Choose a version:
27% The original file has 1635480 bytes (1,597.1k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 437344 bytes (427.1k, 27%).

After GZIP compression these minified files vary in size:
Boot
  138434 bytes (135.2k)
CDN
gzip -6 (default)
  112381 bytes (109.7k)
local copy
gzip -9
  112004 bytes (109.4k)
local copy
cdnhttps
  111836 bytes (109.2k)
CDN
cdnjs
  111836 bytes (109.2k)
CDN
7zip -mx=9 -tgzip
  108228 bytes (105.7k)
local copy
libdeflate -12
  108164 bytes (105.6k)
local copy
pigz -11 -n
  108052 bytes (105.5k)
local copy
kzip -s0 -rn -b5
  107849 bytes (105.3k)
local copy
Zopfli
  107811 bytes (105.3k)
local copy
Zopfli (defluff)
  107810 bytes (105.3k)
local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/ember/ember-2.1.2.min.js

You will automatically get the smallest Ember 2.1.2 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 4025 bytes by using my Ember 2.1.2 Zopfli version instead of the best available CDN (3.73% smaller than cdnjs, 107811 vs. 111836 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 --mls512 --bsr7 --lazy --ohh

(found January 18, 2016)
Description Value Parameter
iterations 100000  --i100000
maximum blocks 8  --mb8
maximum length score 512  --mls512
block splitting recursion 7  --bsr7
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

Even Smaller Files Thanks To Defluff

Zopfli's output can be further optimized by the defluff tool.
In this particular case, defluff saves 1 more byte (107810 bytes).

Verify file integrity

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

MD5:
curl --silent --compressed http://builds.emberjs.com/tags/v2.1.2/ember.min.js --location | md5sum
16ffbb3d9fd945b7d1149ef6ebb8ed9d  -
curl --silent --compressed http://minime.stephan-brumme.com/files/ember/ember-2.1.2.min.zopfli.js.gz | md5sum
16ffbb3d9fd945b7d1149ef6ebb8ed9d  -

SHA1:
curl --silent --compressed http://builds.emberjs.com/tags/v2.1.2/ember.min.js --location | sha1sum
1abacbf55e80452af0809f9ed74486156d1504e3  -
curl --silent --compressed http://minime.stephan-brumme.com/files/ember/ember-2.1.2.min.zopfli.js.gz | sha1sum
1abacbf55e80452af0809f9ed74486156d1504e3  -

All listed CDNs deliver identical contents:
CDN Size (compressed) MD5 (uncompressed) Timestamp
Boot 138434 bytes 16ffbb3d9fd945b7d1149ef6ebb8ed9d January 16, 2016 @ 15:03
cdnhttps 111836 bytes 16ffbb3d9fd945b7d1149ef6ebb8ed9d November 22, 2016 @ 16:33
cdnjs 111836 bytes 16ffbb3d9fd945b7d1149ef6ebb8ed9d January 15, 2016 @ 07:02

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

Other Versions

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

3.0.0,
2.18.2, 2.18.1, 2.18.0,
2.17.2, 2.17.1, 2.17.0,
2.16.2, 2.16.1, 2.16.0,
2.15.3, 2.15.2, 2.15.1, 2.15.0,
2.14.1, 2.14.0,
2.13.4, 2.13.3, 2.13.2, 2.13.1, 2.13.0,
2.12.2, 2.12.1, 2.12.0,
2.11.3, 2.11.2, 2.11.1, 2.11.0,
2.10.2, 2.10.1, 2.10.0,
2.9.1, 2.9.0,
2.8.3, 2.8.2, 2.8.1, 2.8.0,
2.7.3, 2.7.2, 2.7.1, 2.7.0,
2.6.2, 2.6.1, 2.6.0,
2.5.1, 2.5.0,
2.4.6, 2.4.5, 2.4.4, 2.4.3, 2.4.2, 2.4.1, 2.4.0,
2.3.1, 2.3.0,
2.2.2, 2.2.1, 2.2.0,
2.1.2, 2.1.1, 2.1.0,
2.0.3, 2.0.2, 2.0.1, 2.0.0,
1.13.13, 1.13.12, 1.13.11, 1.13.10, 1.13.9, 1.13.8, 1.13.7, 1.13.6, 1.13.5, 1.13.4, 1.13.3, 1.13.2, 1.13.1, 1.13.0,
1.12.2, 1.12.1, 1.12.0,
1.11.4, 1.11.3, 1.11.2, 1.11.1, 1.11.0,
1.10.1, 1.10.0,
1.9.1, 1.9.0,
1.8.1, 1.8.0,
1.7.1, 1.7.0,
1.6.1, 1.6.0,
1.5.1, 1.5.0,
1.4.0,
1.3.2, 1.3.1, 1.3.0,
1.2.2, 1.2.1, 1.2.0,
1.0.1, 1.0.0

