Claude

Claude is Anthropic's most capable AI for complex reasoning, coding, and tool use — the top choice for developers building agentic systems.

Why Vibe Coders Use It

• Powers Claude Code and Cursor — the industry-leading AI coding assistants
• Best-in-class tool use — call external APIs and functions reliably with extended thinking
• Excellent at agentic loops — multi-turn reasoning for autonomous workflows
• 200K context window — process entire codebases and long documents
• Multimodal — understands images for UI design and architecture reviews

Key Specs

Getting Started

1. Create an Account & API Key

Sign up at console.anthropic.com and generate an API key.

2. Set Your Environment Variable

ANTHROPIC_API_KEY=sk-ant-your-key-here

3. Install the SDK

npm install @anthropic-ai/sdk

4. Build an Agentic Workflow

import Anthropic from '@anthropic-ai/sdk';

const client = new Anthropic();

const tools = [
  {
    name: 'search_docs',
    description: 'Search your codebase for a function or file',
    input_schema: {
      type: 'object',
      properties: {
        query: { type: 'string', description: 'What to search for' },
      },
      required: ['query'],
    },
  },
];

async function runAgent(userMessage: string) {
  let messages = [{ role: 'user' as const, content: userMessage }];

  while (true) {
    const response = await client.messages.create({
      model: 'claude-opus-4-20250514',
      max_tokens: 4096,
      tools,
      messages,
    });

    // If model wants to use a tool, simulate the result
    if (response.stop_reason === 'tool_use') {
      const toolBlocks = response.content.filter(
        (block: any) => block.type === 'tool_use'
      );
      const toolResults = toolBlocks.map((block: any) => ({
        type: 'tool_result' as const,
        tool_use_id: block.id,
        content: `Simulated result for: ${block.input.query}`,
      }));

      messages = [
        ...messages,
        { role: 'assistant' as const, content: response.content },
        { role: 'user' as const, content: toolResults },
      ];
    } else {
      // Model finished reasoning
      return response.content
        .filter((block: any) => block.type === 'text')
        .map((block: any) => block.text)
        .join('\n');
    }
  }
}

await runAgent('Review the authentication module and suggest improvements');

Using with AI SDK (Next.js)

import { anthropic } from '@ai-sdk/anthropic';
import { streamText } from 'ai';

export async function POST(req: Request) {
  const { messages } = await req.json();

  const result = streamText({
    model: anthropic('claude-opus-4-20250514'),
    messages,
  });

  return result.toDataStreamResponse();
}

When to Use This vs. Alternatives

Use Claude when you need the strongest reasoning, tool use, and agentic capabilities — ideal for multi-step workflows and complex problem-solving. Use GPT if you prioritize speed and cost for general-purpose tasks. Use Gemini if you need a very long context (1M tokens) for entire codebases.

Models

• Claude Opus 4 — Most capable; best for complex reasoning and agentic loops
• Claude Sonnet 4 — Best balance of performance and cost
• Claude Haiku 3.5 — Fastest and cheapest for high-volume tasks

Resources

• Anthropic Official Documentation
• Claude API Console
• Claude Model Cards
• Prompt Engineering Guide
• See the full Claude profile on LLMReference →