PDF without Chrome: an alternative to PuppeteerSharp and Playwright

PuppeteerSharp and Playwright make PDFs by driving a real Chromium browser. That gives exactly what Chrome prints, but it means downloading and patching a browser of about 600 MB, keeping it running, and paying for a browser tab for every document. PdfPecker converts HTML to PDF in your own process with its own engine written in C#, so there is no browser to install and a one-page document takes about a millisecond. It is free under the MIT license.

dotnet add package PdfPecker

Replace the browser code

From PuppeteerSharp

// Before
await new BrowserFetcher().DownloadAsync();
await using var browser = await Puppeteer.LaunchAsync(new LaunchOptions { Headless = true });
await using var page = await browser.NewPageAsync();
await page.SetContentAsync(html);
byte[] pdf = await page.PdfDataAsync(new PdfOptions
{
    Format = PaperFormat.A4,
    PrintBackground = true,
    MarginOptions = new MarginOptions { Top = "20mm", Bottom = "20mm" },
});

// After
using PdfPecker;

byte[] pdf = HtmlToPdf.Convert(html, new PdfOptions
{
    PageSize = PageSize.A4,
    Margins = PageMargins.FromMillimeters(20, 15, 20, 15),
});

Both libraries have a class called PdfOptions, so in a project that still uses PuppeteerSharp elsewhere, write PdfPecker.PdfOptions.

From Playwright

// Before
using var playwright = await Playwright.CreateAsync();
await using var browser = await playwright.Chromium.LaunchAsync();
var page = await browser.NewPageAsync();
await page.SetContentAsync(html);
byte[] pdf = await page.PdfAsync(new PagePdfOptions { Format = "A4", PrintBackground = true });

// After
byte[] pdf = await HtmlToPdf.ConvertAsync(html);

A4 is the default page size. HtmlToPdf is static and thread-safe, so there is no browser to share between requests and nothing to dispose or restart.

Headers and footers

Chrome's headerTemplate and footerTemplate use <span class="pageNumber"> and <span class="totalPages">, and need displayHeaderFooter. In PdfPecker, use FooterText = "{page} of {pages}" for a simple line, or put the header or footer in the page itself:

<footer data-sticky-pdf-footer height="12mm">Page {page} of {total-pages}</footer>

Unlike Chrome's templates, these elements use the page's own stylesheets, fonts and images. @page margin boxes with counter(page) work too. See Paged media.

Options side by side

What changes

• JavaScript is not run, so content that scripts build in the browser has to be in the HTML before conversion. Templates rendered on the server (Razor, Scriban, string building) work as they are.
• Nothing is downloaded. Images and stylesheets come from files under BaseDirectory, data: URLs, or your own ResourceResolver for http URLs, so a template can't make the server fetch other addresses.
• PdfPecker supports a large part of modern CSS, including flexbox, grid, variables and @page rules, but not all of what Chrome does. Limitations lists the gaps.

Speed and memory

On the same 4-core Linux VM, a one-page invoice took 153 ms with PuppeteerSharp, 194 ms with Playwright and 0.86 ms with PdfPecker. Peak memory was 829 MB and 967 MB for the browser-based converters and 90 MB for PdfPecker, and in parallel PdfPecker made 818 documents per second against 8 to 11. The full table is on the Performance page.

When to keep Chrome

PuppeteerSharp and Playwright are free and print exactly what Chrome shows, so they remain the right tool for pages that build themselves with JavaScript, for live websites and for screenshots. Compare shows what each converter can do.