The Convert PDF to Grayscale tool runs entirely inside your current browser tab and turns a single local PDF up to 25 MB into a brand-new raster PDF whose every visible pixel is replaced with a luminance value computed as round(0.2126 × red + 0.7152 × green + 0.0722 × blue). Nothing is uploaded to a server, nothing requires an account, and the only file you need to bring is the one already saved on your computer. The output is not a recolored copy of the original PDF — it is a fresh document built from PNG page images placed back onto pages of the original size, which means it preserves the look of the page but does not preserve the selectable text layer, live links, form fields, vector paths, original metadata, or the source's compression. Treat the result as a flattened visual monochrome copy: excellent for a print preview, a quick monochrome upload, or a reference PDF, but not a replacement for a professionally edited source. That single sentence — "local, browser-only, raster output" — is the whole mental model behind getting started.
Most people who land on this guide have a color PDF and a clear reason to want it in grayscale: a cheaper print run, an upload form that only accepts monochrome files, an internal reference document where color is noise, or a quick preview to share. Before you click anything, the real "getting started" work happens on your end — opening the right file in the right browser tab, checking the size, and understanding what the tool will and will not keep.
Check Your File Against the Limits Before You Open It
The tool enforces five hard limits before any page canvas is even allocated, so a file that pushes past any of them will simply not convert:
| Limit | Maximum | Why it matters |
|---|---|---|
| File size | 25 MB | One local PDF only |
| Page count | 40 pages | Long jobs are refused |
| Pixel side | 12,000 px on either side of one page | Prevents unbounded buffers |
| Per-page pixels | 40 megapixels | Caps single-page render cost |
| Total pixels | 100 megapixels across the document | Caps total memory use |
If your PDF is over 25 MB, trim it first with a tool that can split or compress locally — for example, the PDF page counter to confirm size before you commit. Password-protected, malformed, or signed PDFs may fail outright because the tool does not bypass passwords and does not validate signatures; unlock the file or remove the signature with a separate tool before retrying.
The 1.5 render scale is also worth knowing: each page is rendered at 1.5× the source's pixel dimensions, then placed back onto a page whose size matches the original. The visual detail is locked at that scale and is not adjustable. If the original page is unusually small, the resulting raster image will still be small; if the original page is very large, you may run into the per-page or document-wide megapixel caps even when the file is well under 25 MB.
Walk Through the Three-Step Conversion
Once your file is ready, the actual conversion is short. Open the Grayscale PDF Converter in a fresh browser tab so that the conversion has the full tab memory budget, then run the three steps:
- Choose one local PDF up to 25 MB from your computer and load it into the tool.
- Select Convert to grayscale and wait while every bounded page is rendered locally, pixel by pixel, in your browser.
- Review the raster PDF and download it when the flattened pages meet your needs.
The rendering step is the slowest one. Starting a new conversion cancels the previous job, and any canvases, PDF worker tasks, and temporary download URLs are released once the job ends or is cancelled, so you are not leaving large buffers sitting in memory if you walk away.
How the Grayscale Math Actually Works
After the browser renders a page with PDF.js, the tool walks every pixel of that page and replaces the red, green, and blue channels with the same luminance value. The formula is the Rec. 709 / sRGB luminance weights, rounded to the nearest 8-bit value:
- Red contribution: 0.2126
- Green contribution: 0.7152
- Blue contribution: 0.0722
For a single pixel of pure red at (255, 0, 0), the math is 0.2126 × 255 + 0.7152 × 0 + 0.0722 × 0 = 54.213, which rounds to 54. That single value is then written into the red, green, and blue channels, so the pixel becomes (54, 54, 54) — visibly gray, and visibly darker than the same shade coming from pure green (which would land around 182) or pure blue (around 18). The same per-pixel arithmetic means black stays at (0, 0, 0), white stays at (255, 255, 255), and any input gray stays at its own gray.
This explicit arithmetic — the same pattern used by the CanvasRenderingContext2D getImageData and putImageData round-trip — is the reason the tool does not lean on a CSS filter: grayscale(100%) shortcut. The CSS path is not a dependable default in Safari, and the explicit arithmetic makes the result testable: feed in pure red, pure green, pure blue, black, and white, and you can predict the output to the nearest integer.
What the Output Keeps and What It Drops
The clearest way to think about the new file is as a "page-sized photo of every page." That analogy is exact: each output page is a PNG image wrapped in a PDF page of the original dimensions, with the original selectable text, links, form fields, annotations, vector paths, layers, original image encodings, accessibility structure, metadata, and the source's compression not carried over.
| Property | Source PDF | Output from the tool |
|---|---|---|
| Page size and orientation | As authored | Preserved |
| Visual look (color to monochrome) | Full color | Grayscale via fixed formula |
| Selectable text layer | Yes | No — flattened into image |
| Live links and form fields | Yes | No |
| Vector paths and layers | Yes | No — rasterized |
| Original metadata and compression | Yes | No — replaced by PNG pages |
| Output file size | Varies | Often larger than source |
If any of the dropped properties matter — for example, you need the text selectable because the file feeds into a translation tool, or you need PDF/A conformance for archival — use a professional PDF editor instead. The browser tool's job is the quick, flattened copy, and it tells you that explicitly: it does not claim archival or legal conformance, does not bypass passwords, and does not validate signatures.
The output is often larger than the source because the source's fonts, vectors, and image compression are replaced by uncompressed-by-comparison PNG page images. If size matters more than visual fidelity, the right next step is to run the new file through a size-targeted compressor in your browser rather than re-rendering it differently — the getting started guide for compressing a PDF to a target size walks through that workflow end to end.
When a Browser Grayscale Copy Is the Wrong Tool
A quick visual monochrome copy is the right answer when the file is a print preview, an internal reference document, a draft that needs to be sent without color noise, or an upload that explicitly asks for grayscale. It is the wrong answer when you need any of the following from the final file:
- Selectable, searchable text that can be copied or fed into a translation or extraction tool
- Live links that still navigate when clicked
- Form fields that can still be completed
- Vector graphics that scale cleanly at any zoom level
- Accessibility tags, logical reading order, or PDF/A conformance
- Original metadata, signatures, or print color management
- A file size that is guaranteed to be smaller than the source
For any of those, stay in a professional PDF editor that preserves the underlying structure of the original document. The browser grayscale converter is the quick path when structure does not need to survive.
Confirm the Result Before You Share the File
A few seconds of checking the downloaded file saves a lot of cleanup later. Open the grayscale PDF, flip through each page that had color content in the source, and confirm that the contrast still reads well — color photographs sometimes look flatter after grayscale, and a page that was balanced for color may need its source adjusted rather than its output. If a page looks wrong, the most common cause is that the source page had an unusual mix of colors or a low-contrast palette that grayscale flattens further; the next step is to fix the source and re-export, not to try to re-grayscale the grayscale file. Also confirm the page count matches the source: the tool reopens the output to verify the page count before letting you download, but a manual count is still cheap insurance for a 40-page job.
Keep your source file alongside the new one until you are sure the grayscale version is what you want to share. The tool runs locally, never uploads your PDF, and does not keep a copy on any server, so the only grayscale file you will have is the one you actually downloaded.
Related reading: How to Get Started Removing PDF Annotations.