Three publishable runs under the exclusive bench queue, one per filtered group, 3 repetitions (tools 3), run on the build of 2026-11-02 (main 33c568d plus the bench rows). Groups: A mp4-mux, mp4-remux, mkv-mux; B tagw-id3, tagw-flac, tagw-ilst, tagw-riff, tagw-mkv; C remux-mem, playable. Comparators: SharpMP4 0.3.2, TagLib# 2.3.0, ATL.NET 7.19. Method: docs/bench.md.

Against the bar

Bar: geometric mean of the ratios <= 1.00x and no file worse than 1.25x, against the fastest managed comparator per file.

Notes

• Each group ran as its own job under the exclusive benchmark queue, alone on the machine. Mac corpus: the files of the 0.2 run plus h264-aac-tags.mp4, audio-aac-priming.m4a, h264-two-audio-tags.mkv and the two files over 4 GiB, regenerated with tools/corpus/generate.sh (ffmpeg 8.0 there, so the bytes differ from the x64 corpus' files of the same name). flac-cover-tags.flac is not in the Mac corpus, so tagw-flac has two files.
• Tag-write rows edit the same four common fields of a file held in memory; the ratio is against TagLib#, the fastest managed comparator on every file. ATL.NET is 1.6x to 14x slower than TagLib# (its ratio column), least on WAV and most on ilst.
• tagw-mkv is the slowest row on this machine: 1.09x on its one file, inside the 1.25x per-row limit; the tag-write geomean stays under 1.0x (TagLib# is the only comparator that can edit it; ATL.NET fails on both Matroska files and TagLib# on test1.mkv). On x64 the same row is 0.99x. A row with one file and a comparator that handles almost no Matroska is thin evidence in either direction.
• Rows mkv-mux, remux-mem and playable have no managed comparator, so they pass; times next to ffmpeg 8.0 and mkvmerge 103.0 are information.
• remux-mem: the bench's own peak figure is blank on macOS (the runtime reports no peak working set there), so the peaks were taken separately with /usr/bin/time -l around one child process per cell (one cold run plus the big-file samples each): 4.08 GiB MP4 to Matroska 73.3 MiB, to MP4 70.5 MiB; 4.14 GiB Matroska to Matroska 96.4 MiB, to MP4 90.6 MiB (ffmpeg -c copy 99.7 MiB and 40.7 MiB). Times on this Mac (10 to 27 s) are bound by the disk write, not by the remuxer.
• playable: 13.4 ms here, against 31.1 ms for ffmpeg 8.0 (information).

Group A: mp4-mux, mp4-remux, mkv-mux

Run 2026-11-02 04:35Z

• Machine: Apple M2; 8 logical cores; macOS 27.0.1 Arm64
• ISA: neon crc32 aes dotprod
• Storage of the corpus: /dev/disk3s5; every file is read once before its rows, so it is in the page cache for all of them
• .NET: 10.0.11 (.NET 10.0.11)
• Load average (1 min): before 3.90; during mean 3.51, max 3.90 over 11 samples at 5 s; after 3.22
• Elapsed: 0.8 min; repetitions 3 (managed), 3 (tools); files over 1 GiB get one warm-up run and 3 samples of one run each, and tools once
• Packages as restored: SharpMP4 0.3.2; RecursiveAI.OneAudio 1.6.1; MetadataExtractor 2.9.3; TagLibSharp 2.3.0.0; z440.atl.core 7.19; Matroska 0.0.12
• Tools: ffmpeg version 8.0 Copyright (c) 2000-2025 the FFmpeg developers; MP4Box - GPAC version 26.07-revrelease; mkvmerge v103.0 ('The Darkness') 64-bit; MediaInfo MediaInfo Command line,; ExifTool 13.60

Ratio is OneMedia's median over the fastest managed comparator's median on the same file (below 1.00× is faster). Peak is the process's peak resident set; Baseline is its resident set before any work, so Peak − Baseline is what the work added.

