This document explains the Extract Data node, which uses AI to pull specific information from text content.
Required Fields
- Text: Content to extract data from (documents, scraped website content, etc.)
- Data Fields to Extract: Define what you want to extract:
- Name: Label for the data (e.g., “location”)
- Type: Format of the data (text/number/boolean)
- Description: Help the AI understand what to extract
Optional Fields
- Extract List: Enable to get multiple items instead of single values
- Additional Context: Extra information to guide the extraction
- Temperature: Controls AI creativity (0-1)
- 0: More focused, consistent
- 1: More creative, varied
- Cache Response: Save responses for reuse
The node allows you to configure certain parameters as dynamic inputs. You can enable these in the “Configure Inputs” section:
-
Extract List?: Boolean
- true/false to enable/disable list extraction
- When enabled, extracts data as a list of items
-
Additional Context: String
- Extra information to guide the extraction process
- Example: “The text contains company names and their founding years”
-
model_preference: String
- Name of the AI model to use
- Accepted values: “Claude 4.5 Sonnet”, “Claude 4.5 Haiku”, “GPT-5”, “GPT-4.1”, etc.
-
Cache Response: Boolean
- true/false to enable/disable response caching
- Helps reduce API calls for identical inputs
-
Temperature: Number
- Value between 0 and 1
- Controls extraction consistency
When enabled as inputs, these parameters can be dynamically set by previous nodes in your workflow. If not enabled, the values set in the node configuration will be used.
AI Model Fallback
Under Show More Options, configure automatic fallback when your selected AI model is unavailable. Fallback is enabled by default.
When an error occurs (rate limits, provider outages, timeouts), the system retries based on severity, then falls back to the next model. Fallback models are always from different providers for true redundancy.
| Error Type | Retries Before Fallback |
|---|
| Rate Limit | 2 |
| Provider 5xx | 1 |
| Network Error | 0 (immediate) |
| Timeout | 1 |
- Expert → Claude Opus 4.5 → Gemini 3 Pro → GPT-5.2
- Fastest → Gemini 3 Flash → Claude Haiku 4.5 → GPT-4.1
- Recommended → Claude Sonnet 4.5 → Gemini 3 Flash → GPT-5.2
Override: Enable to manually select up to 2 fallback models with drag-and-drop priority.
Disabling fallback means your node will fail if the primary model is unavailable.
Node Output
- Extracted Data Fields: Single value or list based on your settings
Node Functionality
The Extract Data node:
- Analyzes text using AI
- Finds specific information
- Returns structured data
- Handles single or multiple items
- Supports various data types
Available AI Models
| Tier | Models |
|---|
| Expert | GPT-5.2, GPT-5.1, GPT-5, OpenAI o3, Claude 4.5/4.1/4 Opus, Claude 3.7 Sonnet Thinking, Gemini 3 Pro, Grok 4 |
| Advanced | GPT-4.1, OpenAI o4-mini, Claude 4.5/4/3.7 Sonnet, Gemini 2.5 Pro, Grok 3, Perplexity Sonar Pro, LLaMA 3 405B |
| Standard | GPT-4.1 Mini/Nano, GPT-5 Mini/Nano, Claude 4.5 Haiku, Gemini 3/2.5 Flash, Grok 3 Mini, DeepSeek V3/R1, Mixtral 8x7B |
| Special | Auto-Select, Azure OpenAI (requires credentials) |
Auto-Select uses third-party routing to choose models based on cost and performance. Not ideal when consistent behavior is required.
Example Use Cases
- Contact Information:
Extract: Email, Phone, Address
From: Company websites or documents
- Product Details:
Extract: Price, Features, Specifications
From: Product descriptions
- Data Extraction from Documents:
Extract: Date, invoice amount, vendor, address
From: Financial documents or invoices
Important Considerations
- Expert models (OpenAI o3) cost 30 credits, advanced models (GPT-4.1, Claude 3.7 & Grok 4) cost 20 credits, and standard models cost 2 credits per run
- You can drop the credit cost to 1 by providing your own API key under the credentials page
- Enable “Extract List” when you need multiple items
- Be specific in your data descriptions for accurate outputs
In summary, the Extract Data node is your tool for pulling structured information from unstructured text, whether you need single values or lists of data. 
