HTML to PDF conversion using OpenhtmlTopdf with ByteArrayOutputStream

Question

I am trying to convert HTML to PDF as encoded string. I am using openhtmltopdf library. I don't want to create a new file in users environment, so I am using ByteArrayOutputStream.

Following is my code:

Document document = Jsoup.parse(html, "UTF-8");
document.outputSettings().syntax(Document.OutputSettings.Syntax.xml);
document.outputSettings().prettyPrint(false);
// File outputpdf = new File("output.pdf");
try (ByteArrayOutputStream os = new ByteArrayOutputStream()) {
    PdfRendererBuilder pdfRendererBuilder = new PdfRendererBuilder();
    pdfRendererBuilder.toStream(os);
    pdfRendererBuilder.withW3cDocument(new W3CDom().fromJsoup(document), "/");
    pdfRendererBuilder.run();
    // os.writeTo(new FileOutputStream(outputpdf));
    byte[] encoded = java.util.Base64.getEncoder().encode(os.toString().getBytes());
    String encodedString = new String(encoded);

I used an online base64 string to PDF decoder and generated PDF while testing. My PDF is coming as empty. When I replaced the ByteArrayOutputStream with FileOutputStream(<fileName>). It is creating a proper PDF file and also when I decode the string it is coming correct.

What am I missing in ByteArrayOutputStream?