Claude API 说明文档

# Claude API — Python ## 安装 bash pip install anthropic ## 客户端初始化 python import anthropic # 默认（使用 ANTHROPIC_API_KEY 环境变量） client = anthropic.Anthropic() # 显式 API 密钥 client = anthropic.Anthropic(api_key="your-api-key") # 异步客户端 async_client = anthropic.AsyncAnthropic() --- ## 基础消息请求 python response = client.messages.create( model="claude-opus-4-7", max_tokens=16000, messages=[ {"role": "user", "content": "What is the capital of France?"} ] ) # response.content 是内容块对象（TextBlock, ThinkingBlock, ToolUseBlock 等）的列表。 # 在访问 .text 之前请检查 .type。 for block in response.content: if block.type == "text": print(block.text) --- ## 系统提示词 (System Prompts) python response = client.messages.create( model="claude-opus-4-7", max_tokens=16000, system="You are a helpful coding assistant. Always provide examples in Python.", messages=[{"role": "user", "content": "How do I read a JSON file?"}] ) --- ## 视觉能力 (Images) ### Base64 python import base64 with open("image.png", "rb") as f: image_data = base64.standard_b64encode(f.read()).decode("utf-8") response = client.messages.create( model="claude-opus-4-7", max_tokens=16000, messages=[{ "role": "user", "content": [ { "type": "image", "source": { "type": "base64", "media_type": "image/png", "data": image_data } }, {"type": "text", "text": "What's in this image?"} ] }] ) ### URL python response = client.messages.create( model="claude-opus-4-7", max_tokens=16000, messages=[{ "role": "user", "content": [ { "type": "image", "source": { "type": "url", "url": "https://example.com/image.png" } }, {"type": "text", "text": "Describe this image"} ] }] ) --- ## 提示词缓存 (Prompt Caching) 缓存大上下文以降低成本（最高可节省 90%）。**缓存是前缀匹配** — 前缀中任何位置的字节更改都会使之后的所有内容失效。关于放置模式、架构指导（固定的系统提示词、确定性的工具顺序、易变内容的位置）以及静默失效 au