You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
42942e0
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reactant.jl Benchmarks
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1325006950
ns1253914991
ns1.06
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1235896790
ns1313550312
ns0.94
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1260870880
ns1289530032
ns0.98
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2354818253
ns2471847121
ns0.95
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Lux
208350792
ns214028961
ns0.97
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
5855977297
ns5274924702
ns1.11
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant
5221394418
ns6071911149
ns0.86
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
5167730933
ns5130491661
ns1.01
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
7329557931
ns6918679637
ns1.06
ViT base (256 x 256 x 3 x 32)/forward/CPU/Lux
33799289512
ns36679290663
ns0.92
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1288968381
ns1311164623
ns0.98
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1367739710.5
ns1254537508.5
ns1.09
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1241844781
ns1191311446.5
ns1.04
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2383293970
ns2328680582
ns1.02
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Lux
8825641
ns8826361
ns1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1947817634
ns1565760477
ns1.24
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant
1582676667
ns1552005190
ns1.02
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1587980197
ns1537805276
ns1.03
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2835816387
ns2759960523
ns1.03
ViT small (256 x 256 x 3 x 4)/forward/CPU/Lux
3059951855
ns2921175858.5
ns1.05
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1383755457.5
ns1474222933
ns0.94
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1446515309
ns1239362666.5
ns1.17
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1261596782
ns1313069664
ns0.96
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2589819311
ns2508320871
ns1.03
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Lux
22684780
ns22697774
ns1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
2197648970
ns2145152905
ns1.02
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant
2228789213
ns2175744176
ns1.02
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
2201537016
ns2159120670
ns1.02
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
3498815180
ns3386821489
ns1.03
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Lux
5675046852
ns5778911115
ns0.98
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1286548945
ns1223003140.5
ns1.05
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1286795640
ns1247984959.5
ns1.03
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1307418594.5
ns1259430848
ns1.04
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2562260509
ns2391121753
ns1.07
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Lux
7586742.5
ns6544743
ns1.16
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1477416147
ns1415937281
ns1.04
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant
1459717161
ns1416011303
ns1.03
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
1427023275
ns1405580370
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
2666201644
ns2623237530
ns1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Lux
1191242722
ns1265262383
ns0.94
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1237383506.5
ns1237811756
ns1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1281328564
ns1325665225
ns0.97
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1283366518
ns1275756769.5
ns1.01
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2571097612
ns2532118794
ns1.02
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Lux
12172844
ns12249082
ns0.99
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
1744895489
ns1700918483
ns1.03
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant
1744209257
ns1688391116
ns1.03
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
1741523121
ns1691224758
ns1.03
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
2952374286
ns2909754771
ns1.01
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Lux
3161120339.5
ns3130725340.5
ns1.01
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1246653286
ns1426574699
ns0.87
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1235082506.5
ns1185837525
ns1.04
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1256777245.5
ns1211526304.5
ns1.04
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2464259574
ns2297841025
ns1.07
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Lux
27204139
ns27328050.5
ns1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
2222730518
ns2168675734
ns1.02
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant
2224096501
ns2140302273
ns1.04
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
2192837165
ns2158768532
ns1.02
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
3433421300
ns3378118459
ns1.02
ViT small (256 x 256 x 3 x 16)/forward/CPU/Lux
5511062562
ns6074321132.5
ns0.91
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme)
1275826558
ns1190081980
ns1.07
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant
1269696500.5
ns1244933579.5
ns1.02
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme)
1268437005
ns1238379610.5
ns1.02
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme)
2335849488
ns2441888225
ns0.96
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Lux
52614815
ns52611669
ns1.00
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme)
3004532618
ns3057790981
ns0.98
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant
3029206535
ns3095002996
ns0.98
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme)
3106275502
ns3013138214
ns1.03
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme)
4523804192
ns4316364644
ns1.05
ViT small (256 x 256 x 3 x 32)/forward/CPU/Lux
13248514271
ns15676777261
ns0.85
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme)
1287364685
ns1226707265
ns1.05
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant
1270728075
ns1243310175
ns1.02
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme)
1211872705.5
ns1236011148
ns0.98
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme)
2307284213
ns2376709482
ns0.97
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Lux
70823960.5
ns70675509
ns1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme)
3242934004
ns3385043788
ns0.96
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant
3261225461
ns3436194343
ns0.95
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme)
3276733542
ns3204367776
ns1.02
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme)
4672103022
ns4566944678
ns1.02
ViT base (256 x 256 x 3 x 16)/forward/CPU/Lux
12733986436
ns15182292720
ns0.84
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme)
1198686032
ns1229933217
ns0.97
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant
1202407868
ns1239348780.5
ns0.97
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme)
1268714295.5
ns1219935097.5
ns1.04
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme)
2558419074
ns2336056718
ns1.10
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Lux
20788095
ns20675012.5
ns1.01
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme)
1877899848
ns1848417271
ns1.02
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant
1956925868
ns1863893342
ns1.05
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme)
2314364652
ns1862653583
ns1.24
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme)
3172070657
ns3062369690
ns1.04
ViT base (256 x 256 x 3 x 4)/forward/CPU/Lux
4073780422
ns3894158827
ns1.05
This comment was automatically generated by workflow using github-action-benchmark.