7-Zip Archive

7z's default LZMA/LZMA2 algorithm uses a much larger compression dictionary and range coding instead of ZIP's Huffman coding, which generally finds more redundancy and encodes it more efficiently, at the cost of taking longer to compress.

Yes — the 7-Zip application is licensed under the LGPL and the underlying LZMA SDK is in the public domain, which is a key reason DAM pipelines can build 7z support into server-side tooling without the licensing questions that apply to RAR.

Neither operating system supported 7z out of the box for most of its history. Windows 11's File Explorer gained native 7z extraction in 2023 through the open-source libarchive library, and the 24H2 update added creating 7z archives too, but Windows 10 and earlier Windows 11 builds still need a separate tool. macOS's built-in Archive Utility never added 7z support, so opening one there still requires Keka, The Unarchiver, or command-line p7zip.

Yes. 7z encrypts file contents with AES-256, using a key derived from the password by hashing it with SHA-256 roughly 500,000 times to slow brute-force attempts. Checking the 'encrypt filenames' option goes further and encrypts the archive's header, which stores the file list—so anyone without the password sees no filenames, folder structure, or file count at all. Standard ZIP encryption cannot do this; it always exposes the file list even when contents are locked.

LZMA and LZMA2 by default, with optional support for PPMd, BZip2, and a few others — the .7z extension refers to the container format, which can wrap different underlying compression algorithms per archive.

Yes, with caveats. LZMA's large dictionary and range coding typically shrink DAM assets—especially batches of similar file types like TIFFs or PSDs—more than ZIP, cutting cold-storage costs. The LGPL license and public-domain LZMA SDK also lower the risk of the format becoming unreadable decades from now. The tradeoff is solid compression: pulling one file out of a solid 7z archive can require decompressing the whole block, so DAMs restoring individual assets should size archives per project or disable solid mode.

Yes. Several native libraries parse the 7z format directly in-process, without spawning an external binary. In Python, py7zr implements the format natively and supports LZMA2-compressed archives. In Java, Apache Commons Compress ships a SevenZFile reader that extracts entries directly from the API. .NET has SharpCompress, and C/C++ projects can link libarchive. This matters for DAM pipelines because avoiding a CLI call means fewer subprocess dependencies, better error handling, and no risk of shell-injection from untrusted filenames.

Igor Pavlov began developing LZMA in 1998. It became the default compression method three years later, in 2001, with the introduction of the 7z archive format. LZMA combines LZ77-style dictionary matching with range coding and context modeling, giving it better ratios than ZIP's Huffman-based DEFLATE. Pavlov later built LZMA2, a container format that splits data into independent chunks for multithreaded compression. In December 2008, he placed the LZMA SDK in the public domain, letting any project reuse the code freely.