Choose a version:
31% The original file has 51359 bytes (50.2k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 16143 bytes (15.8k, 31%).

After GZIP compression these minified files vary in size:
Boot
  6389 bytes (6.2k)
CDN
jsdelivr
  6389 bytes (6.2k)
CDN
cdnjs
  5690 bytes (5.6k)
CDN
gzip -6 (default)
  5687 bytes (5.6k)
local copy
gzip -9
  5673 bytes (5.5k)
local copy
unpkg
  5667 bytes (5.5k)
CDN
cdnhttps
  5663 bytes (5.5k)
CDN
Yandex
  5649 bytes (5.5k)
CDN
7zip -mx=9 -tgzip
  5505 bytes (5.4k)
local copy
libdeflate -12
  5502 bytes (5.4k)
local copy
pigz -11 -n
  5498 bytes (5.4k)
local copy
kzip -s0 -rn -b0
  5495 bytes (5.4k)
local copy
Zopfli
  5484 bytes (5.4k)
local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/underscorejs/underscore-1.8.1.min.js

You will automatically get the smallest UnderscoreJS 1.8.1 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 165 bytes by using my UnderscoreJS 1.8.1 Zopfli version instead of the best available CDN (3.01% smaller than Yandex, 5484 vs. 5649 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 --mls4 --bsr14 --lazy --ohh

(found October 21, 2015)
Description Value Parameter
iterations 100000  --i100000
maximum blocks 8  --mb8
maximum length score 4  --mls4
block splitting recursion 14  --bsr14
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/1.8.1/underscore-min.js --location | md5sum
6959908db2ddae758885b6c2cb2f07a5  -
curl --silent --compressed http://minime.stephan-brumme.com/files/underscorejs/underscore-1.8.1.min.zopfli.js.gz | md5sum
6959908db2ddae758885b6c2cb2f07a5  -

SHA1:
curl --silent --compressed https://raw.githubusercontent.com/jashkenas/underscore/1.8.1/underscore-min.js --location | sha1sum
3debb1117d56240c44bbeaac8bae20e0a6bcb27c  -
curl --silent --compressed http://minime.stephan-brumme.com/files/underscorejs/underscore-1.8.1.min.zopfli.js.gz | sha1sum
3debb1117d56240c44bbeaac8bae20e0a6bcb27c  -

All listed CDNs deliver identical contents:
CDN Size (compressed) MD5 (uncompressed) Timestamp
Boot 6389 bytes 6959908db2ddae758885b6c2cb2f07a5 (invalid)
jsdelivr 6389 bytes 6959908db2ddae758885b6c2cb2f07a5 (invalid)
cdnjs 5690 bytes 6959908db2ddae758885b6c2cb2f07a5 (invalid)
unpkg 5667 bytes 6959908db2ddae758885b6c2cb2f07a5 July 11, 2016 @ 16:49
cdnhttps 5663 bytes 6959908db2ddae758885b6c2cb2f07a5 December 24, 2015 @ 07:36
Yandex 5649 bytes 6959908db2ddae758885b6c2cb2f07a5 December 22, 2015 @ 17:09

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

Other Versions

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

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
5484 bytes -1 byte zopfli --i100000 --mls4 --bsr14 --lazy --ohh October 21, 2015 @ 13:34
5485 bytes -2 bytes zopfli --i10000 --mls4 --bsr11 --lazy --ohh September 22, 2015 @ 07:16
5487 bytes -1 byte zopfli --i10000 --mls32 --bsr12 --lazy --ohh September 19, 2015 @ 02:24
5488 bytes -1 byte zopfli --i100000 --mls8192 --bsr8 --lazy --ohh September 8, 2015 @ 21:52
5489 bytes -1 byte zopfli --i10000 --mls8192 --bsr16 --lazy --ohh September 8, 2015 @ 11:22
5490 bytes -1 byte zopfli --i1000 --mls8192 --bsr4 --lazy --ohh September 8, 2015 @ 10:39
5491 bytes -1 byte zopfli --i1000 --mls512 --bsr8 --lazy --ohh September 8, 2015 @ 10:39
5492 bytes zopfli --i100 --mls8192 --bsr8 --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 11, 2016 @ 16:49.

