率，而换门可以抓住主持人排除错误选项带来的机会，将概率提升到 2/3。 <b data-path-to-node="33,0" data-index-in-node="127"结论：</b 逻辑非常清晰，完全正确！</p
</blockquote
<h3 data-path-to-node="34"测试三：前端代码编程</h3
<blockquote data-path-to-node="35"
<p data-path-to-node="35,0"<b data-path-to-node="35,0" data-index-in-node="0"要求：</b 编写一个 3D 鱼缸场景，水、水草和鱼要有真实感。 <b data-path-to-node="35,0" data-index-in-node="30"结果：</b 它不仅提供了 HTML，还一并写好了 CSS 样式和 JS 代码。将代码复制到电脑上运行后，视觉效果非常逼真。作为一个手机端本地运行的小模型，能达到这种渲染水准令人惊艳。</p
<p data-path-to-node="35,0"<img src="https://www.freedidi.com/wp-content/uploads/2026/04/2026-04-06-16-39-57.00_07_56_19.Still018-scaled.webp" alt="2026 04 06 16 39 57.00 07 56 19.Still018 scaled" width="2560" height="1440" class="alignnone size-full wp-image-23680" decoding="async"</p
</blockquote
<h3 data-path-to-node="36"测试四：医疗物品识别安全机制</h3
<p data-path-to-node="37"我拍了一盒“瑞巴派特片”给它。它准确识别出了药片名称，但立刻触发了安全机制，表示“由于涉及身体和健康产品，无法提供医疗建议，请务必咨询医生”。表现得非常严谨。</p
</div
<p data-path-to-node="37"<img src="https://www.freedidi.com/wp-content/uploads/2026/04/%E5%B1%8F%E5%B9%95%E6%88%AA%E5%9B%BE-2026-04-06-211503.webp" alt="屏幕截图 2026 04 06 211503" width="929" height="651" class="alignnone size-full wp-image-23679 aligncenter" decoding="async"</p
<div
<h2 data-path-to-node="39"终极考验：断网飞行模式测试</h2
<p data-path-to-node="40"为了验证它的纯离线真伪，我<b data-path-to-node="40" data-index-in-node="13"断开了所有网络并开启了飞行模式</b。</p
<ol start="1" data-path-to-node="41"

写长篇小说： 让它写一篇5000字左右的恐怖小说。在完全离线状态下，它使用繁体字分章输出，仅用时1分钟左右就完成了创作。</p

微距视觉推理： 在离线状态下发给它一张包含大象和蚂蚁的画。它准确识别出了大象头部，并声明“由于细节非常小，对蚂蚁的识别是基于微小尺寸的推测”。离线多模态能力确实靠谱。</p

总结与资源获取</h3

Gemma 4模型在手机端的本地离线表现远远超出了我的预期，无论是生成速度、逻辑推理还是多模态视觉，都达到了相当高的可用级别。</p
</div
<div class="donation-container"
<button class="donation-main-btn" id="donationMainBtn"
打赏支持

<div class="donation-modal" id="donationModal"
<div class="donation-modal-content"
×

感谢您的支持

<div class="donation-options"
<div class="donation-option" data-type="alipay"

<img src="https://www.foxnan.com/wp-content/uploads/2026/06/11c254db930eb75-6.png" alt="alipay" class="payment-icon" decoding="async"</div
<span支付宝</span
</div
<div class="donation-option" data-type="wechat"
<div class="payment-icon-wrapper"<img src="https://www.freedidi.com/wp-content/plugins/donation-button/images/wechat.png" alt="wechat" class="payment-icon" decoding="async"</div
<span微信支付</span
</div
<div class="donation-option" data-type="bitcoin"
<div class="payment-icon-wrapper"<img src="https://www.freedidi.com/wp-content/plugins/donation-button/images/bitcoin.png" alt="bitcoin" class="payment-icon" decoding="async"</div
<spanBitcoin</span
</div
<div class="donation-option" data-type="usdt"
<div class="payment-icon-wrapper"<img src="https://www.freedidi.com/wp-content/plugins/donation-button/images/usdt.png" alt="usdt" class="payment-icon" decoding="async"</div
<spanUSDT</span
</div
</div

<div class="qrcode-display" id="qrcodeDisplay"
<div class="qr-container"

</div
<p id="qrcodeText"</p

<div id="donationAddressArea" style="display:none;"
<div class="address-box"
<input type="text" id="walletAddress" readonly
<button type="button" id="copyBtn"复制</button
</div
<span id="copyMsg" style="font-size:12px; color:#26A17B; display:none;"已复制!</span
</div

<button class="back-btn" id="backBtn"返回选择</button
</div
</div
</div
</div
<div class="code-block code-block-center code-block-1 ai-track" data-ai="WzEsMCwiQmxvY2sgMSIsIiIsMV0="
<div class="ai-wait-for-interaction ai-wait-for-interaction-1" data-code="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" data-class="Y29kZS1ibG9jaw=="</div
</div

<div class="clear"</div