Choose a version:
35% The original file has 35286 bytes (34.5k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 12203 bytes (11.9k, 35%).

After GZIP compression these minified files vary in size:
Boot
  4623 bytes (4.5k)
CDN
Baidu
  4152 bytes (4.1k)
CDN
gzip -6 (default)
  4141 bytes (4.0k)
local copy
gzip -9
  4133 bytes (4.0k)
local copy
cdnjs
  4129 bytes (4.0k)
CDN
unpkg
  4123 bytes (4.0k)
CDN
libdeflate -12
  4025 bytes (3.9k)
local copy
7zip -mx=9 -tgzip
  4021 bytes (3.9k)
local copy
kzip -s0 -rn -b1
  4019 bytes (3.9k)
local copy
zultra
  4017 bytes (3.9k)
local copy
pigz -11 -n
  4010 bytes (3.9k)
local copy
Zopfli
  4004 bytes (3.9k)
local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/underscorejs/underscore-1.2.4.min.js (or via HTTPS)

You will automatically get the smallest UnderscoreJS 1.2.4 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 119 bytes by using my UnderscoreJS 1.2.4 Zopfli version instead of the best available CDN (2.97% smaller than unpkg, 4004 vs. 4123 bytes):
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 --mls1024 --bsr10 --lazy --ohh

(found February 12, 2019)
Description Value Parameter
iterations 100000  --i100000
maximum blocks 8  --mb8
maximum length score 1024  --mls1024
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/1.2.4/underscore-min.js --location | md5sum
ba6fafc612bc3d089d76072aa4c6198f  -
curl --silent --compressed https://minime.stephan-brumme.com/files/underscorejs/underscore-1.2.4.min.zopfli.js.gz | md5sum
ba6fafc612bc3d089d76072aa4c6198f  -

SHA1:
curl --silent --compressed https://raw.githubusercontent.com/jashkenas/underscore/1.2.4/underscore-min.js --location | sha1sum
63f6a94752e5fde78627529763e67ecd4ef78f12  -
curl --silent --compressed https://minime.stephan-brumme.com/files/underscorejs/underscore-1.2.4.min.zopfli.js.gz | sha1sum
63f6a94752e5fde78627529763e67ecd4ef78f12  -

These CDNs send you the original file:
CDN Size (compressed) MD5 (uncompressed) Timestamp
Boot 4623 bytes ba6fafc612bc3d089d76072aa4c6198f (invalid)
cdnjs 4129 bytes ba6fafc612bc3d089d76072aa4c6198f (invalid)
unpkg 4123 bytes ba6fafc612bc3d089d76072aa4c6198f July 11, 2016 @ 16:49

And some CDNs send you a different file:
CDN Size (compressed) MD5 (uncompressed) Comment / Diff Timestamp
Baidu 4152 bytes e74af9e87325f5617c1aae48768eb6fa 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
4004 bytes -1 byte zopfli --i100000 --mls1024 --bsr10 --lazy --ohh February 12, 2019 @ 21:42
4005 bytes -1 byte zopfli --i10000 --mls1024 --bsr8 --lazy --ohh September 8, 2015 @ 15:50
4006 bytes -1 byte zopfli --i1000 --mls1024 --bsr8 --lazy --ohh September 8, 2015 @ 10:57
4007 bytes zopfli --i100 --mls1024 --bsr6 --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 @ 12:49.

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
4009 4009 4009 4013 4014 4013 4015 4013 4013 4013 4020 4008 4009 4009 4009
4009 4009 4008 4008 4008 4009 4005 4013 4008 4006 4006 4006 4005 4005 4005
4009 4009 4008 4007 4008 4007 4008 4008 4007 4005 4007 4004 4006 4005 4005
4008 4008 4009 4008 4007 4008 4004 4006 4006 4006 4005 4005 4009 4008 4008
4009 4009 4007 4007 4008 4008 4007 4008 4008 4005 4006 4007 4006 4005 4005
4009 4009 4008 4007 4007 4009 4007 4013 4009 4007 4005 4007 4009 4008 4008
4009 4009 4007 4007 4007 4008 4004 4008 4008 4004 4007 4005 4009 4005 4005
4009 4009 4007 4007 4007 4008 4004 4008 4008 4006 4005 4005 4009 4005 4005
4008 4008 4009 4008 4007 4008 4010 4014 4008 4007 4005 4005 4006 4006 4006
4009 4009 4006 4007 4007 4008 4004 4008 4007 4005 4006 4005 4006 4006 4006
4009 4009 4010 4007 4007 4008 4005 4008 4008 4005 4006 4005 4005 4006 4006
4009 4009 4008 4008 4008 4008 4005 4008 4008 4008 4006 4005 4006 4009 4009
4009 4009 4007 4007 4008 4008 4007 4008 4007 4007 4006 4008 4009 4009 4009
4009 4009 4007 4007 4010 4008 4007 4008 4008 4007 4005 4005 4005 4005 4005
4009 4009 4009 4008 4008 4009 4004 4008 4007 4008 4005 4008 4005 4008 4008
4009 4009 4006 4007 4010 4008 4004 4013 4011 4006 4007 4005 4005 4005 4005
4009 4009 4007 4007 4008 4009 4009 4008 4008 4008 4006 4005 4004 4005 4005
4009 4009 4009 4007 4007 4008 4005 4008 4008 4006 4013 4005 4004 4004 4004
4009 4009 4009 4007 4010 4008 4005 4014 4008 4005 4013 4005 4006 4006 4006
4009 4009 4007 4007 4008 4008 4008 4008 4007 4008 4008 4006 4006 4006 4006
4008 4008 4010 4008 4007 4007 4005 4008 4007 4006 4006 4005 4006 4006 4006
4009 4009 4007 4008 4008 4008 4005 4008 4008 4006 4007 4005 4006 4006 4006
4009 4009 4007 4007 4007 4009 4007 4008 4008 4005 4006 4005 4009 4009 4009

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 4007 bytes 100%
1,000 4006 bytes -1 byte 100%
10,000 4005 bytes -1 byte 100%
100,000 4004 bytes -1 byte 19.42%
1,000,000 4004 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
4019 bytes +15 bytes (+0.37%)
4019 bytes +15 bytes (+0.37%)
4039 bytes +35 bytes (+0.87%) +20 bytes
4035 bytes +31 bytes (+0.77%) +16 bytes
4071 bytes +67 bytes (+1.67%) +52 bytes
4093 bytes +89 bytes (+2.22%) +74 bytes
4125 bytes +121 bytes (+3.02%) +106 bytes
4151 bytes +147 bytes (+3.67%) +132 bytes
4184 bytes +180 bytes (+4.50%) +165 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 - but your browser doesn't support it.
Algorithm Program Parameters Size Compared To Best Zopfli
RAR (proprietary) RAR rar a -m5 -md64m -mc63:128t -mt1 3588 bytes -416 bytes (-10.39%)
Brotli (Wikipedia) brotli brotli -q 11 3691 bytes -313 bytes (-7.82%)
PPMd (Wikipedia) 7zip 7za a -mx=9 -m0=ppmd 3957 bytes -47 bytes (-1.17%)
Zstandard (Wikipedia) zstd zstd -19 4038 bytes +34 bytes (+0.85%)
Burrows-Wheeler transform (Wikipedia) bzip2 bzip2 -9 4048 bytes +44 bytes (+1.10%)
LZMA2 (Wikipedia) xz xz -9 4100 bytes +96 bytes (+2.40%)
ZPAQ (Wikipedia) zpaq zpaq -method 69 4440 bytes +436 bytes (+10.89%)

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):


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 2024.
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.