Choose a version:
21% The original file has 641547 bytes (626.5k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 135581 bytes (132.4k, 21%).

After GZIP compression these minified files vary in size:
Boot
  46857 bytes (45.8k)
CDN
gzip -6 (default)
  39598 bytes (38.7k)
local copy
cdnhttps
  39553 bytes (38.6k)
CDN
cdnjs
  39553 bytes (38.6k)
CDN
jsdelivr
  39553 bytes (38.6k)
CDN
unpkg
  39553 bytes (38.6k)
CDN
gzip -9
  39444 bytes (38.5k)
local copy
libdeflate -12
  38157 bytes (37.3k)
local copy
7zip -mx=9 -tgzip
  38136 bytes (37.2k)
local copy
pigz -11 -n
  38094 bytes (37.2k)
local copy
kzip -s0 -rn -b8
  38051 bytes (37.2k)
local copy
Zopfli
  37919 bytes (37.0k)
local copy
Zopfli (defluff)
  37918 bytes (37.0k)
local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/react/react-0.14.7.min.js

You will automatically get the smallest React 0.14.7 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 1634 bytes by using my React 0.14.7 Zopfli version instead of the best available CDN (4.31% smaller than unpkg, 37919 vs. 39553 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 --i1000000 --mb8 --mls4 --bsr17 --lazy --ohh

(found January 29, 2016)
Description Value Parameter
iterations 1000000  --i1000000
maximum blocks 8  --mb8
maximum length score 4  --mls4
block splitting recursion 17  --bsr17
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 (37918 bytes).

Verify file integrity

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

MD5:
curl --silent --compressed https://fb.me/react-0.14.7.min.js --location | md5sum
9ba549a9a66ab9c7e8eae26d39fcdc5f  -
curl --silent --compressed http://minime.stephan-brumme.com/files/react/react-0.14.7.min.zopfli.js.gz | md5sum
9ba549a9a66ab9c7e8eae26d39fcdc5f  -

SHA1:
curl --silent --compressed https://fb.me/react-0.14.7.min.js --location | sha1sum
8e11452bef99c160335fbcedde40cf7f0643fa26  -
curl --silent --compressed http://minime.stephan-brumme.com/files/react/react-0.14.7.min.zopfli.js.gz | sha1sum
8e11452bef99c160335fbcedde40cf7f0643fa26  -

All listed CDNs deliver identical contents:
CDN Size (compressed) MD5 (uncompressed) Timestamp
Boot 46857 bytes 9ba549a9a66ab9c7e8eae26d39fcdc5f January 29, 2016 @ 11:05
cdnhttps 39553 bytes 9ba549a9a66ab9c7e8eae26d39fcdc5f November 22, 2016 @ 16:40
cdnjs 39553 bytes 9ba549a9a66ab9c7e8eae26d39fcdc5f January 29, 2016 @ 02:02
jsdelivr 39553 bytes 9ba549a9a66ab9c7e8eae26d39fcdc5f January 28, 2016 @ 21:40
unpkg 39553 bytes 9ba549a9a66ab9c7e8eae26d39fcdc5f July 11, 2016 @ 15:33

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

Other Versions

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

16.3.1, 16.3.0,
16.2.0,
16.1.1, 16.1.0, 16.0.0,
15.6.2, 15.6.1, 15.6.0,
15.5.2, 15.5.1, 15.5.0,
15.4.2, 15.4.1, 15.4.0,
15.3.2, 15.3.1, 15.3.0,
15.2.1, 15.2.0,
15.1.0,
15.0.2, 15.0.1, 15.0.0,
0.14.8, 0.14.7, 0.14.6, 0.14.5, 0.14.4, 0.14.3, 0.14.2, 0.14.1, 0.14.0,
0.13.3, 0.13.2, 0.13.1, 0.13.0,
0.12.2, 0.12.1, 0.12.0,
0.11.2, 0.11.1, 0.11.0,
0.10.0

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, Socket.IO, ThreeJS, UnderscoreJS and Vue.

Changelog

Best Zopfli parameters so far:
Size Improvement Parameters Found
37919 bytes -3 bytes zopfli --i1000000 --mls4 --bsr17 --lazy --ohh January 29, 2016 @ 19:25
37922 bytes -5 bytes zopfli --i100000 --mls4 --bsr17 --lazy --ohh January 29, 2016 @ 11:30
37927 bytes -6 bytes zopfli --i10000 --mls4 --bsr17 --lazy --ohh January 29, 2016 @ 10:28
37933 bytes -14 bytes zopfli --i1000 --mls4 --bsr17 --lazy --ohh January 29, 2016 @ 10:20
37947 bytes zopfli --i100 --mls4 --bsr17 --lazy --ohh January 29, 2016 @ 10:15

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 @ 16:40.

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
37956 37957 38045 38048 38025 38026 38030 38046 38046 38048 38046 38046 38044 38046 38060
38044 38047 38048 38049 38045 38043 38045 38045 38047 38046 38047 38050 38046 38044 38045
37954 37965 38013 38020 38014 38015 38022 38042 38046 38046 38048 38043 38045 38042 38045
38025 38046 38046 38017 38042 38003 38025 38045 38046 38045 38048 38047 38042 38043 38049
37973 37974 37956 38016 38019 37946 38023 38047 38046 38044 38048 38047 38042 38042 38047
38048 38046 38046 38049 38043 38042 38042 38045 38044 38045 38048 38050 38048 38043 38044
38045 38044 38046 38042 38045 38042 38044 38047 38044 38046 38048 38046 38042 38043 38043
37968 38046 38021 38016 38018 38022 38022 38047 38045 38044 38048 38050 38048 38042 38045
37980 37957 38045 37947 38008 38001 38011 38047 38044 38044 38042 38045 38046 38046 38042
37964 37965 37963 37955 38024 38021 38011 38047 38041 38042 38043 38047 38045 38047 38054
37962 37960 37947 37955 38002 38016 38009 38042 38046 38044 38042 38046 38048 38045 38045
37945 38045 38046 37955 38023 38042 38042 38047 38046 38045 38048 38050 38049 38048 38043
38048 38046 38046 38020 38006 38003 38021 38047 38042 38044 38048 38049 38042 38043 38043
37946 37919 37953 37952 38022 38002 38021 38047 38046 38044 38045 38050 38049 38050 38044
37972 38047 38045 38044 38042 38042 38045 38042 38045 38046 38045 38046 38045 38042 38050
37955 38014 37946 37953 38021 38000 38022 38047 38045 38045 38045 38046 38042 38042 38045
37961 37974 37955 37954 38023 38022 38021 38047 38045 38046 38048 38046 38042 38045 38042
37954 37963 38021 37952 38007 38023 37941 38042 38046 38044 38041 38052 38045 38042 38046
37956 38044 38021 37957 38020 38024 38021 38045 38044 38042 38045 38046 38046 38039 38045
37962 37965 37946 37946 38021 37945 38005 38042 38045 38045 38048 38052 38045 38044 38043
37956 38044 37943 38044 38043 38045 38044 38045 38043 38042 38041 38050 38045 38044 38046
37970 38044 38046 37944 38021 38022 38013 38047 38043 38044 38041 38051 38046 38047 38049
37955 37963 37946 37996 38021 38003 38022 38047 38046 38046 38048 38050 38042 38045 38048

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 37947 bytes 100%
1,000 37933 bytes -14 bytes 100%
10,000 37927 bytes -6 bytes 100%
100,000 37922 bytes -5 bytes 4.06%
1,000,000 37919 bytes -3 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
38148 bytes +229 bytes (+0.60%) +97 bytes
38143 bytes +224 bytes (+0.59%) +92 bytes
38162 bytes +243 bytes (+0.64%) +111 bytes
38141 bytes +222 bytes (+0.59%) +90 bytes
38102 bytes +183 bytes (+0.48%) +51 bytes
38088 bytes +169 bytes (+0.45%) +37 bytes
38092 bytes +173 bytes (+0.46%) +41 bytes
38056 bytes +137 bytes (+0.36%) +5 bytes
38051 bytes +132 bytes (+0.35%)

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 27938 bytes -9981 bytes (-26.32%)
RAR (proprietary) RAR rar a -m5 -md64m -mc63:128t -mt1 30029 bytes -7890 bytes (-20.81%)
PPMd (Wikipedia) 7zip 7za a -mx=9 -m0=ppmd 31844 bytes -6075 bytes (-16.02%)
Brotli (Wikipedia) brotli brotli -q 11 33541 bytes -4378 bytes (-11.55%)
Burrows-Wheeler transform (Wikipedia) bzip2 bzip2 -9 33963 bytes -3956 bytes (-10.43%)
LZMA2 (Wikipedia) xz xz -9 34680 bytes -3239 bytes (-8.54%)
ZSTD (Wikipedia) zstd zstd -19 36139 bytes -1780 bytes (-4.69%)

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