UnderscoreJS 1.3.1 - as small as it gets (near-perfect GZIP compression)

Choose a version:

The original file has 35168 bytes (34.3k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 12140 bytes (11.9k, 35%).

After GZIP compression these minified files vary in size:

Boot	4604 bytes (4.5k)	CDN
Baidu	4130 bytes (4.0k)	CDN
gzip -6 (default)	4125 bytes (4.0k)	local copy
gzip -9	4115 bytes (4.0k)	local copy
cdnjs	4113 bytes (4.0k)	CDN
unpkg	4099 bytes (4.0k)	CDN
libdeflate -12	4012 bytes (3.9k)	local copy
7zip -mx=9 -tgzip	4008 bytes (3.9k)	local copy
kzip -s0 -rn -b1	4006 bytes (3.9k)	local copy
zultra	4006 bytes (3.9k)	local copy
pigz -11 -n	3996 bytes (3.9k)	local copy
Zopfli	3990 bytes (3.9k)	local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/underscorejs/underscore-1.3.1.min.js (or via HTTPS)

You will automatically get the smallest UnderscoreJS 1.3.1 file, ETag caching is available and
if your browser doesn't support GZIP decompression then the uncompressed version will be sent.

Currently best Zopfli settings

Save 109 bytes by using my UnderscoreJS 1.3.1 Zopfli version instead of the best available CDN (2.73% smaller than unpkg, 3990 vs. 4099 bytes):

Download

You can use my super-compressed files for whatever purpose you like as long as you respect the library's original license agreement.
There are no restrictions from my side - but please avoid hot-linking if you run a high-traffic website.

These command-line settings yielded the best compression ratio so far (Linux version of zopfli-krzymod):
zopfli --i100000 --mb8 --mls2048 --bsr10 --lazy --ohh

(found February 8, 2019)

Description	Value	Parameter
iterations	100000	--i100000
maximum blocks	8	--mb8
maximum length score	2048	--mls2048
block splitting recursion	10	--bsr10
lazy matching in LZ77	yes	--lazy
optimized Huffman headers	yes	--ohh
initial random W for iterations	1	--rw1
initial random Z for iterations	2	--rz2

Verify file integrity

After decompression, my uncompressed files are identical to the original ones:

MD5:
curl --silent --compressed https://raw.githubusercontent.com/jashkenas/underscore/1.3.1/underscore-min.js --location | md5sum
b538b8a3ed2c5519b19409eace25c38b -
curl --silent --compressed https://minime.stephan-brumme.com/files/underscorejs/underscore-1.3.1.min.zopfli.js.gz | md5sum
b538b8a3ed2c5519b19409eace25c38b -

SHA1:
curl --silent --compressed https://raw.githubusercontent.com/jashkenas/underscore/1.3.1/underscore-min.js --location | sha1sum
0a15952061f6fb7a5493281f6ff8eb916a417e2f -
curl --silent --compressed https://minime.stephan-brumme.com/files/underscorejs/underscore-1.3.1.min.zopfli.js.gz | sha1sum
0a15952061f6fb7a5493281f6ff8eb916a417e2f -

These CDNs send you the original file:

CDN	Size (compressed)	MD5 (uncompressed)	Timestamp
Boot	4604 bytes	b538b8a3ed2c5519b19409eace25c38b	(invalid)
cdnjs	4113 bytes	b538b8a3ed2c5519b19409eace25c38b	(invalid)
unpkg	4099 bytes	b538b8a3ed2c5519b19409eace25c38b	July 11, 2016 @ 16:49

And some CDNs send you a different file:

CDN	Size (compressed)	MD5 (uncompressed)	Comment / Diff	Timestamp
Baidu	4130 bytes	5831dba73eb2a76d1b662b244f2c2844	only whitespaces differ	(invalid)

Note: only the MD5 hashes are shown to keep things simple.

