← 返回

使用 LLM API

最快开始使用 LLM 的方式是调用 API。无需本地部署，按需付费。

主流 API 服务对比

服务 模型 输入价格 输出价格 特点
OpenAI GPT-4o $2.50 $10.00 最强综合
OpenAI GPT-4o-mini $0.15 $0.60 性价比高
Anthropic Claude 3.5 Sonnet $3.00 $15.00 编码最强
Anthropic Claude 3 Haiku $0.25 $1.25 快速便宜
Google Gemini 1.5 Pro $1.25 $5.00 1M 上下文
阿里云 Qwen-Plus ¥0.4 ¥1.2 中文优化
智谱 GLM-4 ¥0.5 ¥0.5 便宜
DeepSeek DeepSeek-V3 $0.14 $0.28 超低价

价格为每 1M tokens (USD)，实际以官方为准

如何选择 API？

OpenAI API (Python)

from openai import OpenAI

# 初始化
client = OpenAI(api_key="your-api-key")

# 基础调用
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "你是一个有帮助的助手。"},
        {"role": "user", "content": "什么是量子计算？"}
    ],
    max_tokens=500,
    temperature=0.7
)

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

# 流式输出
for chunk in client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "讲个故事"}],
    stream=True
):
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Claude API (Python)

import anthropic

client = anthropic.Anthropic(api_key="your-api-key")

message = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)

print(message.content[0].text)

国内 API (以 DeepSeek 为例)

from openai import OpenAI

# DeepSeek 兼容 OpenAI SDK
client = OpenAI(
    api_key="your-deepseek-key",
    base_url="https://api.deepseek.com"
)

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "你好"}]
)

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

Function Calling

# 定义函数
tools = [
    {
        "type": "function",
        "function": {
            "name": "get_weather",
            "description": "获取城市天气",
            "parameters": {
                "type": "object",
                "properties": {
                    "city": {"type": "string", "description": "城市名"}
                },
                "required": ["city"]
            }
        }
    }
]

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "北京天气怎么样？"}],
    tools=tools
)

# 模型决定调用函数
tool_call = response.choices[0].message.tool_calls[0]
function_name = tool_call.function.name
arguments = json.loads(tool_call.function.arguments)

# 执行函数
result = get_weather(arguments["city"])

# 把结果返回给模型
final_response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "user", "content": "北京天气怎么样？"},
        response.choices[0].message,
        {
            "role": "tool",
            "tool_call_id": tool_call.id,
            "content": json.dumps(result)
        }
    ]
)

成本估算

import tiktoken

def estimate_cost(text, model="gpt-4o"):
    enc = tiktoken.encoding_for_model(model)
    tokens = len(enc.encode(text))
    input_cost = tokens / 1e6 * 2.50  # 输入价格
    output_cost = tokens / 1e6 * 10.00  # 输出价格
    return tokens, input_cost, output_cost

tokens, input_cost, output_cost = estimate_cost("你的文本...")
print(f"Token 数: {tokens}")
print(f"输入成本: ${input_cost:.4f}")
print(f"输出成本: ${output_cost:.4f}")

实用技巧

1. 减少成本

2. 提升质量

3. 错误处理

安全建议

官方文档