发布时间：2025 年 1 月 21 日
流式 LLM 响应由增量和连续发出的数据组成。 在服务器和客户端上，流式数据的外观不同。
从服务器
为了了解流式响应的内容，我使用命令行工具 curl 提示 Gemini 告诉我一个长笑话。请考虑以下对 Gemini API 的调用。如果您要试用，请务必将网址中的 {GOOGLE_API_KEY} 替换为您的 Gemini API 密钥。
$ curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:streamGenerateContent?alt=sse&key={GOOGLE_API_KEY}" \
-H 'Content-Type: application/json' \
--no-buffer \
-d '{ "contents":[{"parts":[{"text": "Tell me a long T-rex joke, please."}]}]}'
此请求会以事件流格式记录以下（截断的）输出。每行都以 data: 开头，后跟消息载荷。具体格式实际上并不重要，重要的是文本块。
//
data: {"candidates":[{"content": {"parts": [{"text": "A T-Rex"}],"role": "model"},
"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}],
"usageMetadata": {"promptTokenCount": 11,"candidatesTokenCount": 4,"totalTokenCount": 15}}
data: {"candidates": [{"content": {"parts": [{ "text": " walks into a bar and orders a drink. As he sits there, he notices a" }], "role": "model"},
"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}],
"usageMetadata": {"promptTokenCount": 11,"candidatesTokenCount": 21,"totalTokenCount": 32}}
第一个载荷为 JSON。请仔细查看突出显示的 candidates[0].content.parts[0].text：
{
"candidates": [
{
"content": {
"parts": [
{
"text": "A T-Rex"
}
],
"role": "model"
},
"finishReason": "STOP",
"index": 0,
"safetyRatings": [
{
"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
"probability": "NEGLIGIBLE"
},
{
"category": "HARM_CATEGORY_HATE_SPEECH",
"probability": "NEGLIGIBLE"
},
{
"category": "HARM_CATEGORY_HARASSMENT",
"probability": "NEGLIGIBLE"
},
{
"category": "HARM_CATEGORY_DANGEROUS_CONTENT",
"probability": "NEGLIGIBLE"
}
]
}
],
"usageMetadata": {
"promptTokenCount": 11,
"candidatesTokenCount": 4,
"totalTokenCount": 15
}
}
第一个 text 条目是 Gemini 回答的开头。当您提取更多 text 条目时，响应会以换行符分隔。
以下代码段显示了多个 text 条目，其中显示了模型的最终响应。
"A T-Rex"
" was walking through the prehistoric jungle when he came across a group of Triceratops. "
"\n\n\"Hey, Triceratops!\" the T-Rex roared. \"What are"
" you guys doing?\"\n\nThe Triceratops, a bit nervous, mumbled,
\"Just... just hanging out, you know? Relaxing.\"\n\n\"Well, you"
" guys look pretty relaxed,\" the T-Rex said, eyeing them with a sly grin.
\"Maybe you could give me a hand with something.\"\n\n\"A hand?\""
...
但是，如果您不问霸王龙笑话，而是问模型一些稍微复杂的问题，会怎么样？例如，让 Gemini 编写一个 JavaScript 函数来确定数字是偶数还是奇数。text: 分块看起来略有不同。
输出现在包含 Markdown 格式，以 JavaScript 代码块开头。以下示例包含与上文相同的预处理步骤。
"```javascript\nfunction"
" isEven(number) {\n // Check if the number is an integer.\n"
" if (Number.isInteger(number)) {\n // Use the modulo operator"
" (%) to check if the remainder after dividing by 2 is 0.\n return number % 2 === 0; \n } else {\n "
"// Return false if the number is not an integer.\n return false;\n }\n}\n\n// Example usage:\nconsole.log(isEven("
"4)); // Output: true\nconsole.log(isEven(7)); // Output: false\nconsole.log(isEven(3.5)); // Output: false\n```\n\n**Explanation:**\n\n1. **`isEven("
"number)` function:**\n - Takes a single argument `number` representing the number to be checked.\n - Checks if the `number` is an integer using `Number.isInteger()`.\n - If it's an"
...
更具挑战性的是，一些标记的项从一个分块开始，在另一个分块结束。某些标记是嵌套的。在以下示例中，突出显示的函数会拆分到两行：**isEven( 和 number) function:**。合并后，输出为 **isEven("number) function:**。这意味着，如果您想输出格式化的 Markdown，就不能仅使用 Markdown 解析器单独处理每个代码段。
从客户端
如果您在客户端上使用 MediaPipe LLM 等框架运行 Gemma 等模型，则流式数据会通过回调函数传入。
例如：
llmInference.generateResponse(
inputPrompt,
(chunk, done) => {
console.log(chunk);
});
借助 Prompt API，您可以通过迭代 ReadableStream 以分块形式获取流式数据。
const languageModel = await self.ai.languageModel.create();
const stream = languageModel.promptStreaming(inputPrompt);
for await (const chunk of stream) {
console.log(chunk);
}
后续步骤
您是否在想如何高效且安全地渲染流式数据？请参阅呈现 LLM 回答的最佳实践。