Other Versions

Available UnderscoreJS versions at minime.stephan-brumme.com:

1.13.6, 1.13.5, 1.13.4, 1.13.3, 1.13.2, 1.13.1, 1.13.0,
1.12.1, 1.12.0,
1.11.0,
1.10.2, 1.10.1, 1.10.0,
1.9.2, 1.9.1, 1.9.0,
1.8.3, 1.8.2, 1.8.1,
1.7.0,
1.6.0,
1.5.2, 1.5.0,
1.4.4, 1.4.3, 1.4.2, 1.4.1, 1.4.0,
1.3.3, 1.3.2, 1.3.1, 1.3.0,
1.2.4, 1.2.3, 1.2.2, 1.2.1, 1.2.0,
1.1.7, 1.1.5, 1.1.4, 1.1.3, 1.1.2, 1.1.1, 1.1.0,
1.0.4, 1.0.3, 1.0.2, 1.0.1, 1.0.0,
0.6.0,
0.5.7, 0.5.6, 0.5.5, 0.5.4, 0.5.3, 0.5.2, 0.5.1, 0.5.0,
0.4.7, 0.4.6, 0.4.5, 0.4.4, 0.4.3, 0.4.2, 0.4.1, 0.4.0,
0.3.3, 0.3.2, 0.3.1, 0.3.0,
0.2.0,
0.1.1, 0.1.0

The project site contains an overview how well these versions were compressed.
Other interesting projects are AngularJS, BackboneJS, Bootstrap, D3, Dojo, Ember, jQuery, Knockout, lodash, React, Socket.IO, ThreeJS and Vue.

Changelog

Best Zopfli parameters so far:

Size	Improvement	Parameters	Found
3990 bytes	-1 byte	zopfli --i100000 --mls2048 --bsr10 --lazy --ohh	February 8, 2019 @ 15:45
3991 bytes	-1 byte	zopfli --i1000 --mls2048 --bsr9 --lazy --ohh	September 8, 2015 @ 10:55
3992 bytes	-2 bytes	zopfli --i1000 --mls512 --bsr4 --lazy --ohh	September 8, 2015 @ 10:55
3994 bytes	–	zopfli --i100 --mls4096 --bsr16 --lazy --ohh	September 8, 2015 @ 10:32

If there are multiple parameter sets yielding the same compressed size, only the first one found is shown.

Most recent activity on July 20, 2020 @ 12:51.

Heatmaps

This Zopfli heatmap visualizes how compression changes when modifying the --bsr and --mls parameter.
Cell's contents is the best filesize achieved (in bytes, hover with mouse over cells to see number of iterations).

Good parameters are green, bad are red. The best and worst are bold as well.
The brightness of the blue background color indicates how many iterations were processed:
10,000, 100,000 or 1,000,000.

bsr \ mls	2	4	8	16	32	64	128	256	512	1024	2048	4096	8192	16384	32768
bsr \ mls	2	4	8	16	32	64	128	256	512	1024	2048	4096	8192	16384	32768
2	3995	3995	3995	3998	3998	3998	3997	3997	3997	3997	3998	3996	3993	3994	3994
4	3990	3990	3994	3994	3992	3994	3993	3991	3991	3990	3993	3991	3993	3992	3992
6	3993	3993	3990	3993	3993	3993	3993	3993	3992	3997	3990	3991	3993	3993	3993
7	3990	3990	3990	3990	3993	3993	3992	3990	3992	3993	3993	3993	3993	3993	3993
8	3994	3994	3993	3993	3993	3992	3994	3990	3993	3991	3994	3990	3993	3993	3993
9	3990	3990	3990	3990	3990	3992	3992	3994	3993	3992	3990	3993	3993	3995	3995
10	3990	3990	3993	3993	3993	3992	3992	3990	3993	3990	3990	3994	3991	3990	3990
11	3990	3990	3994	3993	3993	3994	3994	3990	3992	3992	3990	3990	3993	3994	3994
12	3992	3992	3991	3993	3993	3993	3993	3992	3993	3995	3990	3992	3990	3993	3993
13	3990	3990	3993	3990	3992	3994	3994	3990	3992	3992	3990	3990	3993	3993	3993
14	3990	3990	3994	3996	3993	3994	3994	3990	3993	3992	3993	3993	3994	3993	3993
15	3990	3990	3990	3990	3993	3994	3993	3990	3993	3993	3993	3993	3993	3993	3993
16	3990	3990	3990	3992	3993	3992	3993	3990	3993	3993	3990	3990	3993	3993	3993
17	3990	3990	3990	3990	3997	3992	3999	3990	3995	3993	3990	3993	3994	3990	3990
18	3990	3990	3993	3990	3990	3993	3992	3992	3993	3991	3990	3994	3993	3994	3994
19	3993	3993	3990	3991	3993	3992	3992	3992	3993	3990	3991	3993	3990	3990	3990
20	3993	3993	3990	3993	3993	3993	3992	3990	3993	3990	3990	3990	3994	3994	3994
21	3993	3993	3994	3991	3993	3992	3992	3992	3993	3992	3990	3990	3993	3993	3993
22	3991	3991	3990	3997	3993	3993	3992	3990	3993	3993	3991	3990	3990	3994	3994
23	3990	3990	3993	3993	3998	3992	3992	3992	3993	3993	3990	3992	3994	3994	3994
25	3991	3991	3992	3993	3993	3993	3994	3991	3993	3991	3993	3992	3990	3994	3994
30	3990	3990	3991	3996	3993	3992	3994	3990	3993	3990	3991	3993	3993	3993	3993
40	3990	3990	3990	3993	3993	3992	3993	3990	3993	3996	3990	3993	3994	3990	3990

