Navigation

Mobile Docs

Thinking

Extended reasoning with step-by-step thought process.

Uses extended thinking to show the model's internal reasoning process before producing a final answer. Supports Anthropic Claude (claude-sonnet-4-6, claude-opus-4) with native extended thinking tokens, and OpenAI o-series reasoning models (o3, o4-mini). The thinking process and final response are both available as outputs.

Core Blocksthinking

Usage

01Add the block to your workflow and connect it to the upstream step.
02Configure any required credentials or tokens in the inputs.
03Fill in required inputs and optional parameters for the run.
04Run a test execution, inspect outputs, and iterate before deploying.
05Deploy the thinking block with monitoring enabled in production.

Inputs (UI)

Model

dropdown

Layout: half

Options: Claude Sonnet 4.6, Claude Opus 4, Claude Sonnet 4, OpenAI o4-mini

API Key

short-input

Placeholder: sk-ant-... or sk-...

Layout: half

Thinking Budget (tokens)

slider

Layout: half

Condition: model = ["claude-sonnet-4-6","claude-opus-4-20250514","claude-sonnet-4-20250514"]

Show Thinking Process

switch

Layout: half

Prompt

long-input

Placeholder: What should the model reason about? Ask a question or describe a task...

Layout: full

System Prompt (optional)

long-input

Placeholder: Optional instructions to guide the model's reasoning style...

Layout: full

Hidden by default

Inputs (API)

model

string

Required

Model to use for extended thinking

apiKey

string

Required

API key for the selected provider

prompt

string

Required

The question or task to reason about

systemPrompt

string

Optional

Optional system prompt

thinkingBudget

number

Optional

Extended thinking token budget (Anthropic only)

showThinkingProcess

boolean

Optional

Whether to include reasoning steps in the output

Outputs

Primary response type:

{
  "thinkingContent": "string",
  "response": "string",
  "model": "string",
  "tokens": "json",
  "acknowledgedThought": "string"
}

Tool Access

thinking_tool