[GH-PAGES] Updated website

triton-lang · Jun 3, 2024 · fce5ac3 · fce5ac3
1 parent be03d8b
commit fce5ac3
Show file tree

Hide file tree

Showing 55 changed files with 403 additions and 415 deletions.
diff --git a/main/.doctrees/environment.pickle b/main/.doctrees/environment.pickle
diff --git a/main/.doctrees/getting-started/tutorials/01-vector-add.doctree b/main/.doctrees/getting-started/tutorials/01-vector-add.doctree
diff --git a/main/.doctrees/getting-started/tutorials/02-fused-softmax.doctree b/main/.doctrees/getting-started/tutorials/02-fused-softmax.doctree
diff --git a/main/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree b/main/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree
diff --git a/main/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree b/main/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree
diff --git a/main/.doctrees/getting-started/tutorials/05-layer-norm.doctree b/main/.doctrees/getting-started/tutorials/05-layer-norm.doctree
diff --git a/main/.doctrees/getting-started/tutorials/06-fused-attention.doctree b/main/.doctrees/getting-started/tutorials/06-fused-attention.doctree
diff --git a/main/.doctrees/getting-started/tutorials/07-extern-functions.doctree b/main/.doctrees/getting-started/tutorials/07-extern-functions.doctree
diff --git a/main/.doctrees/getting-started/tutorials/08-grouped-gemm.doctree b/main/.doctrees/getting-started/tutorials/08-grouped-gemm.doctree
diff --git a/main/.doctrees/getting-started/tutorials/sg_execution_times.doctree b/main/.doctrees/getting-started/tutorials/sg_execution_times.doctree
diff --git a/main/.doctrees/python-api/generated/triton.language.arange.doctree b/main/.doctrees/python-api/generated/triton.language.arange.doctree
diff --git a/main/.doctrees/python-api/generated/triton.language.trans.doctree b/main/.doctrees/python-api/generated/triton.language.trans.doctree
diff --git a/main/.doctrees/python-api/triton.language.doctree b/main/.doctrees/python-api/triton.language.doctree
diff --git a/main/.doctrees/sg_execution_times.doctree b/main/.doctrees/sg_execution_times.doctree
diff --git a/main/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip b/main/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip
diff --git a/main/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip b/main/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip
diff --git a/main/_images/sphx_glr_01-vector-add_001.png b/main/_images/sphx_glr_01-vector-add_001.png
diff --git a/main/_images/sphx_glr_01-vector-add_thumb.png b/main/_images/sphx_glr_01-vector-add_thumb.png
diff --git a/main/_images/sphx_glr_02-fused-softmax_001.png b/main/_images/sphx_glr_02-fused-softmax_001.png
diff --git a/main/_images/sphx_glr_02-fused-softmax_thumb.png b/main/_images/sphx_glr_02-fused-softmax_thumb.png
diff --git a/main/_images/sphx_glr_03-matrix-multiplication_001.png b/main/_images/sphx_glr_03-matrix-multiplication_001.png
diff --git a/main/_images/sphx_glr_03-matrix-multiplication_002.png b/main/_images/sphx_glr_03-matrix-multiplication_002.png
diff --git a/main/_images/sphx_glr_03-matrix-multiplication_thumb.png b/main/_images/sphx_glr_03-matrix-multiplication_thumb.png
diff --git a/main/_images/sphx_glr_05-layer-norm_001.png b/main/_images/sphx_glr_05-layer-norm_001.png
diff --git a/main/_images/sphx_glr_05-layer-norm_thumb.png b/main/_images/sphx_glr_05-layer-norm_thumb.png
diff --git a/main/_images/sphx_glr_06-fused-attention_001.png b/main/_images/sphx_glr_06-fused-attention_001.png
diff --git a/main/_images/sphx_glr_06-fused-attention_002.png b/main/_images/sphx_glr_06-fused-attention_002.png
diff --git a/main/_images/sphx_glr_06-fused-attention_003.png b/main/_images/sphx_glr_06-fused-attention_003.png
diff --git a/main/_images/sphx_glr_06-fused-attention_thumb.png b/main/_images/sphx_glr_06-fused-attention_thumb.png
diff --git a/main/_images/sphx_glr_08-grouped-gemm_001.png b/main/_images/sphx_glr_08-grouped-gemm_001.png
diff --git a/main/_images/sphx_glr_08-grouped-gemm_thumb.png b/main/_images/sphx_glr_08-grouped-gemm_thumb.png
diff --git a/main/_sources/getting-started/tutorials/01-vector-add.rst.txt b/main/_sources/getting-started/tutorials/01-vector-add.rst.txt
@@ -230,20 +230,20 @@ We can now run the decorated function above. Pass `print_data=True` to see the p
 
