dupeGuru vs fclones

dupeGuru is a desktop tool for finding duplicate files on Linux, macOS, and Windows. Its Standard, Music, and Picture modes offer different ways to compare files. Filename scans can detect similar names, while Music mode can compare names or selected tags and show music-specific details. Picture mode looks for similar images, including different-sized images when that preference is enabled. Contents scans identify exact matches, first using file sizes to rule out files that cannot match and then comparing contents with MD5 hashes. A recursive contents scan can also locate duplicate folders. Users can filter and sort results, preview files, and delete, move, or copy marked duplicates. Reference folders and duplicate groups are designed to protect reference files and leave at least one item in each group. The documentation recommends checking results before deletion: similar-word matching can cause false positives and slow scans, and picture comparisons can be CPU-intensive. dupeGuru is free, with packages listed for Windows, macOS, Debian, RPM, Arch Linux, and Ubuntu PPA.

fclones is a free command-line utility for locating groups of identical files and handling redundant copies. It runs on Linux, macOS, and Windows, and can scan multiple directory roots, standard-input file lists, and mounted or removable devices. Search controls include recursion limits, glob and regular-expression filters, and minimum or maximum file sizes. It can compare files using size, inode or file identifiers, prefix and suffix hashes, and full-content hashes. Results are available as text, fdupes-compatible output, CSV, or JSON. Once duplicates are identified, fclones can remove or move them, or replace them with soft or hard links. On supported file systems, it can deduplicate file data with native copy-on-write reflinks, a capability unavailable on Windows. Users can inspect and alter duplicate lists before removal, and dry-run mode reports changes without executing them. The Rust implementation parallelizes I/O and CPU-heavy work and adjusts processing for SSDs and HDDs. Optional persistent hash caching can help repeated scans and resumed work; the project is released under the MIT license.