Introducing AI Agent SDK
LeapOCR
Transform any PDF document into structured data with AI-powered OCR. Powered by Google Gemini LLM for exceptional accuracy and intelligent document processing.
Process your first document in minutes
import { LeapOCR } from '@leapocr/sdk';
const client = new LeapOCR({
apiKey: 'your-api-key'
});
const uploadDocument = async (file: File) => {
try {
const response = await client.upload({
file: file,
category: 'invoice',
customInstructions: 'Extract vendor, total, and date'
});
console.log('Job ID:', response.jobId);
return response.jobId;
} catch (error) {
console.error('Upload failed:', error);
}
};
// Usage
const fileInput = document.getElementById('pdf-file');
uploadDocument(fileInput.files[0]);
import { LeapOCR } from '@leapocr/sdk';
const client = new LeapOCR({
apiKey: 'your-api-key'
});
const monitorJob = async (jobId: string) => {
const pollStatus = setInterval(async () => {
try {
const status = await client.getStatus(jobId);
console.log('Progress:', status.progress);
if (status.status === 'completed') {
clearInterval(pollStatus);
const result = await client.getResult(jobId);
console.log('OCR Results:', result.data);
// Process extracted data
result.pages.forEach((page, index) => {
console.log(`Page ${index + 1}:`, page.extractedText);
});
} else if (status.status === 'failed') {
clearInterval(pollStatus);
console.error('Job failed:', status.error);
}
} catch (error) {
console.error('Status check failed:', error);
}
}, 2000);
};
import { LeapOCR } from '@leapocr/sdk';
const client = new LeapOCR({
apiKey: 'your-api-key'
});
// Define a custom schema for invoice processing
const invoiceSchema = {
type: 'object',
properties: {
vendor: { type: 'string' },
invoiceNumber: { type: 'string' },
date: { type: 'string', format: 'date' },
total: { type: 'number' },
tax: { type: 'number' },
lineItems: {
type: 'array',
items: {
type: 'object',
properties: {
description: { type: 'string' },
quantity: { type: 'number' },
unitPrice: { type: 'number' },
total: { type: 'number' }
}
}
}
},
required: ['vendor', 'total', 'date']
};
const processInvoice = async (file: File) => {
const response = await client.upload({
file: file,
schema: invoiceSchema,
category: 'invoice'
});
return response.jobId;
};
import { LeapOCR } from '@leapocr/sdk';
const client = new LeapOCR({
apiKey: 'your-api-key'
});
class DocumentProcessor {
private jobs: Map<string, any> = new Map();
async processBatch(files: File[], category: string = 'default') {
const jobPromises = files.map(async (file, index) => {
try {
const response = await client.upload({
file: file,
category: category,
customInstructions: 'Extract all text and key data points'
});
this.jobs.set(response.jobId, {
filename: file.name,
status: 'processing',
startTime: new Date()
});
return response.jobId;
} catch (error) {
console.error(`Failed to upload ${file.name}:`, error);
return null;
}
});
const jobIds = await Promise.all(jobPromises);
return jobIds.filter(id => id !== null);
}
async monitorBatch(jobIds: string[]) {
const results = [];
for (const jobId of jobIds) {
const result = await this.waitForCompletion(jobId);
results.push(result);
}
return results;
}
private async waitForCompletion(jobId: string) {
return new Promise((resolve) => {
const poll = setInterval(async () => {
const status = await client.getStatus(jobId);
if (status.status === 'completed') {
clearInterval(poll);
const result = await client.getResult(jobId);
resolve(result);
}
}, 1000);
});
}
}
// Usage
const processor = new DocumentProcessor();
const files = Array.from(document.getElementById('file-input').files);
processor.processBatch(files, 'invoice');
Document Types We Process
See how LeapOCR accurately extracts structured data from various document types. Our AI-powered OCR handles complex layouts, tables, and domain-specific content with exceptional precision.
Invoice
Extract vendor details, line items, and totals from invoices
Sample Text Preview
Extracted Data
94.2% confidenceReceipt
Process retail receipts and expense reports automatically
Sample Text Preview
Extracted Data
91.8% confidenceContract
Extract key terms, dates, and parties from legal contracts
Sample Text Preview
Extracted Data
87.3% confidenceMedical Record
Digitize patient records, lab results, and medical reports
Sample Text Preview
Extracted Data
89.5% confidenceFinancial Statement
Process balance sheets, income statements, and financial reports
Sample Text Preview
Extracted Data
96.1% confidenceGovernment Form
Process tax forms, permits, and regulatory documents with precision
Sample Text Preview
Extracted Data
92.7% confidenceOriginal Document
Extracted Data
Live PDF Processing Comparison
See LeapOCR in action with real documents. Select a PDF below to view the original page alongside our AI-extracted markdown. Notice how our OCR maintains formatting, structure, and even handles complex layouts with high accuracy.
Select a document to analyze:
US Constitution - Page 1
Preamble and Article I

Extracted Markdown
THE CONSTITUTION OF THE UNITED STATES
WE THE PEOPLE
We the People of the United States, in Order to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common Defence, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America.
ARTICLE I
Section 1
All legislative Powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives.
Section 2
The House of Representatives shall be composed of Members chosen every second Year by the People of the several States, and the Electors in each State shall have the Qualifications requisite for Electors of the most numerous Branch of the State Legislature.
Our Technology in Action
Hybrid Processing for Maximum Accuracy
Combine text extraction with AI-powered image analysis for maximum accuracy.
[2023-12-15 14:23:45] INFO
OCR job initialized. Starting document processing.
