Yes — the Grayscale PDF Converter turns a PDF into a grayscale file entirely in your current browser tab, never uploads the source, and applies the explicit per-pixel formula 0.2126R + 0.7152G + 0.0722B to every rendered pixel. The tool opens your local PDF using PDF.js, draws each bounded page on an in-memory canvas at a 1.5x scale, and replaces the red, green, and blue channels of every pixel with a single luminance value calculated from those three weights. That grayscale canvas is encoded as a PNG, then placed back at the original page dimensions inside a new PDF built with pdf-lib. Because every step — page rendering, pixel math, PNG encoding, and final assembly — runs in the same tab that loaded the page, the source file never travels across the network or onto a remote server. The converter is also explicit about the format of its output: the result is a raster PDF rather than a recolored copy of your original, so the page is clear about which document properties survive the trip and which ones get flattened to a pixel image. The sections that follow walk through the exact click path, the math used for each pixel, the limits enforced before any canvas is allocated, and the situations where a desktop PDF editor is the better choice.
Running the tool only takes a few clicks, and you can inspect the output in a preview before you download anything. The steps below work on any modern desktop browser. If you start a new conversion while an old one is still running, the previous job is canceled automatically and its canvases, PDF worker tasks, and temporary download URLs are released so they do not pile up inside the tab.
Convert a PDF to grayscale in your browser
The Grayscale PDF Converter walks you through a three-stage flow: choose a file, run the conversion, then review and download. The steps below come straight from the verified operating steps for the tool, so they reflect the controls and limits you will actually see on screen.
- Open the Grayscale PDF Converter at /pdf/grayscale-pdf/ in your browser.
- Click the file picker and choose one local PDF that is at most 25 MB; the file is read with the browser File API and never uploads, because the page uses the PDF.js library to parse it locally.
- Select "Convert to grayscale" and wait while every bounded page is rendered locally on a 1.5x canvas, recolored through the per-pixel formula, and reassembled into a new PDF.
- Review the raster PDF once rendering is complete; if a page looks wrong, open the source PDF in another tab to confirm it is not already corrupted, then start a fresh job — the previous run is canceled automatically and its temporary resources are released.
- Download the grayscale PDF once rendering is done, open it in any PDF viewer, and check the page count, page dimensions, and visual content before you share the file.
The per-pixel grayscale transform explained
You can verify the exact output by sampling known input colors against the tool, because the converter deliberately avoids the CSS canvas filter property — which Safari does not expose as a dependable default — and instead reads pixels directly through the Canvas 2D getImageData call. The W3C Filter Effects specification defines the same grayscale equivalent with the same weights inside filter:grayscale(), which is why the converter's numbers line up with what a stylesheet would produce on browsers that do support it.
- The formula. For each pixel returned from the canvas, the tool computes round(0.2126 × R + 0.7152 × G + 0.0722 × B) and writes that integer into all three channels.
- The weights. 0.2126, 0.7152, and 0.0722 are the BT.709 luminance coefficients; they roughly match human brightness perception, so greens contribute more than reds, and reds more than blues.
- The result on the primary colors. A pure red sample at (255, 0, 0) becomes (54, 54, 54) because 0.2126 × 255 = 54.213, which rounds down; a pure green (0, 255, 0) becomes (182, 182, 182) because 0.7152 × 255 ≈ 182.376; and a pure blue (0, 0, 255) becomes (18, 18, 18) because 0.0722 × 255 ≈ 18.411. You can confirm each of these against any test canvas in DevTools.
- The storage step. The recolored pixels are encoded as PNG, then placed on an output canvas at the original page size in points so the output PDF keeps the source's page box even though the source was rendered at 1.5x detail.
Using the arithmetic transform rather than a CSS filter makes the behavior testable for black, white, gray, and the primary colors, and it sidesteps Safari's tendency to leave filter property defaults undefined.
What the raster output keeps and what it flattens
Because the conversion is visual rather than structural, several PDF properties that you might expect to survive do not. The table below shows which document features pass through unchanged and which ones are flattened to a pixel image, so you can decide before you run a long document through the tool.
| Document feature | Status after grayscale conversion |
|---|---|
| Original page dimensions in points | Preserved |
| Render detail | Rendered at 1.5x then placed at original page size |
| Selectable text layer | Flattened to PNG image |
| Hyperlinks | Removed |
| Form fields | Removed |
| Annotations and comments | Removed |
| Vector paths and drawings | Flattened to pixels |
| Original image encodings (JPEG, JBIG2, etc.) | Replaced with PNG |
| PDF/A compliance | Not preserved |
| Accessibility tags and structure tree | Removed |
| Document metadata (title, author, etc.) | Not preserved |
| Original compression and file size hints | Replaced; output is often larger |
If any row in the right-hand column matters for your use case — for example you need search-and-select text or a signed PDF to stay valid — this table tells you to look at a different tool rather than the browser converter.
File, page, and pixel limits enforced before rendering
The tool enforces its limits before allocating a single output canvas, which prevents a malformed or oversized document from exhausting browser memory. Every run is checked against the same set of budgets:
| Limit | Value | Why it exists |
|---|---|---|
| Source file size | 25 MB max | Keeps the full file inside the browser's PDF.js worker |
| Page count | 40 max | Bounds the number of pages that have to be rendered in one tab |
| Pixels per side on one rendered page | 12,000 max | Avoids allocating an unreasonably large canvas |
| Megapixels per page | 40 max | Caps the per-page image size after the 1.5x scale |
| Total megapixels across the document | 100 max | Bounds the cumulative image data held in memory |
| Render scale | 1.5x | Trades fine detail for predictable memory use |
If your PDF exceeds any of these values, the converter will not start the job. Password-protected, signed, malformed, or unusually complex PDFs may also fail; this tool does not bypass passwords, validate signatures, or claim archival or legal conformance.
When to skip a browser tool and reach for a desktop editor
A browser converter cannot do the work of a professional PDF editor when you need the document to stay interactive, compliant, or signed. The Grayscale PDF Converter is meant for a quick visual monochrome copy, a print preview, an upload requirement, or a reference document where flattened pages are acceptable. Switch tools if any of the following must survive the conversion:
- Preserve selectable text, an accessible reading order, or PDF/A conformance for an archive or regulator.
- Keep live hyperlinks, form fields, e-signatures, or annotations intact for ongoing collaboration.
- Maintain vector drawings, layers, transparency groups, or print color management for a press workflow.
- Control compression, final file size, or color profile embedding for a delivery deadline.
- Open a password-protected source PDF, since this tool does not bypass passwords or accept credentials.
Before distributing the grayscale PDF, open a few representative pages alongside the original so you can sanity-check the contrast and confirm fine type, line art, or color-coded charts still read clearly. If a page looks muddy or too dark, regenerate it from a higher-resolution source, or move to a vector-preserving workflow in a desktop editor instead.