一、并发请求多个模型
import asyncio
async def ask_model(question, model):
resp = await async_client.messages.create(
model=model, max_tokens=1024,
messages=[{"role": "user", "content": question}]
)
return model, resp.content[0].text
async def compare_models(question):
models = ["claude-sonnet-4-20250514", "claude-haiku-4-5"]
tasks = [ask_model(question, m) for m in models]
results = await asyncio.gather(*tasks)
for model, answer in results:
print(f"[{model}] {answer[:100]}")
动手练习
练习 1:多模型对比
同一问题问多个模型
asyncio.gather(*tasks)
查看答案
练习 2:速率限制
信号量控制并发
semaphore = asyncio.Semaphore(5)
查看答案
打卡验收
打卡方式: 全部勾选后,告诉我 "Day 33 打卡",并贴出你的代码。