Due to the Monte Carlo design of my search algorithm, not all parameters have reached the same number of iterations yet:

Iterations	Min. Bytes	Reduction	Coverage
100	3993 bytes	–	100%
1,000	3991 bytes	-2 bytes	100%
10,000	3991 bytes	–	100%
100,000	3990 bytes	-1 byte	33.62%
1,000,000	3990 bytes	–	0.29%
10,000,000	–	–	–

KZIP has far less options available for tuning/optimization. I only played around with the number of blocks (parameter -n):

Blocks	Min. Bytes	Compared To Best Zopfli	Compared To Best KZIP
(any)	4006 bytes	+16 bytes (+0.40%)	–
1	4006 bytes	+16 bytes (+0.40%)	–
2	4024 bytes	+34 bytes (+0.85%)	+18 bytes
3	4022 bytes	+32 bytes (+0.80%)	+16 bytes
4	4042 bytes	+52 bytes (+1.30%)	+36 bytes
5	4075 bytes	+85 bytes (+2.13%)	+69 bytes
6	4108 bytes	+118 bytes (+2.96%)	+102 bytes
7	4140 bytes	+150 bytes (+3.76%)	+134 bytes
8	4173 bytes	+183 bytes (+4.59%)	+167 bytes

Non-DEFLATE Algorithms

Archivers based on completely different compression algorithms often produce superior results.
Unfortunately, browsers only support gzip compression at the moment.
However, support for Brotli is constantly growing - for example, your browser actually supports it !

Algorithm	Program	Parameters	Size	Compared To Best Zopfli
RAR (proprietary)	RAR	rar a -m5 -md64m -mc63:128t -mt1	3571 bytes	-419 bytes (-10.50%)
Brotli (Wikipedia)	brotli	brotli -q 11	3677 bytes	-313 bytes (-7.84%)
PPMd (Wikipedia)	7zip	7za a -mx=9 -m0=ppmd	3952 bytes	-38 bytes (-0.95%)
Zstandard (Wikipedia)	zstd	zstd -19	4025 bytes	+35 bytes (+0.88%)
Burrows-Wheeler transform (Wikipedia)	bzip2	bzip2 -9	4037 bytes	+47 bytes (+1.18%)
LZMA2 (Wikipedia)	xz	xz -9	4084 bytes	+94 bytes (+2.36%)
ZPAQ (Wikipedia)	zpaq	zpaq -method 69	4423 bytes	+433 bytes (+10.85%)

Detailled Analysis

I wrote a DEFLATE decoder in Javascript. Click the button below to start a client-side analysis of the smallest gzipped files (may take a second):

Load and Parse Data

Notes: pigz is a fast open source multi-threaded implementation of gzip written by one of the original authors of gzip.
However, when using compression level 11, pigz actually switches to the slower Zopfli algorithm and isn't multi-threaded anymore.
KrzyMOD's extensions to Zopfli offer the highest level of configuration and is therefore used for my brute-force search.
Ken Silverman wrote the closed-source KZIP compression program and Jonathon Fowler ported it to Linux.
Defluff was created by Joachim Henke; DeflOpt is a tool by Ben Jos Walbeehm.

website made by Stephan Brumme in 2015 and still improving in 2026.
all timestamps are displayed in central european time. see my changelog.
no flash, not even images or external css files - and everything squeezed into a single html file.
which was handsomely compressed before releasing it into the wild internet - obviously.

please visit my homepage and my blog, too.
email: minime (at) stephan-brumme.com

All trademarks are property of their respective owners. You know, the boring legal stuff.