Overview
Anthropic’s Claude models are renowned for complex reasoning, careful instruction following, and exceptionally long context windows. Claude is an excellent choice for applications that involve nuanced conversation, long document analysis, and code generation.Setup
Get an API Key
Sign up or log in at console.anthropic.com. Navigate to API Keys and create a new key.
Configure in Nadoo
Go to Admin > Model Providers > Anthropic and enter:
| Field | Required | Description |
|---|---|---|
| API Key | Yes | Your Anthropic API key (starts with sk-ant-) |
Available Models
| Model | Context Window | Strengths |
|---|---|---|
claude-3.5-sonnet | 200K tokens | Best balance of intelligence, speed, and cost. Recommended default. |
claude-3-opus | 200K tokens | Highest reasoning capability. Best for complex, multi-step tasks. |
claude-3-haiku | 200K tokens | Fastest and most cost-efficient. Ideal for high-volume, simpler tasks. |
Capabilities
Chat Completion
Conversational AI with streaming support and careful instruction following.
Long Context
Process up to 200K tokens — equivalent to roughly 500 pages of text in a single request.
Code Generation
Strong performance on code writing, review, debugging, and explanation tasks.
Streaming
Real-time token-by-token streaming for responsive user interfaces.
When to Use Anthropic
Complex reasoning tasks
Complex reasoning tasks
Claude excels at multi-step logical reasoning, mathematical problem-solving, and tasks that require careful analysis of competing considerations.
Long document processing
Long document processing
With a 200K token context window, Claude can ingest entire codebases, legal documents, or research papers without chunking. This makes it ideal for summarization, Q&A over long texts, and document comparison.
Code generation and review
Code generation and review
Claude produces high-quality code across many languages and is particularly strong at understanding existing codebases, explaining logic, and suggesting improvements.
Careful, nuanced responses
Careful, nuanced responses
Claude is designed to be helpful, harmless, and honest. It performs well in applications that require balanced, well-reasoned responses — such as customer-facing support or advisory bots.
Recommended Models by Use Case
| Use Case | Recommended Model | Reason |
|---|---|---|
| General chatbot | claude-3.5-sonnet | Best all-around performance and cost balance |
| Complex analysis | claude-3-opus | Highest reasoning capability |
| High-volume support | claude-3-haiku | Fast and cost-efficient |
| Document Q&A | claude-3.5-sonnet | Long context + strong comprehension |
| Code assistant | claude-3.5-sonnet | Strong code generation and understanding |
Environment Variable
When self-hosting, set the API key via environment variable:Rate Limits and Pricing
Anthropic applies rate limits based on your account tier and usage level. Nadoo AI handles rate-limit responses gracefully and will surface errors in the chat interface.Troubleshooting
401 Authentication Error
401 Authentication Error
Your API key is invalid or expired. Generate a new key at console.anthropic.com.
429 Rate Limit
429 Rate Limit
You have exceeded your tier’s rate limit. Wait briefly and retry, or check your usage tier for higher limits.
Overloaded Error
Overloaded Error
Anthropic’s API is temporarily at capacity. Nadoo AI will automatically retry with exponential backoff.