mp4-mux: MP4: write the first AAC track's frames from memory to an MP4 in memory

mp4-remux: MP4: read the first AAC track from a file and write it as an MP4 in memory

Reference tools, run as processes (wall time including start-up, peak resident set); information only:

mkv-mux: Matroska/WebM: read every packet and write it to a new Matroska file (no managed Matroska writer to compare with)

Reference tools, run as processes (wall time including start-up, peak resident set); information only:

Group B: tagw-id3, tagw-flac, tagw-ilst, tagw-riff, tagw-mkv

Run 2026-11-02 04:36Z

• Machine: Apple M2; 8 logical cores; macOS 27.0.1 Arm64
• ISA: neon crc32 aes dotprod
• Storage of the corpus: /dev/disk3s5; every file is read once before its rows, so it is in the page cache for all of them
• .NET: 10.0.11 (.NET 10.0.11)
• Load average (1 min): before 3.22; during mean 2.88, max 3.22 over 13 samples at 5 s; after 2.78
• Elapsed: 1.0 min; repetitions 3 (managed), 3 (tools); files over 1 GiB get one warm-up run and 3 samples of one run each, and tools once
• Packages as restored: SharpMP4 0.3.2; RecursiveAI.OneAudio 1.6.1; MetadataExtractor 2.9.3; TagLibSharp 2.3.0.0; z440.atl.core 7.19; Matroska 0.0.12
• Tools: ffmpeg version 8.0 Copyright (c) 2000-2025 the FFmpeg developers; MP4Box - GPAC version 26.07-revrelease; mkvmerge v103.0 ('The Darkness') 64-bit; MediaInfo MediaInfo Command line,; ExifTool 13.60

Ratio is OneMedia's median over the fastest managed comparator's median on the same file (below 1.00× is faster). Peak is the process's peak resident set; Baseline is its resident set before any work, so Peak − Baseline is what the work added.

tagw-id3: MP3: change title, artist, album and comment; the edited file in memory

tagw-flac: FLAC: change title, artist, album and comment; the edited file in memory

tagw-ilst: MP4/M4A: change title, artist, album and comment in ilst; the edited file in memory

tagw-riff: WAV: change LIST INFO title, artist, album and comment; the edited file in memory

tagw-mkv: Matroska: change title, artist, album and comment; the edited file in memory

Group C: remux-mem, playable

Run 2026-11-02 04:42Z

• Machine: Apple M2; 8 logical cores; macOS 27.0.1 Arm64
• ISA: neon crc32 aes dotprod
• Storage of the corpus: /dev/disk3s5; every file is read once before its rows, so it is in the page cache for all of them
• .NET: 10.0.11 (.NET 10.0.11)
• Load average (1 min): before 2.78; during mean 1.84, max 2.78 over 74 samples at 5 s; after 1.66
• Elapsed: 6.2 min; repetitions 3 (managed), 3 (tools); files over 1 GiB get one warm-up run and 3 samples of one run each, and tools once
• Packages as restored: SharpMP4 0.3.2; RecursiveAI.OneAudio 1.6.1; MetadataExtractor 2.9.3; TagLibSharp 2.3.0.0; z440.atl.core 7.19; Matroska 0.0.12
• Tools: ffmpeg version 8.0 Copyright (c) 2000-2025 the FFmpeg developers; MP4Box - GPAC version 26.07-revrelease; mkvmerge v103.0 ('The Darkness') 64-bit; MediaInfo MediaInfo Command line,; ExifTool 13.60

Ratio is OneMedia's median over the fastest managed comparator's median on the same file (below 1.00× is faster). Peak is the process's peak resident set; Baseline is its resident set before any work, so Peak − Baseline is what the work added.

remux-mem: Remux of the files past 4 GiB into MP4 and Matroska, output to disk: peak resident memory

Reference tools, run as processes (wall time including start-up, peak resident set); information only:

playable: MakePlayable (generic MP4): Matroska H.264+AAC to MP4; information

Reference tools, run as processes (wall time including start-up, peak resident set); information only: