python调用火山引擎大模型

1. 安装模块

pip3.12  install 'volcengine-python-sdk[ark]'  --break-system-packages

　

2. 代码

第一种方式：

非流式生成，生成全部内容后，一次性输出

import os
import time
from volcenginesdkarkruntime import Ark
start=time.time()
client = Ark(api_key="2ce863e9-ccad-4c3a-a2b8")
completion = client.chat.completions.create(
# 可按需替换为 Model ID
      model="doubao-seed-1-6-250615",
      messages=[
      {"role": "system", "content": "什么是猫传腹？"}
      ]
)
print(completion.choices[0].message)
print("耗时:",time.time()-start)

　　

第二种方式：

流式生成，即一边生成一边输出

from volcenginesdkarkruntime import Ark
client = Ark(api_key="2ce863e9-ccad-4c3a")
completion = client.chat.completions.create(
    model="doubao-seed-1-6-250615",
    messages=[{"role": "system", "content": "什么是猫传腹？"}],
    stream=True,
)
for chunk in completion:
    if delta := chunk.choices[0].delta:
        print(delta.content, end="", flush=True)