Skip to main content
The simple() strategy performs extraction in a single pass without chunking. Use this for documents small enough to fit within your model’s context window.

​
Usage

import { extract, simple } from 'struktur';
import { openai } from '@ai-sdk/openai';

const result = await extract({
  artifacts,
  schema,
  strategy: simple({
    model: openai('gpt-4o'),
  }),
});

​
Configuration

​
model
LanguageModel
required
The AI SDK language model to use for extraction.
​
outputInstructions
string
Additional instructions to guide the model’s output format or behavior.
​
execute
function
Custom retry executor function. Defaults to runWithRetries.
​
strict
boolean
Enable strict mode for structured output validation. Defaults to false.

​
When to use

  • Your document fits within the model’s context window
  • You have a single page or small document
  • You want the fastest extraction with lowest token usage
  • You don’t need parallel processing

​
Trade-offs

Advantages:
  • Fastest extraction time
  • Lowest token usage
  • Simplest configuration
  • No merge complexity
Limitations:
  • Cannot handle large documents
  • No parallel processing
  • Fails if input exceeds context window

​
Performance characteristics

The strategy estimates 3 steps:
  1. Prepare
  2. Extract
  3. Complete

​
Example with options

import { extract, simple } from 'struktur';
import { anthropic } from '@ai-sdk/anthropic';

const result = await extract({
  artifacts: [singlePageArtifact],
  schema: productSchema,
  strategy: simple({
    model: anthropic('claude-3-5-sonnet-20241022'),
    outputInstructions: 'Focus on extracting prices in USD',
    strict: true,
  }),
  events: {
    onStep: ({ step, total, label }) => {
      console.log(`Step ${step}/${total}: ${label}`);
    },
  },
});

Build docs developers (and LLMs) love

Get started for freeTalk to us