     vector-add-performance:
                size       Triton        Torch
-    0        4096.0     8.000000     8.000000
+    0        4096.0     8.000000     9.600000
     1        8192.0    15.999999    15.999999
     2       16384.0    31.999999    31.999999
     3       32768.0    63.999998    63.999998
     4       65536.0   127.999995   127.999995
     5      131072.0   219.428568   219.428568
     6      262144.0   384.000001   384.000001
     7      524288.0   614.400016   614.400016
-    8     1048576.0   819.200021   819.200021
+    8     1048576.0   819.200021   768.000002
     9     2097152.0  1023.999964  1023.999964
-    10    4194304.0  1228.800031  1260.307736
+    10    4194304.0  1260.307736  1228.800031
     11    8388608.0  1424.695621  1404.342820
     12   16777216.0  1560.380965  1560.380965
-    13   33554432.0  1631.601649  1624.859540
+    13   33554432.0  1624.859540  1624.859540
     14   67108864.0  1669.706983  1662.646960
     15  134217728.0  1684.008546  1678.616907
 
@@ -253,7 +253,7 @@ We can now run the decorated function above. Pass `print_data=True` to see the p
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 6.091 seconds)
+   **Total running time of the script:** (0 minutes 12.734 seconds)
 
 
 .. _sphx_glr_download_getting-started_tutorials_01-vector-add.py:

diff --git a/main/_sources/getting-started/tutorials/02-fused-softmax.rst.txt b/main/_sources/getting-started/tutorials/02-fused-softmax.rst.txt
@@ -303,104 +303,104 @@ We will then compare its performance against (1) :code:`torch.softmax` and (2) t
 
