# COMPRESSION_LZFSE The LZFSE compression algorithm, which is recommended for use on Apple platforms. ``` var COMPRESSION_LZFSE: compression_algorithm { get } ``` ## Discussion LZFSE is the Apple proprietary, general-purpose compression algorithm. It pairs an LZ77-style match finder with Finite State Entropy (FSE, a variant of tabled Asymmetric Numeral Systems) to encode literals and match metadata, which lets it match the compression ratio of [`COMPRESSION_ZLIB`](/documentation/Compression/COMPRESSION_ZLIB) (zlib level 5) while running 2x to 3x faster for both encoding and decoding. LZFSE is also substantially more energy-efficient than [`COMPRESSION_ZLIB`](/documentation/Compression/COMPRESSION_ZLIB), which makes it a good default for mobile and battery-powered devices. LZFSE sits at a balanced point in the speed-ratio tradeoff: it compresses better than [`COMPRESSION_LZ4`](/documentation/Compression/COMPRESSION_LZ4) and runs faster than [`COMPRESSION_ZLIB`](/documentation/Compression/COMPRESSION_ZLIB) or [`COMPRESSION_LZMA`](/documentation/Compression/COMPRESSION_LZMA), but doesn’t match the ratio of [`COMPRESSION_LZMA`](/documentation/Compression/COMPRESSION_LZMA) or the raw throughput of [`COMPRESSION_LZ4`](/documentation/Compression/COMPRESSION_LZ4). For new Apple-only code, prefer [`COMPRESSION_LZMESH`](/documentation/Compression/COMPRESSION_LZMESH) — it typically compresses a bit better than LZFSE, encodes significantly faster, and decodes slightly faster. If a payload needs to be decoded on another platform such as Linux or Windows, use [`COMPRESSION_LZ4`](/documentation/Compression/COMPRESSION_LZ4), [`COMPRESSION_LZMA`](/documentation/Compression/COMPRESSION_LZMA), or [`COMPRESSION_ZLIB`](/documentation/Compression/COMPRESSION_ZLIB) instead. Apple does publish an open-source LZFSE reference implementation, but it isn’t as widely available as [`COMPRESSION_ZLIB`](/documentation/Compression/COMPRESSION_ZLIB) or [`COMPRESSION_LZMA`](/documentation/Compression/COMPRESSION_LZMA). LZFSE is supported by both the buffer API — [`compression_encode_buffer(_:_:_:_:_:_:)`](/documentation/Compression/compression_encode_buffer(_:_:_:_:_:_:)) and [`compression_decode_buffer(_:_:_:_:_:_:)`](/documentation/Compression/compression_decode_buffer(_:_:_:_:_:_:)) — and the streaming API via [`compression_stream_init(_:_:_:)`](/documentation/Compression/compression_stream_init(_:_:_:)), so you can use it for in-memory payloads or for processing data incrementally as it arrives. --- Copyright © 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)