Baidu 千帆 API - 永久免费国产 AI API 服务

from openai import OpenAI

# 配置百度千帆（使用 OpenAI SDK）
client = OpenAI(
    api_key="YOUR_BAIDU_API_KEY",
    base_url="https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop"
)

# 调用永久免费的 ERNIE-3.5-8K
response = client.chat.completions.create(
    model="ernie-3.5-8k",
    messages=[
        {"role": "system", "content": "你是一个有帮助的助手"},
        {"role": "user", "content": "介绍一下量子计算"}
    ]
)

print(response.choices[0].message.content)

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_BAIDU_API_KEY",
    base_url="https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop"
)

# 使用永久免费的 ERNIE-3.5
response = client.chat.completions.create(
    model="ernie-3.5-8k",
    messages=[
        {"role": "user", "content": "什么是人工智能？"}
    ]
)

print(response.choices[0].message.content)

# 流式输出（适合实时显示）
stream = client.chat.completions.create(
    model="ernie-3.5-8k",
    messages=[
        {"role": "user", "content": "写一篇关于 AI 的文章"}
    ],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

messages = [
    {"role": "system", "content": "你是一个 Python 编程助手"}
]

# 第一轮对话
messages.append({"role": "user", "content": "如何读取 CSV 文件？"})
response = client.chat.completions.create(
    model="ernie-3.5-8k",
    messages=messages
)
messages.append({
    "role": "assistant", 
    "content": response.choices[0].message.content
})

# 第二轮对话
messages.append({"role": "user", "content": "如何处理缺失值？"})
response = client.chat.completions.create(
    model="ernie-3.5-8k",
    messages=messages
)
print(response.choices[0].message.content)

# 安装官方 SDK
# pip install qianfan

import qianfan

# 通过 API Key 和 Secret Key 初始化
chat_comp = qianfan.ChatCompletion(
    api_key="YOUR_API_KEY",
    secret_key="YOUR_SECRET_KEY"
)

# 调用 ERNIE-3.5-8K
resp = chat_comp.do(
    model="ERNIE-3.5-8K",
    messages=[{
        "role": "user",
        "content": "你好，请介绍一下自己"
    }]
)

print(resp["result"])

# 获取 access_token
curl -X POST \
  'https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials&client_id=YOUR_API_KEY&client_secret=YOUR_SECRET_KEY'

# 调用 API（使用获取的 access_token）
curl -X POST \
  'https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop/chat/ernie-3.5-8k?access_token=YOUR_ACCESS_TOKEN' \
  -H 'Content-Type: application/json' \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "你好，请介绍一下自己"
      }
    ]
  }'

import hashlib
import json

cache = {}

def cached_completion(model, messages):
    key = hashlib.md5(
        json.dumps({"model": model, "messages": messages}).encode()
    ).hexdigest()
    
    if key in cache:
        return cache[key]
    
    response = client.chat.completions.create(
        model=model,
        messages=messages
    )
    
    cache[key] = response
    return response

client = OpenAI(
    api_key="YOUR_BAIDU_API_KEY",
    base_url="https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop"
)

def ai_customer_service(user_question):
    """智能客服助手"""
    response = client.chat.completions.create(
        model="ernie-3.5-8k",  # 使用永久免费模型
        messages=[
            {
                "role": "system",
                "content": "你是一个专业的客服助手，负责回答用户关于产品的问题"
            },
            {
                "role": "user",
                "content": user_question
            }
        ]
    )
    return response.choices[0].message.content

# 使用
answer = ai_customer_service("你们的产品有什么特点？")
print(answer)

def summarize_document(document_text):
    """生成文档摘要"""
    response = client.chat.completions.create(
        model="ernie-3.5-8k",
        messages=[
            {
                "role": "system",
                "content": "你是一个文档分析专家，擅长提取关键信息"
            },
            {
                "role": "user",
                "content": f"请总结以下文档的核心内容：\n\n{document_text}"
            }
        ]
    )
    return response.choices[0].message.content

def content_moderation(user_content):
    """内容审核"""
    response = client.chat.completions.create(
        model="ernie-speed-8k",  # 使用极速模型
        messages=[
            {
                "role": "system",
                "content": "你是一个内容审核专家，判断内容是否合规"
            },
            {
                "role": "user",
                "content": f"请判断以下内容是否合规：\n{user_content}"
            }
        ]
    )
    return response.choices[0].message.content