     softmax-performance:
               N       Triton        Torch
-    0     256.0   465.463076   675.883749
-    1     384.0   603.079841   830.079190
-    2     512.0   753.161944   914.066855
-    3     640.0   794.511138   948.204968
-    4     768.0   876.875005  1028.022329
-    5     896.0   924.292555  1063.979061
-    6    1024.0   992.233743  1107.052681
-    7    1152.0  1106.869379   617.406361
-    8    1280.0  1137.945170   667.999010
-    9    1408.0  1162.981028   729.775076
-    10   1536.0  1185.488174   786.231422
-    11   1664.0  1216.467604   818.384227
-    12   1792.0  1244.282941   858.164107
-    13   1920.0  1255.540209   914.369932
-    14   2048.0  1276.738954   964.849517
-    15   2176.0  1264.639989   976.216613
-    16   2304.0  1266.730342  1013.633062
-    17   2432.0  1295.641882  1056.499041
-    18   2560.0  1297.153183  1093.460116
-    19   2688.0  1313.661261  1111.337888
-    20   2816.0  1329.564965  1140.557943
-    21   2944.0  1331.614829  1176.775172
-    22   3072.0  1341.051904  1191.683308
-    23   3200.0  1349.888142  1202.503826
-    24   3328.0  1358.588367  1229.027717
-    25   3456.0  1377.428691  1250.427581
-    26   3584.0  1374.510821  1268.784121
-    27   3712.0  1377.176722  1279.071948
-    28   3840.0  1381.287203  1301.882679
-    29   3968.0  1389.064106  1318.881448
-    30   4096.0  1395.474538  1331.512185
-    31   4224.0  1345.556647  1166.650002
-    32   4352.0  1342.369741  1174.156135
-    33   4480.0  1351.710851  1178.985727
-    34   4608.0  1354.019193  1194.362631
-    35   4736.0  1362.612599  1202.738098
-    36   4864.0  1369.436098  1220.977976
-    37   4992.0  1375.796474  1232.807729
-    38   5120.0  1376.351827  1256.846607
-    39   5248.0  1379.287461  1265.840070
-    40   5376.0  1373.963366  1284.049845
-    41   5504.0  1382.889222  1296.432798
-    42   5632.0  1393.714894  1316.893384
-    43   5760.0  1398.826036  1327.094235
-    44   5888.0  1396.044465  1343.728873
-    45   6016.0  1399.249264  1350.663263
-    46   6144.0  1408.159052  1373.836110
-    47   6272.0  1415.220403  1377.667866
-    48   6400.0  1415.589407  1390.342081
-    49   6528.0  1415.519673  1395.398534
-    50   6656.0  1418.895634  1404.581657
-    51   6784.0  1418.203402  1413.922009
-    52   6912.0  1420.279835  1426.416621
-    53   7040.0  1425.235022  1428.082469
-    54   7168.0  1431.983130  1438.079289
-    55   7296.0  1432.106994  1445.071214
-    56   7424.0  1426.862629  1449.840203
-    57   7552.0  1429.480220  1455.603978
-    58   7680.0  1436.278797  1463.003647
-    59   7808.0  1434.404832  1464.748231
-    60   7936.0  1432.440983  1470.346752
-    61   8064.0  1431.421053  1473.393801
-    62   8192.0  1432.095196  1484.318597
-    63   8320.0  1387.450876  1402.386272
-    64   8448.0  1384.443465  1405.896562
-    65   8576.0  1392.396541  1396.134852
-    66   8704.0  1390.734237  1403.150266
-    67   8832.0  1391.962230  1404.885386
-    68   8960.0  1392.497146  1414.524497
-    69   9088.0  1405.587214  1418.987296
-    70   9216.0  1403.700211  1425.223331
-    71   9344.0  1402.034764  1423.639872
-    72   9472.0  1391.304006  1433.051585
-    73   9600.0  1402.086866  1437.097020
-    74   9728.0  1404.221575  1444.898518
-    75   9856.0  1405.529715  1441.766156
-    76   9984.0  1392.189223  1451.555775
-    77  10112.0  1413.514547  1459.581911
-    78  10240.0  1411.416652  1469.622297
-    79  10368.0  1417.430944  1465.406049
-    80  10496.0  1423.622408  1469.707135
-    81  10624.0  1404.217003  1469.395778
-    82  10752.0  1394.721768  1473.584987
-    83  10880.0  1394.388858  1478.529736
-    84  11008.0  1414.236727  1477.103531
-    85  11136.0  1428.479102  1484.901506
-    86  11264.0  1413.707974  1487.465253
-    87  11392.0  1419.858471  1491.152346
-    88  11520.0  1418.984065  1497.066641
-    89  11648.0  1427.240185  1503.509443
-    90  11776.0  1424.484961  1506.276858
-    91  11904.0  1435.529476  1506.144435
-    92  12032.0  1412.830212  1506.302588
-    93  12160.0  1414.303931  1512.440974
-    94  12288.0  1431.206105  1393.859097
-    95  12416.0  1444.323508  1389.925161
-    96  12544.0  1444.659330  1394.155304
