Claude Model Block
The Claude Model Block enables using Claude models for text generation, chat completions, and structured outputs within your workflows.
⚠️
API Key Required. Get your API key from the Claude Console (opens in a new tab).
Available Models
| Model | Best For |
|---|---|
| Claude Opus 4 | Complex analysis, math, coding, extended thinking |
| Claude Sonnet 4 | Balanced performance and speed |
| Claude Haiku 3.5 | Fast responses, lightweight tasks |
Configuration
- Model: Select from available Claude models
- System Prompt: Set the AI assistant's behavior and persona (supports
{{variables}}) - User Prompt: The message/question to send (supports
{{variables}}) - Temperature: Controls randomness (0 = deterministic, 1 = creative)
- Max Tokens: Maximum response length (up to 8192)
- Top P: Nucleus sampling parameter (0-1)
- Output Format: Define structured JSON output schema
- Response Variable Mapping: Map response fields to workflow variables
Output
{
"statusCode": 200,
"data": {
"completion": "The AI-generated response text...",
"usage": {
"inputTokens": 150,
"outputTokens": 300,
"model": "claude-sonnet-4-20250514",
"organizationName": "Claude"
}
}
}Use Cases
- Generating content, summaries, and reports
- Analyzing data and extracting insights
- Building conversational AI workflows
- Structured data extraction with output format schemas
- Code generation and review automation