Turn PDFs into structured data at scale. Powered by frontier open-weight OCR models with a type-safe TypeScript SDK.

• Best-in-class OCR - PaddleOCR-VL-0.9B for accurate text extraction
• Structured extraction - Define schemas, get JSON back
• Built for scale - Queue-based processing for thousands of documents
• Type-safe SDK - Full TypeScript support with React hooks
• Real-time updates - WebSocket notifications for job progress (client-side)
• Self-hostable - Run on your own infrastructure

npm install ocrbase

# .env
OCRBASE_API_KEY=sk_xxx

Important: Jobs are processed asynchronously. Use WebSocket for real-time status updates.

import { createClient } from "ocrbase";

const { parse, extract, ws } = createClient({
  apiKey: process.env.OCRBASE_API_KEY,
});

// Start parsing (returns job with status "pending")
const job = await parse({ file: document });

// Subscribe to real-time updates via WebSocket
ws.subscribeToJob(job.id, {
  onComplete: (result) => {
    console.log(result.markdownResult); // string - the parsed markdown
  },
});

Use the useJobSubscription hook for real-time updates:

import { useParse, useJobSubscription } from "ocrbase/react";

function DocumentParser() {
  const [jobId, setJobId] = useState<string | null>(null);
  const parse = useParse();

  const handleFile = (file: File) => {
    parse.mutate({ file }, { onSuccess: (job) => setJobId(job.id) });
  };

  const { status, job } = useJobSubscription(jobId!, { enabled: !!jobId });

  if (status === "completed" && job) {
    return <pre>{job.markdownResult}</pre>;
  }

  return (
    <div>
      <input type="file" onChange={(e) => handleFile(e.target.files![0])} />
      {status && <p>Status: {status}</p>}
    </div>
  );
}

Best practice: Parse documents with ocrbase before sending to LLMs. Raw PDF binary wastes tokens and produces poor results.

Pattern: Parse on client with WebSocket → send markdown to your API → call LLM.

See SDK documentation for React hooks and advanced usage.

See Self-Hosting Guide for deployment instructions.

Requirements: Docker, Bun

MIT - See LICENSE for details.

For API access, on-premise deployment, or questions: adammajcher20@gmail.com