baidu.md

百度文心一言

通过环境变量进行鉴权

import os 
os.environ["ERNIE_CLIENT_ID"] = "your-client-id"
os.environ["ERNIE_CLIENT_SECRET"] = "your-client-secret"

非流式调用

from unionllm import unionchat

# model call
response = unionchat(
    provider="wenxin",
    model="completions_pro", 
    messages = [{ "content": "Hello, how are you?","role": "user"}],
    stream=False
)

print(response)

流式调用

from unionllm import unionchat

# model call
response = unionchat(
    provider="wenxin",
    model="completions_pro", 
    messages = [{ "content": "Hello, how are you?","role": "user"}],
    stream=True
)

for chunk in response:
    print(chunk)

在接口中直接传入鉴权参数调用

# model call
response = unionchat(
    provider="wenxin",
    model="completions_pro", 
    client_id="your-client-id",
    client_secret="your-client-secret",
    messages = [{ "content": "Hello, how are you?","role": "user"}]
)

支持模型

支持百度文心平台的所有模型，注意传入模型的名称必须是请求地址结尾的部分。

例如： https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop/chat/completions_pro

那么，需要传入的模型名称是completions_pro

参考文档：

• API文档
• 模型价格