总算等到了，Gemini Nano 马上就要面向所有 Chrome 用户开放了（我最早被误导以为是在 Chrome 138——但我自己核实了一下，从 Chrome 137 开始，它已经在部分场景下开始无需 flag 即可使用）。我是被这篇 HN 帖子提醒的。预计到年底就会完全默认开放。

我不太喜欢 Google 写文档的方式，所以这篇博文基本上就是我按自己的思路把他们的文档重写了一遍。

他们提供了几个针对常用场景的 API，但说实话，作为工程师你真正会关心的，主要还是最灵活、最开放的那个——Prompt API。

配置

和当初过度承诺的 window.ai 不同（它还催生了一堆仿制的垫片，比如 xander 和 chromeai，现在都没人维护了），现在正式发布的实现可没那么“优雅”。总之，下面是目前的配置方法。

1. 确保你装的是 Chrome 137 及以上版本
2. 前往 chrome://flags/#prompt-api-for-gemini-nano 并启用（遗憾的是需要重启 Chrome）
3. 然后首次调用 LangaugeModel.create() 来下载模型——在家用 Wi-Fi 下大概要几分钟。Gemini 说“下载大小约为 1.5 GB 至 2.4 GB”，这么看大概是个 4 到 60 亿参数、做了 4 到 8 比特量化的模型。

const session = await LanguageModel.create({
monitor(m) {
  m.addEventListener("downloadprogress", (e) => {
    console.log(`Downloaded ${e.loaded * 100}%`);
  });
},
// // uncomment if want multimodal input https://developer.chrome.com/docs/ai/prompt-api#multimodal_capabilities 
// expectedInputs: [
//  { type: "audio" },
//  { type: "image" }
//  ]
})

需要了解的基础要点

加载后的模型上下文是 6k token（不带任何 initialPrompts 直接查 inputQuota 就能看到）：

session.inputQuota
// 6144

和 Gemini Nano 团队不同，我恰好是个认为函数调用 / JSON 输出非常重要的人，所以来看看在 Gemini Nano 里怎么实现这个，prompt 示例搬自 Hamel 和 Jason：

const JSONschema = `<schema>
{
    "description": "Correctly extracted \`UserDetail\` with all the required parameters with correct types",
    "name": "UserDetail",
    "parameters": {
        "properties": {
            "age": {
                "title": "Age",
                "type": "integer"
            },
            "name": {
                "title": "Name",
                "type": "string"
            }
        },
        "required": [
            "age",
            "name"
        ],
        "type": "object"
    }
}
</schema>`
const JSONsession = await LanguageModel.create({
  initialPrompts: [
    { role: 'system', content: 'You are a helpful LLM that only responds in valid JSON fitting a schema: ' + JSONschema },
    { role: 'user', content: "Extract Jason is 35 years old" },
    { role: 'assistant', content: '{age: 35, name: Jason}'},
  ]
});

const result1 = await JSONsession.prompt("Extract sarah is 22 years old");
console.log(result1);
// {age: 22, name: Sarah}

坑点

它的指令遵循能力不太好，所以必填字段其实并不会被严格遵守：

const result1 = await JSONsession.prompt("its been a year since vibhu's birthday, he was 28 last year, guess how old he is now");
console.log(result1);
// { "age": 29 }

另一点是，session 默认是有状态的，如果忘了这点会很坑。所以一个无状态的版本大概长这样：

const baseSession = await LanguageModel.create({
  initialPrompts: // blah blah, as above
})

// you can also implement this as a class if you want to force users to use`new` keyword to make super clear it is stateless
const statelessSession = {  
    async prompt(str) {
        const clonedSession = await session.clone()
        return clonedSession.prompt(str)
    }
}

// these are all stateless calls now! yay repeatability and predictability!
const result1 = await statelessSession.prompt("Extract sarah is 22 years old");
console.log(result1);
const result2 = await statelessSession.prompt("Extract tanisha is 30 years old");
console.log(result2);

正是因为这些坑，你大概会想自己手写一个小的封装库，或者参考 https://github.com/kstonekuan/simple-chromium-ai

最后，给不熟悉 JS 的同学一个小技巧：如何在浏览器环境中（也就是不用 npm install 或打包步骤）用 ESM 语法引入这些封装库（可能需要 <script type="module">——在 localhost 或 CSP 比较宽松的网站上运行）：

// alternatively use https://cdn.jsdelivr.net/npm/[email protected]/dist/simple-chromium-ai.mjs
const ChromiumAI = await import('https://unpkg.com/[email protected]/dist/simple-chromium-ai.mjs');

const ai = await ChromiumAI.initialize("You are a friendly assistant");
const response = await ChromiumAI.prompt(ai, "Tell me a joke");
console.log(response);
const ChromiumAI = await import('https://unpkg.com/[email protected]/dist/simple-chromium-ai.mjs');

const ai = await ChromiumAI.initialize("You are a friendly assistant");
const response = await ChromiumAI.prompt(ai, "Tell me a joke");
console.log(response);
// Why don't scientists trust atoms?  Because they make up everything! 

// and of course... the structured output implementation now works:
const schema = {
  type: "object",
  properties: {
    sentiment: {
      type: "string",
      enum: ["positive", "negative", "neutral"]
    },
    confidence: {
      type: "number",
      minimum: 0,
      maximum: 1
    },
    keywords: {
      type: "array",
      items: { type: "string" },
      maxItems: 5
    }
  },
  required: ["sentiment", "confidence", "keywords"]
};

// Create session with response constraint
const response = await ChromiumAI.prompt(
  ai, 
  "Analyze the sentiment of this text: 'I love this new feature!'",
  undefined, // no timeout
  { responseConstraint: schema }
);

// Response will be valid JSON matching the schema
const result = JSON.parse(response);
console.log(result);