[2023-12-15 14:23:47] ACTION
Extracting text from PDF pages...
[2023-12-15 14:23:50] DECISION
Gemini LLM processing. Accuracy: 94%
[2023-12-15 14:23:52] WARNING
Low quality scan detected on page 3.
[2023-12-15 14:23:55] ERROR
Page 5 processing failed. Retrying...
Monitor Every Step in Real-Time
Track and analyze your document processing with detailed activity logs.
Enterprise-Ready Processing
Scalable document processing that works seamlessly across different platforms.
Features
AI-Powered OCR
Leverage Google Gemini LLM for intelligent document recognition and data extraction.
Learn more >Hybrid Processing
Combines text extraction with image analysis for maximum accuracy on any document type.
Learn more >Structured Data Output
Extract data into custom JSON schemas or predefined formats for seamless integration.
Learn more >Reliable Pipeline
Temporal workflows ensure reliable processing with automatic retries and error handling.
Learn more >Custom Instructions
Define specific processing instructions and document categories for your use case.
Learn more >Real-time Processing
Monitor progress in real-time with parallel processing for faster document handling.
Learn more >Enterprise-Grade OCR with Google Vertex AI
Power your mission-critical document processing with Google's enterprise AI infrastructure. Designed for organizations that need scale, security, and reliability.
Google Vertex AI Integration
Leverage Google's enterprise-grade AI infrastructure for maximum scalability and performance
- 99.9% uptime SLA with Google Cloud reliability
- Auto-scaling to handle millions of documents
- Global edge processing for reduced latency
- Enterprise-grade security and compliance
On-Premise Deployment
Deploy LeapOCR within your private cloud or on-premise infrastructure
- Complete data sovereignty and control
- Air-gapped deployment options available
- Custom security configurations
- Direct integration with your infrastructure
Advanced Security & Compliance
Enterprise-grade security features and compliance certifications
- SOC 2 Type II certified
- GDPR, HIPAA, and PCI DSS compliant
- End-to-end encryption in transit and at rest
- Advanced audit logging and monitoring
Custom AI Models
Domain-specific AI models trained on your document types and requirements
- Models fine-tuned for your specific use cases
- Higher accuracy on domain-specific documents
- Continuous learning from your data
- Priority access to latest AI model updates
Dedicated Support & SLA
Premium support with guaranteed response times and dedicated account management
- 24/7 priority support with 1-hour response
- Dedicated customer success manager
- Custom implementation and training
- Direct access to engineering team
Advanced Analytics & Reporting
Comprehensive analytics dashboard with custom reporting and insights
- Real-time processing metrics and KPIs
- Custom dashboard with role-based access
- Automated compliance and audit reports
- API for integrating with BI tools
Ready to Scale Your Document Processing?
Join hundreds of enterprises already processing millions of documents with LeapOCR. Get a custom demonstration and implementation plan tailored to your specific requirements.
Trusted by enterprises worldwide
Testimonials
Sarah Chen
FinanceFlow Solutions
LeapOCR has transformed our invoice processing workflow. The accuracy with Gemini LLM is outstanding, and we've reduced manual data entry by 90%.
Michael Rodriguez
LegalDoc Systems
The hybrid OCR processing handles our mixed document types perfectly. From scanned contracts to digital forms, everything gets processed accurately.
Emma Thompson
DataBridge Corp
The API integration was seamless, and the custom schemas feature allows us to extract exactly the data we need for our CRM system.
James Wilson
ProcessFlow Inc
Real-time processing monitoring and the reliability of Temporal workflows gives us confidence in our document processing pipeline.
Dr. Lisa Park
MedTech Analytics
LeapOCR's custom instructions feature lets us process specialized medical documents with domain-specific requirements. Game-changing for healthcare.
Robert Kim
ScaleData Solutions
The parallel processing capability handles our high-volume document batches efficiently. We've cut processing time from hours to minutes.
Alex Foster
ComplianceFirst
The analytics dashboard provides valuable insights into our document processing operations. Perfect for compliance reporting.
Rachel Martinez
Enterprise Solutions
LeapOCR's enterprise features and SLA guarantees give us the confidence to process mission-critical documents at scale.
David Chang
IntegrateNow
The structured JSON output format makes it incredibly easy to integrate OCR results into our existing systems and workflows.
Pricing
Simple & Scalable
One credit = 1 page. A fraction of the cost of legacy OCR providers. All plans include API access, dashboard UI, custom prompts, team features, and private processing.
Best for high-speed text extraction
Clean OCR → Markdown/JSON • Simple docs
- Fast document processing
- Markdown & JSON output
- API & dashboard access
- Custom schema builder
- Team & workspace support
- 50 pages free trial
Best for invoices, receipts, and forms
Logical extraction • Tables • Invoices
- Chain-of-thought reasoning
- Semantic field extraction
- Enhanced table accuracy
- Custom instructions
- Complex document handling
- All Standard features
Gemini 2.5 Pro
Complex layouts • Multilingual docs
- Advanced layout handling
- Multilingual support
- Deep prompt-guided output
- Superior table extraction
- Complex document reasoning
- All Pro features
Google Vertex AI + Enterprise
Custom models • On-premise • Advanced security
- Google Vertex AI infrastructure
- 99.9% uptime SLA guarantee
- On-premise deployment options
- Custom AI model training
- SOC 2, GDPR, HIPAA compliance
- 24/7 dedicated support
- Advanced analytics & reporting
- Priority feature development
Community
We're grateful for the amazing open-source community that helps make our project better every day.
Ready to transform your documents with AI?