> ## Documentation Index
> Fetch the complete documentation index at: https://docs.inb.cc/llms.txt
> Use this file to discover all available pages before exploring further.

# 流式响应

> 用 Server-Sent Events 实时接收模型输出

# 流式响应

设置 `stream: true` 即可逐 token 接收模型输出，适合聊天 UI 实时显示、减少用户等待感知。

## 快速示例

```bash theme={null}
curl https://api.getinfinityblue.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "stream": true,
    "messages": [{"role": "user", "content": "数到 10"}]
  }'
```

返回 Server-Sent Events，每行格式为 `data: {json}`，流以 `data: [DONE]` 结束。

每个事件的 JSON 结构如下：

```json theme={null}
{
  "id": "chatcmpl-9f3a8b2e1c0d",
  "object": "chat.completion.chunk",
  "created": 1717430000,
  "model": "gpt-5.4",
  "choices": [
    {
      "index": 0,
      "delta": {
        "content": "1"
      },
      "finish_reason": null
    }
  ]
}
```

最后一个事件的 `finish_reason` 会变为 `"stop"`（或 `"length"`），之后紧跟 `data: [DONE]`。

## Python 示例

```python theme={null}
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.getinfinityblue.com/v1",
)

with client.chat.completions.stream(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "数到 10"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)
```

## JavaScript / Node 示例

```javascript theme={null}
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://api.getinfinityblue.com/v1",
});

const stream = client.chat.completions.stream({
  model: "gpt-5.4",
  messages: [{ role: "user", content: "数到 10" }],
});

for await (const chunk of stream) {
  const content = chunk.choices[0]?.delta?.content ?? "";
  process.stdout.write(content);
}
```

## 浏览器端

`EventSource` 不支持自定义请求头，需改用 `fetch` + `ReadableStream`：

```javascript theme={null}
const response = await fetch("https://api.getinfinityblue.com/v1/chat/completions", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "gpt-5.4",
    stream: true,
    messages: [{ role: "user", content: "数到 10" }],
  }),
});

const reader = response.body.getReader();
const decoder = new TextDecoder();
let buffer = "";

while (true) {
  const { done, value } = await reader.read();
  if (done) break;
  buffer += decoder.decode(value, { stream: true });
  const lines = buffer.split("\n");
  buffer = lines.pop() || ""; // 保留最后一个可能不完整的行
  for (const line of lines) {
    if (line.startsWith("data: ") && line !== "data: [DONE]") {
      const chunk = JSON.parse(line.slice(6));
      const content = chunk.choices[0]?.delta?.content ?? "";
      // 更新 DOM，例如：
      // document.getElementById("output").textContent += content;
      console.log(content); // 或其他浏览器端渲染方式
    }
  }
}
```

<Tip>
  流式模式下 `usage` 字段默认不出现在中间 chunk 中。如果需要 token 统计，请在最后一个 chunk 或单独的非流式请求中获取。
</Tip>