python实战豆包大模型-文本模型

Fxouz2024-12-05 14:13

准备

注册并登录火山引擎

[官网] https://console.volcengine.com/ark/region:ark+cn-beijing/experience/chat

开通模型

创建推理接入点

调用参数API Key、model

调用

SDK调用

pip install --upgrade 'volcengine-python-sdk[ark]'

python 复制代码

from volcenginesdkarkruntime import Ark

client = Ark(
    base_url="https://ark.cn-beijing.volces.com/api/v3",
    api_key="上面获取的api-key"
)

# Non-streaming:
print("----- standard request -----")
completion = client.chat.completions.create(
    model="上面获取的model",
    messages = [
        {"role": "system", "content": "你是豆包，是由字节跳动开发的 AI 人工智能助手"},
        {"role": "user", "content": "常见的十字花科植物有哪些？"},
    ],
)
print(completion.choices[0].message.content)

# Streaming: 流式调用
print("----- streaming request -----")
stream = client.chat.completions.create(
    model="上面获取的model",
    messages = [
        {"role": "system", "content": "你是豆包，是由字节跳动开发的 AI 人工智能助手"},
        {"role": "user", "content": "常见的十字花科植物有哪些？"},
    ],
    stream=True
)
for chunk in stream:
    if not chunk.choices:
        continue
    print(chunk.choices[0].delta.content, end="")

HTTP调用

url https://ark.cn-beijing.volces.com/api/v3

json_date { model:"上面获取的model",messages = [{"role": "system", "content": "你是豆包，是由字节跳动开发的 AI 人工智能助手"},{"role": "user", "content": "常见的十字花科植物有哪些？"},]}

sk 上面获取的api-key

python 复制代码

async def get_common_content(url,json_data,sk):
  try:
    async with aiohttp.ClientSession() as session:
        authorization = "Bearer "+ (str(sk))
        # print(f"**** 调用接口 -{url}-{json_data}-{authorization}****")
        headers = {"Content-Type": 'application/json',"Authorization": authorization }
        async with session.post(url,data = json.dumps(json_data), headers = headers) as response:
            return await response.text()
  except Exception as e:
       print(f"**ERROR 调用模型失败....{e}**")