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 or 100,000.
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
5500 5493 5501 5500 5501 5500 5501 5493 5493 5493 5493 5492 5491 5492 5492
5493 5493 5493 5492 5493 5502 5493 5493 5493 5493 5493 5490 5488 5493 5493
5487 5503 5493 5493 5493 5497 5493 5493 5493 5493 5493 5489 5490 5491 5491
5503 5504 5501 5501 5500 5504 5493 5493 5493 5493 5493 5491 5491 5492 5492
5500 5493 5501 5500 5500 5498 5492 5492 5489 5493 5493 5490 5488 5492 5492
5503 5499 5501 5494 5501 5498 5492 5493 5493 5493 5493 5494 5490 5493 5493
5502 5493 5501 5500 5501 5492 5492 5493 5489 5493 5493 5490 5490 5492 5492
5502 5485 5499 5500 5499 5501 5492 5493 5493 5493 5493 5490 5490 5490 5490
5503 5503 5499 5500 5487 5497 5492 5492 5489 5493 5493 5490 5490 5492 5492
5502 5502 5501 5493 5501 5497 5493 5493 5493 5493 5493 5489 5489 5492 5492
5501 5484 5501 5500 5500 5500 5493 5492 5493 5493 5493 5489 5489 5492 5492
5501 5502 5499 5493 5493 5491 5493 5492 5493 5493 5493 5491 5491 5491 5491
5501 5502 5501 5493 5499 5491 5493 5493 5489 5493 5493 5490 5489 5486 5487
5501 5496 5499 5500 5500 5491 5493 5492 5493 5493 5493 5491 5491 5491 5491
5502 5505 5499 5493 5493 5497 5493 5492 5493 5493 5493 5491 5490 5490 5490
5501 5503 5501 5493 5500 5491 5493 5492 5491 5493 5493 5491 5491 5492 5492
5501 5502 5499 5500 5501 5501 5492 5492 5493 5493 5493 5491 5490 5487 5487
5502 5503 5501 5486 5493 5497 5493 5493 5491 5493 5493 5492 5491 5492 5492
5493 5493 5499 5500 5499 5497 5491 5493 5493 5493 5493 5489 5491 5491 5491
5502 5485 5499 5500 5501 5497 5492 5492 5493 5493 5493 5491 5491 5490 5490
5502 5503 5499 5486 5499 5491 5492 5492 5493 5493 5493 5491 5490 5490 5490
5502 5502 5500 5500 5501 5491 5493 5493 5493 5493 5493 5491 5490 5490 5490
5502 5502 5501 5501 5499 5498 5493 5493 5493 5493 5493 5491 5491 5491 5491

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 5490 bytes 100%
1,000 5487 bytes -3 bytes 100%
10,000 5485 bytes -2 bytes 100%
100,000 5484 bytes -1 byte 17.68%
1,000,000
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
5495 bytes +11 bytes (+0.20%)
5495 bytes +11 bytes (+0.20%)
5519 bytes +35 bytes (+0.64%) +24 bytes
5544 bytes +60 bytes (+1.09%) +49 bytes
5575 bytes +91 bytes (+1.66%) +80 bytes
5585 bytes +101 bytes (+1.84%) +90 bytes
5622 bytes +138 bytes (+2.52%) +127 bytes
5654 bytes +170 bytes (+3.10%) +159 bytes
5690 bytes +206 bytes (+3.76%) +195 bytes

Non-DEFLATE Algorithms

Archivers based on completely different compression algorithms often produce superior results.
Unfortunately, browsers only support gzip compression at the moment.
Algorithm Program Parameters Size Compared To Best Zopfli
RAR (proprietary) RAR rar a -m5 -md64m -mc63:128t -mt1 4832 bytes -652 bytes (-11.89%)
Brotli (Wikipedia) brotli brotli -q 11 5119 bytes -365 bytes (-6.66%)
Burrows-Wheeler transform (Wikipedia) bzip2 bzip2 -9 5423 bytes -61 bytes (-1.11%)
PPMd (Wikipedia) 7zip 7za a -mx=9 -m0=ppmd 5453 bytes -31 bytes (-0.57%)
ZSTD (Wikipedia) zstd zstd -19 5500 bytes +16 bytes (+0.29%)
LZMA2 (Wikipedia) xz xz -9 5548 bytes +64 bytes (+1.17%)
ZPAQ (Wikipedia) zpaq zpaq -method 69 5627 bytes +143 bytes (+2.61%)

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 2018.
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