-    97  12672.0  1439.147480  1394.767174
+    0     256.0   464.208270   685.567100
+    1     384.0   613.034238   833.588488
+    2     512.0   742.107385   933.396745
+    3     640.0   784.829491   953.456602
+    4     768.0   883.112141  1029.245440
+    5     896.0   925.616578  1061.595575
+    6    1024.0   988.144964  1107.437062
+    7    1152.0  1095.380654   613.835111
+    8    1280.0  1125.682650   667.566978
+    9    1408.0  1160.258320   728.707704
+    10   1536.0  1186.856355   781.906519
+    11   1664.0  1218.672169   818.538462
+    12   1792.0  1234.958085   864.437934
+    13   1920.0  1253.925816   913.292283
+    14   2048.0  1283.267570   964.876493
+    15   2176.0  1265.208184   980.152298
+    16   2304.0  1276.154236  1015.530882
+    17   2432.0  1297.684672  1060.570442
+    18   2560.0  1305.432667  1088.103332
+    19   2688.0  1303.448743  1110.818692
+    20   2816.0  1333.043184  1138.682445
+    21   2944.0  1334.895568  1171.994555
+    22   3072.0  1338.827475  1190.784357
+    23   3200.0  1356.229960  1202.819132
+    24   3328.0  1363.892928  1224.191464
+    25   3456.0  1373.813634  1249.795156
+    26   3584.0  1374.110970  1267.575254
+    27   3712.0  1381.976421  1277.787412
+    28   3840.0  1388.337080  1300.063924
+    29   3968.0  1387.310679  1316.218574
+    30   4096.0  1396.271471  1336.655074
+    31   4224.0  1337.963082  1165.825985
+    32   4352.0  1344.306639  1174.675905
+    33   4480.0  1346.919186  1179.570076
+    34   4608.0  1353.163760  1193.801619
+    35   4736.0  1366.687041  1201.531024
+    36   4864.0  1373.398011  1224.197699
+    37   4992.0  1373.307182  1231.385849
+    38   5120.0  1378.325890  1255.259130
+    39   5248.0  1386.784266  1266.373325
+    40   5376.0  1375.944222  1287.808601
+    41   5504.0  1386.185269  1292.886755
+    42   5632.0  1388.705630  1317.770250
+    43   5760.0  1392.595882  1323.054006
+    44   5888.0  1394.147605  1341.572919
+    45   6016.0  1407.443309  1355.983540
+    46   6144.0  1405.743196  1377.071142
+    47   6272.0  1415.395700  1378.892315
+    48   6400.0  1412.205899  1387.798460
+    49   6528.0  1415.619340  1393.698736
+    50   6656.0  1414.964809  1406.502041
+    51   6784.0  1421.732775  1416.353765
+    52   6912.0  1422.026364  1424.473579
+    53   7040.0  1423.856017  1428.405822
+    54   7168.0  1427.631832  1437.204447
+    55   7296.0  1437.137368  1441.185345
+    56   7424.0  1424.496347  1446.821089
+    57   7552.0  1428.458077  1457.000314
+    58   7680.0  1437.528114  1465.173283
+    59   7808.0  1435.005803  1468.999534
+    60   7936.0  1429.186342  1468.328022
+    61   8064.0  1434.868908  1475.777155
+    62   8192.0  1434.400445  1487.348939
+    63   8320.0  1391.820963  1404.182529
+    64   8448.0  1382.973572  1406.142272
+    65   8576.0  1389.176158  1394.002859
+    66   8704.0  1393.185900  1400.028952
+    67   8832.0  1391.951855  1401.672919
+    68   8960.0  1389.723705  1413.853767
+    69   9088.0  1401.625259  1417.971826
+    70   9216.0  1401.500746  1424.381804
+    71   9344.0  1407.856930  1423.619060
+    72   9472.0  1395.023459  1437.535148
+    73   9600.0  1403.158579  1436.856888
+    74   9728.0  1402.536318  1440.640117
+    75   9856.0  1405.329217  1443.677466
+    76   9984.0  1392.176246  1451.279340
+    77  10112.0  1416.777083  1459.321080
+    78  10240.0  1410.083478  1470.362934
+    79  10368.0  1416.877320  1462.958361
+    80  10496.0  1420.340864  1471.438067
+    81  10624.0  1404.631452  1470.282238
+    82  10752.0  1397.947393  1474.370728
+    83  10880.0  1396.443035  1479.095530
+    84  11008.0  1412.546116  1477.946304
+    85  11136.0  1425.384037  1485.883593
+    86  11264.0  1422.965083  1486.511253
+    87  11392.0  1421.947287  1492.362407
+    88  11520.0  1421.851704  1496.315547
+    89  11648.0  1431.961825  1500.304749
+    90  11776.0  1433.233056  1500.152813
+    91  11904.0  1435.456897  1507.371476
+    92  12032.0  1413.574913  1507.528158
+    93  12160.0  1424.625752  1510.193976
+    94  12288.0  1430.675218  1391.484278
+    95  12416.0  1445.268023  1390.384811
+    96  12544.0  1447.392631  1393.468635
+    97  12672.0  1446.409532  1394.889076
 
