Gemini 4 Hydrogen生成长文本：开启后台执行，轮询状态并合并全部输出段-Python教程-PHP中文网

需要超过65536 token的思考和输出预算时，可在hydrogen的interactions请求中同时设置background=true和较高的max_output_tokens，保留返回id并轮询到完成，最后合并所有model_output步骤中的文本。统一窗口仍受输入占用影响，后台模式也不是无限输出。

准备SDK与任务预算

在已配置GEMINI_API_KEY且安装最新版google-genai的Python环境执行。下面将预算设为131072作为示例；它是上限，不代表模型一定生成同样长度，实际输入、思考及输出还需装入1048576 Token统一窗口。

谷歌Long decode说明统一窗口与连续生成段

import time
from google import genai

client = genai.Client()
job = client.interactions.create(
    model="gemini-4-hydrogen",
    input="为一个日志处理服务编写分章节的设计说明，覆盖输入、存储、错误处理和验证。",
    background=True,
    generation_config={
        "max_output_tokens": 131072,
        "thinking_level": "high",
    },
)
job_id = job.id
print("Interaction:", job_id)
deadline = time.monotonic() + 1800
while job.status in ("queued", "in_progress"):
    if time.monotonic() >= deadline:
        raise TimeoutError("Still running; keep the interaction ID for later retrieval")
    time.sleep(5)
    job = client.interactions.get(id=job_id)
if job.status != "completed":
    raise RuntimeError(f"Interaction ended with status: {job.status}")
text = "".join(
    part.text
    for step in job.steps
    if step.type == "model_output"
    for part in step.content
    if part.type == "text"
)
print(text)
print(job.usage)