What lossless means here
A PDF stores text instructions, images and other supported objects. Some of that information can be stored more compactly. PDFLane attempts a lossless rewrite and checks the candidate output; it does not lower photo resolution to force a smaller download.
Lossy image reduction changes image data, for example by downsampling or recompressing a photograph at lower quality. A result may still look acceptable while containing less detail. That is a different choice from the lossless operation offered here. A claim that two pages “look the same” is not, by itself, proof of losslessness.
A file can already use compact streams and efficient images. Rewriting it may save nothing or add overhead. The amount saved depends on the actual document, not just its starting size. For technical background, the qpdf stream-compression reference describes the general PDF concept; it is not a description of PDFLane’s engine or its options.
Two measured results
We ran two synthetic files through the actual Compress tool and inspected the downloads independently. The first deliberately repeats harmless drawing instructions, so it is an illustrative compressibility test. It is not a typical-customer saving estimate. The second is a tiny valid PDF for which this operation offered no smaller candidate.
| Synthetic file | Input | Download |
|---|---|---|
| Redundant streams | 198,050 bytes | 5,750 bytes |
| No smaller candidate | 420 bytes | 420 bytes |
The smaller output retained six pages, their text, page geometry and the supported internal link. Its image pixels were independently compared. The no-gain download was byte-for-byte identical to its input. Neither result proves that every PDF will behave the same way.
Returning the original can be a correct result. If the candidate is not smaller and the original fits the result limit, PDFLane makes the original available unchanged. A zero-byte saving is useful information, not a reason to invent a better percentage.
Read your own size comparison
- Open Compress PDF, select a supported document and choose Compress PDF.
- Read Original size and Result size. The panel includes actual byte counts as well as compact size labels. Use the bytes when comparing against a receiving service’s exact limit.
- Read the saving or unchanged-original message. Download the result and check text, images, page count, orientation and important links in a reader.
- Use View actual result when available to review the exported PDF. This is optional: a rendering failure does not remove an otherwise available Download.
A compact label such as KiB uses binary units. A receiving website may instead use decimal MB, rounded labels or its own restrictions. A PDFLane download is not a guarantee that another service will accept the file.
Why a result can be refused
A source PDF can be up to 4 MB and 20 pages. Results together must fit within 2,097,152 bytes (2 MiB). Passing the input checks does not guarantee that the exported result will fit. Keep the original and review every download.
If a valid input cannot produce a result within 2 MiB, the operation can reject the output. It does not relax that ceiling or automatically change your images to squeeze the document through. Unsupported encryption, forms, signatures or active content can also prevent admission.
For a shorter document, consider whether you actually need all the pages. Organize PDF can remove unneeded pages from a supported PDF. That changes the content and is a separate decision; keep an original and do not remove required material merely to meet a size target. Review the per-tool input and output limits before another step.
Practical questions
Can it guarantee a PDF below 100 KB?
No. Lossless storage changes cannot promise a fixed percentage or target size. A scan made mostly of already compressed photos may offer little room for a lossless saving.
Will this reduce image resolution?
This tool does not offer image downsampling or a quality slider. If you need a different workflow that changes image data, review its quality trade-offs separately. Do not mistake an unchanged original for a failed download.
The practice examples use synthetic documents and were checked against PDFLane source 0005122 in desktop Chromium. They are examples, not guarantees for every document. See current browser support.
Keep your originals. For a problem with a tool, contact PDFLane without sending sensitive documents or passwords. Read the document privacy notice.