 
 
@@ -415,7 +415,7 @@ In the above plot, we can see that:
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 35.652 seconds)
+   **Total running time of the script:** (0 minutes 36.993 seconds)
 
 
 .. _sphx_glr_download_getting-started_tutorials_02-fused-softmax.py:

diff --git a/main/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt b/main/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt
@@ -570,45 +570,45 @@ but feel free to arrange this script as you wish to benchmark any other matrix s
     matmul-performance-fp16:
              M       N       K      cuBLAS      Triton
     0    256.0   256.0   256.0    4.096000    4.096000
-    1    384.0   384.0   384.0   12.288000   12.288000
+    1    384.0   384.0   384.0   11.059200   12.288000
     2    512.0   512.0   512.0   26.214401   26.214401
     3    640.0   640.0   640.0   42.666665   42.666665
     4    768.0   768.0   768.0   68.056616   68.056616
-    5    896.0   896.0   896.0   78.051553   87.808000
+    5    896.0   896.0   896.0   78.051553   93.661869
     6   1024.0  1024.0  1024.0  104.857603   99.864382
     7   1152.0  1152.0  1152.0  135.726544  129.825388
     8   1280.0  1280.0  1280.0  163.840004  163.840004
     9   1408.0  1408.0  1408.0  155.765024  132.970149
     10  1536.0  1536.0  1536.0  181.484314  157.286398
     11  1664.0  1664.0  1664.0  183.651271  183.651271
-    12  1792.0  1792.0  1792.0  175.616000  208.137481
-    13  1920.0  1920.0  1920.0  203.294114  168.585369
-    14  2048.0  2048.0  2048.0  226.719125  192.841562
-    15  2176.0  2176.0  2176.0  214.081356  214.081356
-    16  2304.0  2304.0  2304.0  234.194828  234.194828
-    17  2432.0  2432.0  2432.0  205.069087  205.069087
-    18  2560.0  2560.0  2560.0  225.986210  218.453323
-    19  2688.0  2688.0  2688.0  198.602388  201.771569
-    20  2816.0  2816.0  2816.0  213.795141  213.795141
-    21  2944.0  2944.0  2944.0  221.493479  222.482283
-    22  3072.0  3072.0  3072.0  208.941345  211.280236
-    23  3200.0  3200.0  3200.0  219.931269  217.687077
-    24  3328.0  3328.0  3328.0  206.871539  208.067338
-    25  3456.0  3456.0  3456.0  217.308808  223.328435
-    26  3584.0  3584.0  3584.0  222.562856  213.069643
-    27  3712.0  3712.0  3712.0  213.000737  221.011819
-    28  3840.0  3840.0  3840.0  211.456969  210.250955
-    29  3968.0  3968.0  3968.0  211.114084  219.073497
-    30  4096.0  4096.0  4096.0  220.029067  216.829933
+    12  1792.0  1792.0  1792.0  172.914215  212.064605
+    13  1920.0  1920.0  1920.0  203.294114  170.666670
+    14  2048.0  2048.0  2048.0  229.824868  195.083907
+    15  2176.0  2176.0  2176.0  211.827867  214.081356
+    16  2304.0  2304.0  2304.0  229.691080  234.194828
+    17  2432.0  2432.0  2432.0  205.069087  203.583068
+    18  2560.0  2560.0  2560.0  224.438347  219.919464
+    19  2688.0  2688.0  2688.0  200.704002  200.704002
+    20  2816.0  2816.0  2816.0  214.848312  212.752230
+    21  2944.0  2944.0  2944.0  221.493479  224.486628
+    22  3072.0  3072.0  3072.0  208.941345  209.715208
+    23  3200.0  3200.0  3200.0  219.931269  225.352106
