Anthropic API

Introduction

Anthropic's Claude API is the second-most-used LLM API. Claude 3.5 Sonnet is widely considered the best coding model and excels at long-context tasks (200k tokens).

Beginner analogy: Same shape as OpenAI's API but with Claude as the brain.

Understanding the topic

Core concepts to understand:

• Models: claude-3-5-sonnet, claude-3-5-haiku, claude-3-opus.
• 200k token context window.
• Tool use (function calling) supported.
• Prompt caching for huge cost savings.
• Computer use — Claude can drive a virtual computer.

Syntax reference

Visual workflow / architecture:

POST https://api.anthropic.com/v1/messages
x-api-key: sk-ant-...
anthropic-version: 2023-06-01

{
  "model": "claude-3-5-sonnet-20241022",
  "max_tokens": 1024,
  "messages": [
    { "role": "user", "content": "Hello, Claude!" }
  ]
}

Real-world use

Cursor, Replit, Notion AI, and many enterprise products default to Claude for coding and long-document tasks.

Best practices

• Use prompt caching aggressively — 90% cheaper for repeated context.
• Try Sonnet before Opus — Sonnet is often as good and cheaper.

Hands-on exercise

Interview preparation — practice these questions:

• Q1. How does Claude differ from GPT-4?
• Q2. What is prompt caching?
• Q3. What is Anthropic's 'computer use' feature?