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

Changelog

Best Zopfli parameters so far:
Size Improvement Parameters Found
107811 bytes -8 bytes zopfli --i100000 --mls512 --bsr7 --lazy --ohh January 18, 2016 @ 19:06
107819 bytes -10 bytes zopfli --i10000 --mls512 --bsr7 --lazy --ohh January 18, 2016 @ 03:30
107829 bytes -1 byte zopfli --i1000 --mls64 --bsr19 --lazy --ohh January 17, 2016 @ 23:02
107830 bytes -20 bytes zopfli --i1000 --mls64 --bsr18 --lazy --ohh January 17, 2016 @ 21:23
107850 bytes -20 bytes zopfli --i1000 --mls1024 --bsr4 --lazy --ohh January 17, 2016 @ 21:20
107870 bytes zopfli --i100 --mls512 --bsr7 --lazy --ohh January 17, 2016 @ 21:07

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

Most recent activity on November 22, 2016 @ 16:33.

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
107968 107914 107993 107894 107991 107974 108001 108001 107993 107970 107973 107971 107979 107995 107999
107883 107844 107950 107950 107946 107953 107946 107950 107833 107841 107830 107845 107873 107995 107954
107831 107816 107831 107831 107835 107835 107832 107835 107850 107836 107847 107838 107916 107995 107999
107949 107947 107950 107951 107952 107812 107831 107836 107811 107948 107950 107836 107953 107951 107952
107951 107951 107948 107950 107948 107844 107950 107950 107951 107952 107948 107840 107957 107952 107950
107856 107947 107950 107945 107946 107843 107950 107947 107950 107952 107950 107838 107953 107951 107949
107949 107947 107949 107948 107950 107949 107950 107951 107950 107952 107949 107847 107834 107951 107951
107950 107950 107950 107948 107950 107862 107950 107951 107945 107949 107951 107841 107952 107952 107952
107949 107947 107949 107948 107952 107850 107946 107950 107950 107948 107949 107844 107951 107951 107948
107949 107947 107948 107950 107948 107852 107951 107950 107950 107949 107949 107838 107832 107951 107950
107949 107947 107948 107949 107948 107858 107946 107947 107950 107950 107949 107947 107954 107950 107951
107949 107947 107948 107948 107948 107846 107950 107950 107950 107948 107828 107834 107954 107951 107950
107817 107950 107948 107949 107950 107855 107951 107950 107950 107950 107952 107839 107830 107953 107951
107857 107945 107942 107946 107949 107863 107946 107952 107951 107948 107952 107842 107833 107951 107948
107817 107947 107947 107947 107947 107816 107950 107947 107947 107949 107821 107831 107956 107953 107951
107826 107947 107948 107949 107947 107815 107949 107950 107949 107951 107949 107841 107834 107952 107952
107949 107947 107950 107948 107950 107849 107949 107950 107950 107950 107950 107860 107832 107952 107946
107853 107947 107948 107950 107948 107845 107949 107950 107950 107950 107952 107832 107953 107952 107946
107828 107947 107948 107945 107949 107846 107947 107945 107950 107950 107952 107841 107833 107952 107949
107833 107950 107948 107948 107948 107857 107950 107950 107950 107948 107950 107852 107954 107954 107976
107832 107947 107950 107950 107948 107816 107945 107947 107950 107949 107950 107839 107831 107954 107949
107832 107951 107945 107951 107946 107837 107947 107946 107951 107952 107834 107832 107831 107951 107950
107851 107950 107950 107948 107949 107849 107946 107945 107951 107950 107954 107839 107949 107954 107949

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 107870 bytes 100%
1,000 107829 bytes -41 bytes 100%
10,000 107819 bytes -10 bytes 100%
100,000 107811 bytes -8 bytes 2.61%
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
107850 bytes +39 bytes (+0.04%) +1 byte
107979 bytes +168 bytes (+0.16%) +130 bytes
107983 bytes +172 bytes (+0.16%) +134 bytes
107943 bytes +132 bytes (+0.12%) +94 bytes
107853 bytes +42 bytes (+0.04%) +4 bytes
107849 bytes +38 bytes (+0.04%)
107883 bytes +72 bytes (+0.07%) +34 bytes
107916 bytes +105 bytes (+0.10%) +67 bytes
107983 bytes +172 bytes (+0.16%) +134 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
ZPAQ (Wikipedia) zpaq zpaq -method 69 73880 bytes -33931 bytes (-31.47%)
RAR (proprietary) RAR rar a -m5 -md64m -mc63:128t -mt1 79085 bytes -28726 bytes (-26.64%)
PPMd (Wikipedia) 7zip 7za a -mx=9 -m0=ppmd 85007 bytes -22804 bytes (-21.15%)
Burrows-Wheeler transform (Wikipedia) bzip2 bzip2 -9 90014 bytes -17797 bytes (-16.51%)
Brotli (Wikipedia) brotli brotli -q 11 93883 bytes -13928 bytes (-12.92%)
LZMA2 (Wikipedia) xz xz -9 96088 bytes -11723 bytes (-10.87%)
ZSTD (Wikipedia) zstd zstd -19 98771 bytes -9040 bytes (-8.39%)

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