PDF Merge & Split
Explore PDF conversion and page tools with clear processing labels. Browser conversions are available; server merge, split and editing are currently disabled.
Learn how structural PDF compression differs from image downsampling. Server PDF compression is currently disabled; explore available local conversion tasks.
Loading your private workspace…
Recompress uncompressed PDF streams and compare the resulting size.
pikepdf compress_streams, recompress_flate and object streams; preserve page count, reopen/validate and compare bytes.
One unsigned, unencrypted PDF up to 10 MiB and 200 pages; output up to 30 MiB. Structural lossless compression only; no image downsampling, so scans may not shrink. Active content, annotations, forms, attachments and metadata removed. A larger sanitized result is disclosed, never presented as successful size reduction.
A PDF compressor can optimize document structure or reduce embedded image quality, and those are different operations. Mayexa's compressor uses structural, lossless processing rather than scan-image downsampling. Server compression is currently disabled on the live deployment, so this guide explains its scope without promising an available upload workflow.
The current workspace exposes separate browser-local conversion tasks, but those should not be confused with the disabled server compression service. Signed or encrypted PDFs are not supported for editing. Sanitization removes active content and other unsupported objects; it is not a digital-signature preservation service. Do not use a crop or a rendered preview as evidence of secure redaction.
No Mayexa account is required. Review the workspace's processing labels and limits before entering data.
No. Lossless stream and object compression cannot guarantee savings, especially for image-heavy scans. The tool reports actual byte sizes and clearly identifies a result that is not smaller.
No. This compression preserves image resolution. Maximum additionally recompresses Flate streams; neither preset is an image-quality or lossy downsampling control.
Inputs are deleted after processing or failure. Outputs expire after one hour with periodic cleanup, or Clear removes a completed job immediately. Active content, annotations, forms, attachments and metadata are removed; signed/encrypted inputs are refused.
Work through a practical example and understand how to interpret the result before relying on it.
Why a PDF Compressor May Not Make a Scanned Document SmallerOrganize, edit and convert documents. Choose a task to see exactly what happens.
Lossless structural compression.
Only Upload & process sends these files to our private processor. Inputs are removed after processing; results expire within one hour. Clear deletes completed jobs immediately. Annotations, links, forms, attachments and metadata are removed. Encrypted and signed PDFs are refused.
Private worker unavailable. Upload is disabled; local conversions still work.
Drag files here, or choose them from your device.
Up to10 MiB per file ·25 MiB total ·one file · PDFs up to 200 pagesChoose a task, then add documents and review their first-page previews. Reorder files for merging or images-to-PDF. Configure task-specific options. Local tasks never upload your documents. Private tasks upload only when you choose Upload & process. Download the result, then clear files and completed jobs.
Page ranges are one-based:1-3,5. Reorder requires every page exactly once. Keep originals: server processing deliberately removes interactive content and metadata. Preview images are first pages, not a certification of document fidelity.
Server PDF tasks:5files for merge; one for edits/compression,10 MiB each,25 MiB total,200 pages. Protected/signed files refused. Private inputs removed after processing; results expire within 1 hour or on Clear. Local tasks:200PDFpages,2 million extracted characters,50 rendered pages per ZIP,100 MP cumulative,30 MiB output and2-minute operation timeout. Images:JPEG/PNG/staticWebP within 20 MP/10,000 px. DOCX text import: bounded 5 MiB ZIP, text-only. Text PDFs: Latin/WinAnsi only,100,000 characters/100 pages. Original metadata and profiles may change.
| Working here | Merge, extract, split ZIP, remove, reorder, rotate, crop, visible watermarks, numbers, lossless compression; images/text/DOCX text to PDF; PDF images/text/Markdown/DOCX text and text comparison. |
|---|---|
| Needs additional infrastructure | OCR of scans, faithful Word/Excel/PowerPoint conversion, PDF/A certification, repair guarantees, digital signatures, password management, true redaction and AI/translation are not available. These require dedicated engines, validation and sometimes credentials; no fake conversion or third-party document upload is used. |
Crop, stamps and text comparison are not secure redaction or signature verification. Downloads are newly generated documents. No documents, extracted text or capabilities are saved in browser storage or analytics. Other browser/device/production limits need separate validation.