+    24  3328.0  3328.0  3328.0  208.067338  211.739107
+    25  3456.0  3456.0  3456.0  219.080343  222.097984
+    26  3584.0  3584.0  3584.0  217.186932  208.137481
+    27  3712.0  3712.0  3712.0  211.199462  221.501871
+    28  3840.0  3840.0  3840.0  211.862062  213.086708
+    29  3968.0  3968.0  3968.0  212.585252  220.658104
+    30  4096.0  4096.0  4096.0  219.668951  217.180793
     matmul-performance-fp8:
              M       N       K      Triton
-    0    256.0   256.0   256.0    2.978909
+    0    256.0   256.0   256.0    3.276800
     1    384.0   384.0   384.0   10.053818
     2    512.0   512.0   512.0   20.164923
     3    640.0   640.0   640.0   34.133334
     4    768.0   768.0   768.0   42.130286
-    5    896.0   896.0   896.0   61.083825
-    6   1024.0  1024.0  1024.0   61.680940
+    5    896.0   896.0   896.0   58.538665
+    6   1024.0  1024.0  1024.0   63.550060
     7   1152.0  1152.0  1152.0   80.702267
     8   1280.0  1280.0  1280.0  102.400003
     9   1408.0  1408.0  1408.0   82.602666
@@ -618,29 +618,29 @@ but feel free to arrange this script as you wish to benchmark any other matrix s
     13  1920.0  1920.0  1920.0  100.905113
     14  2048.0  2048.0  2048.0  114.912434
     15  2176.0  2176.0  2176.0  121.226797
-    16  2304.0  2304.0  2304.0  135.726544
-    17  2432.0  2432.0  2432.0  133.783163
-    18  2560.0  2560.0  2560.0  148.271497
-    19  2688.0  2688.0  2688.0  118.540803
-    20  2816.0  2816.0  2816.0  129.419013
-    21  2944.0  2944.0  2944.0  140.779756
-    22  3072.0  3072.0  3072.0  144.446699
-    23  3200.0  3200.0  3200.0  140.969165
+    16  2304.0  2304.0  2304.0  134.201527
+    17  2432.0  2432.0  2432.0  132.521057
+    18  2560.0  2560.0  2560.0  146.941707
+    19  2688.0  2688.0  2688.0  118.171514
+    20  2816.0  2816.0  2816.0  129.804192
+    21  2944.0  2944.0  2944.0  139.596724
+    22  3072.0  3072.0  3072.0  144.816120
+    23  3200.0  3200.0  3200.0  139.433550
     24  3328.0  3328.0  3328.0  131.611151
-    25  3456.0  3456.0  3456.0  138.763456
-    26  3584.0  3584.0  3584.0  149.609643
-    27  3712.0  3712.0  3712.0  141.097948
-    28  3840.0  3840.0  3840.0  139.812890
-    29  3968.0  3968.0  3968.0  147.460947
-    30  4096.0  4096.0  4096.0  155.885867
+    25  3456.0  3456.0  3456.0  140.455692
+    26  3584.0  3584.0  3584.0  149.858980
+    27  3712.0  3712.0  3712.0  141.899635
+    28  3840.0  3840.0  3840.0  139.989870
+    29  3968.0  3968.0  3968.0  147.194128
+    30  4096.0  4096.0  4096.0  157.347868
 
 
 
 
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (2 minutes 17.827 seconds)
+   **Total running time of the script:** (2 minutes 18.597 seconds)
 
 
 .. _sphx_glr_download_getting-started_tutorials_03-matrix-multiplication.py:

diff --git a/main/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt b/main/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt
@@ -242,7 +242,7 @@ References
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 0.750 seconds)
+   **Total running time of the script:** (0 minutes 0.767 seconds)
 
 
 .. _sphx_glr_download_getting-started_tutorials_04-low-memory-dropout.py: