Skip to content

Conversation

@xys-syx
Copy link
Collaborator

@xys-syx xys-syx commented Jan 31, 2026

I have uncommented cudart-to-hiprt for mapping cudaFree to hipFree directly (temp maybe)

Copy link
Contributor

@github-actions github-actions bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

EnzymeJAX Benchmarks

Details
Benchmark suite Current: 5a16c94 Previous: 2d3922c Ratio
actmtch / JaXPipe / cpu / Primal 0.000007652479989701533 s 0.000007271839967870619 s 1.05
actmtch / Jax / cpu / Primal 0.000007178480009315536 s 0.000006975079986659693 s 1.03
actmtch / HLOOpt / cpu / Primal 0.00001074128001164354 s 0.000010773200028779685 s 1.00
actmtch / PartOpt / cpu / Primal 0.000007115539956430439 s 0.000007492480008295388 s 0.95
actmtch / IPartOpt / cpu / Primal 0.000006436400017264532 s 0.000007248080019053304 s 0.89
actmtch / DefOpt / cpu / Primal 0.000011509820005812798 s 0.000011577819959711633 s 0.99
actmtch / IDefOpt / cpu / Primal 0.000007375179993687198 s 0.000007383559986919863 s 1.00
actmtch / JaXPipe / cpu / Forward 0.000011184219965798548 s 0.00001151564002611849 s 0.97
actmtch / Jax / cpu / Forward 0.000010098319989992888 s 0.000010873160026676488 s 0.93
actmtch / HLOOpt / cpu / Forward 0.000015319559979616313 s 0.000016168559977813856 s 0.95
actmtch / PartOpt / cpu / Forward 0.00001534057997560012 s 0.00001610137996067351 s 0.95
actmtch / IPartOpt / cpu / Forward 0.000011291780001556616 s 0.000011685460012813565 s 0.97
actmtch / DefOpt / cpu / Forward 0.000015777299986439174 s 0.000015440160022990313 s 1.02
actmtch / IDefOpt / cpu / Forward 0.000011285239988865214 s 0.0000117808800223429 s 0.96
actmtch / JaXPipe / cpu / PreRev 0.000012638959979085484 s 0.000012508640020314488 s 1.01
actmtch / JaXPipe / cpu / PostRev 0.00001147341999967466 s 0.000011967219988946451 s 0.96
actmtch / JaXPipe / cpu / BothRev 0.000011951019978369004 s 0.000011937360013689614 s 1.00
actmtch / Jax / cpu / BothRev 0.000010551060013312964 s 0.000010844939961316411 s 0.97
actmtch / HLOOpt / cpu / PreRev 0.00001196261999211856 s 0.000012242080047144554 s 0.98
actmtch / HLOOpt / cpu / PostRev 0.000016141859960043802 s 0.000016901179997148574 s 0.96
actmtch / HLOOpt / cpu / BothRev 0.000013882880029996158 s 0.000013715339982809382 s 1.01
actmtch / PartOpt / cpu / PreRev 0.00001182265997158538 s 0.000012723280024147243 s 0.93
actmtch / PartOpt / cpu / PostRev 0.00001063382000211277 s 0.00001106601998799306 s 0.96
actmtch / PartOpt / cpu / BothRev 0.000012303600005907355 s 0.000011431200018705567 s 1.08
actmtch / IPartOpt / cpu / PreRev 0.000012121579948143336 s 0.000012506260036388994 s 0.97
actmtch / IPartOpt / cpu / PostRev 0.000011271580005995929 s 0.000010934379988611908 s 1.03
actmtch / IPartOpt / cpu / BothRev 0.000011822499973277444 s 0.000012168219982413576 s 0.97
actmtch / DefOpt / cpu / PreRev 0.000012498920023062964 s 0.000012047140044160188 s 1.04
actmtch / DefOpt / cpu / PostRev 0.000012044160002915306 s 0.000012444979984138626 s 0.97
actmtch / DefOpt / cpu / BothRev 0.000011822779997601174 s 0.000011966699958065874 s 0.99
actmtch / IDefOpt / cpu / PreRev 0.000011691359977703542 s 0.000012383379971652177 s 0.94
actmtch / IDefOpt / cpu / PostRev 0.00001230864000717702 s 0.000011945180058319236 s 1.03
actmtch / IDefOpt / cpu / BothRev 0.000011932120005440085 s 0.000011602779995882883 s 1.03
actmtch / JaXPipe / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / Jax / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / HLOOpt / cuda / Primal 0.000002015 s 0.000002016 s 1.00
actmtch / PartOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / IPartOpt / cuda / Primal 0.000002016 s 0.000002047 s 0.98
actmtch / DefOpt / cuda / Primal 0.000002015 s 0.000002047 s 0.98
actmtch / IDefOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / JaXPipe / cuda / Forward 0.000010399 s 0.00001024 s 1.02
actmtch / Jax / cuda / Forward 0.000010272 s 0.000010431 s 0.98
actmtch / HLOOpt / cuda / Forward 0.000010144 s 0.000010272 s 0.99
actmtch / PartOpt / cuda / Forward 0.00001056 s 0.00001024 s 1.03
actmtch / IPartOpt / cuda / Forward 0.000010112 s 0.0000104 s 0.97
actmtch / DefOpt / cuda / Forward 0.000010304 s 0.000010112 s 1.02
actmtch / IDefOpt / cuda / Forward 0.000010433 s 0.000009856 s 1.06
actmtch / JaXPipe / cuda / PreRev 0.000010496 s 0.000010592 s 0.99
actmtch / JaXPipe / cuda / PostRev 0.000010367 s 0.000012512 s 0.83
actmtch / JaXPipe / cuda / BothRev 0.000010112 s 0.000010016 s 1.01
actmtch / Jax / cuda / BothRev 0.000010207 s 0.000009344 s 1.09
actmtch / HLOOpt / cuda / PreRev 0.000010112 s 0.000010592 s 0.95
actmtch / HLOOpt / cuda / PostRev 0.000010208 s 0.000010111 s 1.01
actmtch / HLOOpt / cuda / BothRev 0.000011328 s 0.000009631 s 1.18
actmtch / PartOpt / cuda / PreRev 0.000009696 s 0.000009984 s 0.97
actmtch / PartOpt / cuda / PostRev 0.000011296 s 0.000009664 s 1.17
actmtch / PartOpt / cuda / BothRev 0.000014432 s 0.00001136 s 1.27
actmtch / IPartOpt / cuda / PreRev 0.000010368 s 0.00001024 s 1.01
actmtch / IPartOpt / cuda / PostRev 0.000011136 s 0.00001056 s 1.05
actmtch / IPartOpt / cuda / BothRev 0.000014049 s 0.000009984 s 1.41
actmtch / DefOpt / cuda / PreRev 0.000011424 s 0.000009824 s 1.16
actmtch / DefOpt / cuda / PostRev 0.000010111 s 0.000010176 s 0.99
actmtch / DefOpt / cuda / BothRev 0.000010079 s 0.000010304 s 0.98
actmtch / IDefOpt / cuda / PreRev 0.00001008 s 0.000015136 s 0.67
actmtch / IDefOpt / cuda / PostRev 0.000013184 s 0.000010145 s 1.30
actmtch / IDefOpt / cuda / BothRev 0.000010176 s 0.000010176 s 1
actmtch / JaXPipe / tpu / Primal 5.631e-7 s 5.63725e-7 s 1.00
actmtch / Jax / tpu / Primal 6.06075e-7 s 6.06725e-7 s 1.00
actmtch / HLOOpt / tpu / Primal 0.0000020960500000000005 s 0.0000020957 s 1.00
actmtch / PartOpt / tpu / Primal 6.064250000000001e-7 s 6.06925e-7 s 1.00
actmtch / IPartOpt / tpu / Primal 5.62625e-7 s 5.6255e-7 s 1.00
actmtch / DefOpt / tpu / Primal 0.000002163125 s 0.000002171175 s 1.00
actmtch / IDefOpt / tpu / Primal 0.00000210075 s 0.0000021082 s 1.00
actmtch / JaXPipe / tpu / Forward 0.00000383275 s 0.0000038315 s 1.00
actmtch / Jax / tpu / Forward 0.0000012187 s 0.000001216425 s 1.00
actmtch / HLOOpt / tpu / Forward 0.000003937025000000001 s 0.000003943575000000001 s 1.00
actmtch / PartOpt / tpu / Forward 0.000003905725 s 0.0000039182 s 1.00
actmtch / IPartOpt / tpu / Forward 0.00000393625 s 0.000003929 s 1.00
actmtch / DefOpt / tpu / Forward 0.0000039234500000000005 s 0.000003917075 s 1.00
actmtch / IDefOpt / tpu / Forward 0.000003938775 s 0.000003942675 s 1.00
actmtch / JaXPipe / tpu / PreRev 0.0000034596 s 0.000003492125 s 0.99
actmtch / JaXPipe / tpu / PostRev 0.000001640175 s 0.0000016393 s 1.00
actmtch / JaXPipe / tpu / BothRev 0.00000349635 s 0.0000034905000000000003 s 1.00
actmtch / Jax / tpu / BothRev 0.000001637325 s 0.0000016388749999999998 s 1.00
actmtch / HLOOpt / tpu / PreRev 0.0000034808 s 0.0000034716 s 1.00
actmtch / HLOOpt / tpu / PostRev 0.0000034136 s 0.000003411375 s 1.00
actmtch / HLOOpt / tpu / BothRev 0.000003461825 s 0.000003464375 s 1.00
actmtch / PartOpt / tpu / PreRev 0.000003417775 s 0.000003414475 s 1.00
actmtch / PartOpt / tpu / PostRev 0.0000015890000000000002 s 0.00000158595 s 1.00
actmtch / PartOpt / tpu / BothRev 0.00000339675 s 0.000003404 s 1.00
actmtch / IPartOpt / tpu / PreRev 0.000003465975 s 0.00000347615 s 1.00
actmtch / IPartOpt / tpu / PostRev 0.000001640725 s 0.000001637025 s 1.00
actmtch / IPartOpt / tpu / BothRev 0.0000034635250000000004 s 0.00000347785 s 1.00
actmtch / DefOpt / tpu / PreRev 0.00000341735 s 0.00000341665 s 1.00
actmtch / DefOpt / tpu / PostRev 0.000003419175 s 0.0000034146 s 1.00
actmtch / DefOpt / tpu / BothRev 0.0000034063 s 0.0000034134 s 1.00
actmtch / IDefOpt / tpu / PreRev 0.000003474575 s 0.000003477675 s 1.00
actmtch / IDefOpt / tpu / PostRev 0.0000034202250000000004 s 0.00000342645 s 1.00
actmtch / IDefOpt / tpu / BothRev 0.000003476925 s 0.00000347685 s 1.00
actmtch / JaXPipe / cpu / Primal 0.000016714 s 0.000007271839967870619 s 2.30
actmtch / Jax / cpu / Primal 0.00001655 s 0.000006975079986659693 s 2.37
actmtch / HLOOpt / cpu / Primal 0.000017860000000000002 s 0.000010773200028779685 s 1.66
actmtch / PartOpt / cpu / Primal 0.000016664000000000002 s 0.000007492480008295388 s 2.22
actmtch / IPartOpt / cpu / Primal 0.00001647 s 0.000007248080019053304 s 2.27
actmtch / DefOpt / cpu / Primal 0.000017603 s 0.000011577819959711633 s 1.52
actmtch / IDefOpt / cpu / Primal 0.000017402 s 0.000007383559986919863 s 2.36
actmtch / JaXPipe / cpu / Forward 0.000023704 s 0.00001151564002611849 s 2.06
actmtch / Jax / cpu / Forward 0.000022525 s 0.000010873160026676488 s 2.07
actmtch / HLOOpt / cpu / Forward 0.000023838 s 0.000016168559977813856 s 1.47
actmtch / PartOpt / cpu / Forward 0.00002362 s 0.00001610137996067351 s 1.47
actmtch / IPartOpt / cpu / Forward 0.000023926 s 0.000011685460012813565 s 2.05
actmtch / DefOpt / cpu / Forward 0.000024858 s 0.000015440160022990313 s 1.61
actmtch / IDefOpt / cpu / Forward 0.000024183 s 0.0000117808800223429 s 2.05
actmtch / JaXPipe / cpu / PreRev 0.000024585 s 0.000012508640020314488 s 1.97
actmtch / JaXPipe / cpu / PostRev 0.000022071 s 0.000011967219988946451 s 1.84
actmtch / JaXPipe / cpu / BothRev 0.000024992 s 0.000011937360013689614 s 2.09
actmtch / Jax / cpu / BothRev 0.000022367 s 0.000010844939961316411 s 2.06
actmtch / HLOOpt / cpu / PreRev 0.000023674 s 0.000012242080047144554 s 1.93
actmtch / HLOOpt / cpu / PostRev 0.000024746 s 0.000016901179997148574 s 1.46
actmtch / HLOOpt / cpu / BothRev 0.000024427 s 0.000013715339982809382 s 1.78
actmtch / PartOpt / cpu / PreRev 0.00002455 s 0.000012723280024147243 s 1.93
actmtch / PartOpt / cpu / PostRev 0.000022315 s 0.00001106601998799306 s 2.02
actmtch / PartOpt / cpu / BothRev 0.000024377 s 0.000011431200018705567 s 2.13
actmtch / IPartOpt / cpu / PreRev 0.000024427 s 0.000012506260036388994 s 1.95
actmtch / IPartOpt / cpu / PostRev 0.000021967 s 0.000010934379988611908 s 2.01
actmtch / IPartOpt / cpu / BothRev 0.000024555 s 0.000012168219982413576 s 2.02
actmtch / DefOpt / cpu / PreRev 0.00002442 s 0.000012047140044160188 s 2.03
actmtch / DefOpt / cpu / PostRev 0.000024793 s 0.000012444979984138626 s 1.99
actmtch / DefOpt / cpu / BothRev 0.000024591 s 0.000011966699958065874 s 2.05
actmtch / IDefOpt / cpu / PreRev 0.000023835 s 0.000012383379971652177 s 1.92
actmtch / IDefOpt / cpu / PostRev 0.000024634 s 0.000011945180058319236 s 2.06
actmtch / IDefOpt / cpu / BothRev 0.000024657 s 0.000011602779995882883 s 2.13
actmtch / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000007271839967870619 s 1.24
actmtch / Jax / cpu / Primal 0.000008999999999999999 s 0.000006975079986659693 s 1.29
actmtch / HLOOpt / cpu / Primal 0.00001 s 0.000010773200028779685 s 0.93
actmtch / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000007492480008295388 s 1.20
actmtch / IPartOpt / cpu / Primal 0.000008999999999999999 s 0.000007248080019053304 s 1.24
actmtch / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000011577819959711633 s 0.78
actmtch / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000007383559986919863 s 1.22
actmtch / JaXPipe / cpu / Forward 0.000014 s 0.00001151564002611849 s 1.22
actmtch / Jax / cpu / Forward 0.000012 s 0.000010873160026676488 s 1.10
actmtch / HLOOpt / cpu / Forward 0.000014 s 0.000016168559977813856 s 0.87
actmtch / PartOpt / cpu / Forward 0.000013 s 0.00001610137996067351 s 0.81
actmtch / IPartOpt / cpu / Forward 0.000014 s 0.000011685460012813565 s 1.20
actmtch / DefOpt / cpu / Forward 0.000014 s 0.000015440160022990313 s 0.91
actmtch / IDefOpt / cpu / Forward 0.000014 s 0.0000117808800223429 s 1.19
actmtch / JaXPipe / cpu / PreRev 0.000013 s 0.000012508640020314488 s 1.04
actmtch / JaXPipe / cpu / PostRev 0.000012 s 0.000011967219988946451 s 1.00
actmtch / JaXPipe / cpu / BothRev 0.000013 s 0.000011937360013689614 s 1.09
actmtch / Jax / cpu / BothRev 0.000012 s 0.000010844939961316411 s 1.11
actmtch / HLOOpt / cpu / PreRev 0.000013 s 0.000012242080047144554 s 1.06
actmtch / HLOOpt / cpu / PostRev 0.000013 s 0.000016901179997148574 s 0.77
actmtch / HLOOpt / cpu / BothRev 0.000014 s 0.000013715339982809382 s 1.02
actmtch / PartOpt / cpu / PreRev 0.000014 s 0.000012723280024147243 s 1.10
actmtch / PartOpt / cpu / PostRev 0.000013 s 0.00001106601998799306 s 1.17
actmtch / PartOpt / cpu / BothRev 0.000013 s 0.000011431200018705567 s 1.14
actmtch / IPartOpt / cpu / PreRev 0.000013 s 0.000012506260036388994 s 1.04
actmtch / IPartOpt / cpu / PostRev 0.000012 s 0.000010934379988611908 s 1.10
actmtch / IPartOpt / cpu / BothRev 0.000013 s 0.000012168219982413576 s 1.07
actmtch / DefOpt / cpu / PreRev 0.000013 s 0.000012047140044160188 s 1.08
actmtch / DefOpt / cpu / PostRev 0.000013 s 0.000012444979984138626 s 1.04
actmtch / DefOpt / cpu / BothRev 0.000013 s 0.000011966699958065874 s 1.09
actmtch / IDefOpt / cpu / PreRev 0.000014 s 0.000012383379971652177 s 1.13
actmtch / IDefOpt / cpu / PostRev 0.000014 s 0.000011945180058319236 s 1.17
actmtch / IDefOpt / cpu / BothRev 0.000013 s 0.000011602779995882883 s 1.12
add_one / JaXPipe / cpu / Primal 0.000007117319928511278 s 0.000007327479988816776 s 0.97
add_one / Jax / cpu / Primal 0.000007010019999142969 s 0.000007088460015438613 s 0.99
add_one / HLOOpt / cpu / Primal 0.000010567860044830011 s 0.000011369959956937236 s 0.93
add_one / PartOpt / cpu / Primal 0.000007076859965309268 s 0.0000071144199955597285 s 0.99
add_one / IPartOpt / cpu / Primal 0.000006986479984334438 s 0.000007482559976779157 s 0.93
add_one / DefOpt / cpu / Primal 0.00001179338001747965 s 0.000010894339966398548 s 1.08
add_one / IDefOpt / cpu / Primal 0.000007253260037032305 s 0.00000700012003107986 s 1.04
add_one / JaXPipe / cpu / Forward 0.000011327899983371026 s 0.00001131887995143188 s 1.00
add_one / Jax / cpu / Forward 0.000011578219982766312 s 0.000011139899997942848 s 1.04
add_one / HLOOpt / cpu / Forward 0.000016212640011872282 s 0.000016387120049330405 s 0.99
add_one / PartOpt / cpu / Forward 0.000015730220002296847 s 0.00001628274001632235 s 0.97
add_one / IPartOpt / cpu / Forward 0.000011479940003482624 s 0.000011611659992922796 s 0.99
add_one / DefOpt / cpu / Forward 0.000016844920019138953 s 0.00001598849999027152 s 1.05
add_one / IDefOpt / cpu / Forward 0.000011602819995459868 s 0.00001159851997726946 s 1.00
add_one / JaXPipe / cpu / PreRev 0.000013341399981072756 s 0.000013755660020251526 s 0.97
add_one / JaXPipe / cpu / PostRev 0.000012323840001045028 s 0.000013645239978359312 s 0.90
add_one / JaXPipe / cpu / BothRev 0.000014015300002938602 s 0.000016904079984669806 s 0.83
add_one / Jax / cpu / BothRev 0.000012325499974394917 s 0.000013961900031063123 s 0.88
add_one / HLOOpt / cpu / PreRev 0.000012730520029435866 s 0.00001307442003962933 s 0.97
add_one / HLOOpt / cpu / PostRev 0.000016182959989237133 s 0.000013742180017288777 s 1.18
add_one / HLOOpt / cpu / BothRev 0.000014046040014363826 s 0.000014313519959614496 s 0.98
add_one / PartOpt / cpu / PreRev 0.000012263359985809077 s 0.000013014620026297051 s 0.94
add_one / PartOpt / cpu / PostRev 0.00001268762003746815 s 0.00001419920000444108 s 0.89
add_one / PartOpt / cpu / BothRev 0.000012126239962526595 s 0.000013423580012386085 s 0.90
add_one / IPartOpt / cpu / PreRev 0.00001784132003194827 s 0.00001846378003392601 s 0.97
add_one / IPartOpt / cpu / PostRev 0.00001286049995542271 s 0.000012843100039390263 s 1.00
add_one / IPartOpt / cpu / BothRev 0.000012023599992971867 s 0.000013335700014067698 s 0.90
add_one / DefOpt / cpu / PreRev 0.00001264882001123624 s 0.0000128624799890531 s 0.98
add_one / DefOpt / cpu / PostRev 0.00001238883996848017 s 0.000013225259999671837 s 0.94
add_one / DefOpt / cpu / BothRev 0.000012719780042971252 s 0.000012803099998563994 s 0.99
add_one / IDefOpt / cpu / PreRev 0.000012817159986298063 s 0.000013157600014892524 s 0.97
add_one / IDefOpt / cpu / PostRev 0.00001289775993427611 s 0.000013262079964988516 s 0.97
add_one / IDefOpt / cpu / BothRev 0.000012793999985660776 s 0.000013631260007969104 s 0.94
add_one / JaXPipe / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / Jax / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
add_one / HLOOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / PartOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / IPartOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / DefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / IDefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / JaXPipe / cuda / Forward 0.000010496 s 0.000010048 s 1.04
add_one / Jax / cuda / Forward 0.000010015 s 0.000009888 s 1.01
add_one / HLOOpt / cuda / Forward 0.0000104 s 0.000009856 s 1.06
add_one / PartOpt / cuda / Forward 0.000010112 s 0.000009792 s 1.03
add_one / IPartOpt / cuda / Forward 0.00001024 s 0.00001024 s 1
add_one / DefOpt / cuda / Forward 0.00001024 s 0.000010017 s 1.02
add_one / IDefOpt / cuda / Forward 0.000009984 s 0.000010144 s 0.98
add_one / JaXPipe / cuda / PreRev 0.000024959 s 0.000025248 s 0.99
add_one / JaXPipe / cuda / PostRev 0.000031296 s 0.000025505 s 1.23
add_one / JaXPipe / cuda / BothRev 0.000024576 s 0.000025696 s 0.96
add_one / Jax / cuda / BothRev 0.000024896 s 0.000025376 s 0.98
add_one / HLOOpt / cuda / PreRev 0.00002544 s 0.00002544 s 1
add_one / HLOOpt / cuda / PostRev 0.00002544 s 0.0000256 s 0.99
add_one / HLOOpt / cuda / BothRev 0.00002496 s 0.000024992 s 1.00
add_one / PartOpt / cuda / PreRev 0.000025312 s 0.00002496 s 1.01
add_one / PartOpt / cuda / PostRev 0.000024032 s 0.000024832 s 0.97
add_one / PartOpt / cuda / BothRev 0.000024576 s 0.000025056 s 0.98
add_one / IPartOpt / cuda / PreRev 0.000024864 s 0.000026016 s 0.96
add_one / IPartOpt / cuda / PostRev 0.000024992 s 0.000025472000000000003 s 0.98
add_one / IPartOpt / cuda / BothRev 0.000024672 s 0.000025536 s 0.97
add_one / DefOpt / cuda / PreRev 0.000025504 s 0.000025056 s 1.02
add_one / DefOpt / cuda / PostRev 0.000024512 s 0.000025055 s 0.98
add_one / DefOpt / cuda / BothRev 0.000024928 s 0.000026496 s 0.94
add_one / IDefOpt / cuda / PreRev 0.000024576 s 0.000026624 s 0.92
add_one / IDefOpt / cuda / PostRev 0.000024928 s 0.000024512 s 1.02
add_one / IDefOpt / cuda / BothRev 0.000024704 s 0.000025953 s 0.95
add_one / JaXPipe / tpu / Primal 0.0000014326 s 0.000001435325 s 1.00
add_one / Jax / tpu / Primal 0.00000141185 s 0.0000014036249999999998 s 1.01
add_one / HLOOpt / tpu / Primal 0.000001428225 s 0.0000014238 s 1.00
add_one / PartOpt / tpu / Primal 0.000001399025 s 0.000001409925 s 0.99
add_one / IPartOpt / tpu / Primal 0.000001427825 s 0.000001430025 s 1.00
add_one / DefOpt / tpu / Primal 0.00000140975 s 0.000001401975 s 1.01
add_one / IDefOpt / tpu / Primal 0.0000014259250000000002 s 0.0000014373250000000005 s 0.99
add_one / JaXPipe / tpu / Forward 0.00000185165 s 0.000001861625 s 0.99
add_one / Jax / tpu / Forward 0.0000018482 s 0.000001850025 s 1.00
add_one / HLOOpt / tpu / Forward 0.000001860075 s 0.00000186145 s 1.00
add_one / PartOpt / tpu / Forward 0.0000018477 s 0.00000185835 s 0.99
add_one / IPartOpt / tpu / Forward 0.00000184365 s 0.000001852475 s 1.00
add_one / DefOpt / tpu / Forward 0.000001835275 s 0.0000018417 s 1.00
add_one / IDefOpt / tpu / Forward 0.00000185455 s 0.000001850325 s 1.00
add_one / JaXPipe / tpu / PreRev 0.0000022477 s 0.0000022454750000000003 s 1.00
add_one / JaXPipe / tpu / PostRev 0.00000225125 s 0.0000022446 s 1.00
add_one / JaXPipe / tpu / BothRev 0.000002232225 s 0.00000223875 s 1.00
add_one / Jax / tpu / BothRev 0.00000223535 s 0.000002231825 s 1.00
add_one / HLOOpt / tpu / PreRev 0.000002237925 s 0.00000223815 s 1.00
add_one / HLOOpt / tpu / PostRev 0.000002244875 s 0.000002236725 s 1.00
add_one / HLOOpt / tpu / BothRev 0.000002235525 s 0.000002242775 s 1.00
add_one / PartOpt / tpu / PreRev 0.00000224435 s 0.00000224535 s 1.00
add_one / PartOpt / tpu / PostRev 0.00000223215 s 0.000002238 s 1.00
add_one / PartOpt / tpu / BothRev 0.0000022334 s 0.000002242125 s 1.00
add_one / IPartOpt / tpu / PreRev 0.000002232375 s 0.0000022485500000000003 s 0.99
add_one / IPartOpt / tpu / PostRev 0.0000022392 s 0.0000022366 s 1.00
add_one / IPartOpt / tpu / BothRev 0.000002234775 s 0.0000022363 s 1.00
add_one / DefOpt / tpu / PreRev 0.0000022422000000000003 s 0.0000022422000000000003 s 1
add_one / DefOpt / tpu / PostRev 0.0000022381750000000004 s 0.000002238425 s 1.00
add_one / DefOpt / tpu / BothRev 0.00000224045 s 0.000002240325 s 1.00
add_one / IDefOpt / tpu / PreRev 0.000002231875 s 0.000002242625 s 1.00
add_one / IDefOpt / tpu / PostRev 0.0000022392 s 0.000002235875 s 1.00
add_one / IDefOpt / tpu / BothRev 0.000002235075 s 0.0000022469 s 0.99
add_one / JaXPipe / cpu / Primal 0.000016244999999999998 s 0.000007327479988816776 s 2.22
add_one / Jax / cpu / Primal 0.000016495 s 0.000007088460015438613 s 2.33
add_one / HLOOpt / cpu / Primal 0.000026244 s 0.000011369959956937236 s 2.31
add_one / PartOpt / cpu / Primal 0.000016408 s 0.0000071144199955597285 s 2.31
add_one / IPartOpt / cpu / Primal 0.00001628 s 0.000007482559976779157 s 2.18
add_one / DefOpt / cpu / Primal 0.000016326 s 0.000010894339966398548 s 1.50
add_one / IDefOpt / cpu / Primal 0.000016209 s 0.00000700012003107986 s 2.32
add_one / JaXPipe / cpu / Forward 0.00002244 s 0.00001131887995143188 s 1.98
add_one / Jax / cpu / Forward 0.000022275 s 0.000011139899997942848 s 2.00
add_one / HLOOpt / cpu / Forward 0.000022377 s 0.000016387120049330405 s 1.37
add_one / PartOpt / cpu / Forward 0.000021879 s 0.00001628274001632235 s 1.34
add_one / IPartOpt / cpu / Forward 0.000022149 s 0.000011611659992922796 s 1.91
add_one / DefOpt / cpu / Forward 0.000022726 s 0.00001598849999027152 s 1.42
add_one / IDefOpt / cpu / Forward 0.00002238 s 0.00001159851997726946 s 1.93
add_one / JaXPipe / cpu / PreRev 0.000024665 s 0.000013755660020251526 s 1.79
add_one / JaXPipe / cpu / PostRev 0.000025105 s 0.000013645239978359312 s 1.84
add_one / JaXPipe / cpu / BothRev 0.000025051 s 0.000016904079984669806 s 1.48
add_one / Jax / cpu / BothRev 0.00002415 s 0.000013961900031063123 s 1.73
add_one / HLOOpt / cpu / PreRev 0.000024696 s 0.00001307442003962933 s 1.89
add_one / HLOOpt / cpu / PostRev 0.000025235 s 0.000013742180017288777 s 1.84
add_one / HLOOpt / cpu / BothRev 0.000024457 s 0.000014313519959614496 s 1.71
add_one / PartOpt / cpu / PreRev 0.000024794 s 0.000013014620026297051 s 1.91
add_one / PartOpt / cpu / PostRev 0.000024219 s 0.00001419920000444108 s 1.71
add_one / PartOpt / cpu / BothRev 0.00002457 s 0.000013423580012386085 s 1.83
add_one / IPartOpt / cpu / PreRev 0.000024535 s 0.00001846378003392601 s 1.33
add_one / IPartOpt / cpu / PostRev 0.00002447 s 0.000012843100039390263 s 1.91
add_one / IPartOpt / cpu / BothRev 0.000024282 s 0.000013335700014067698 s 1.82
add_one / DefOpt / cpu / PreRev 0.000024941 s 0.0000128624799890531 s 1.94
add_one / DefOpt / cpu / PostRev 0.000024818 s 0.000013225259999671837 s 1.88
add_one / DefOpt / cpu / BothRev 0.00002453 s 0.000012803099998563994 s 1.92
add_one / IDefOpt / cpu / PreRev 0.000024847 s 0.000013157600014892524 s 1.89
add_one / IDefOpt / cpu / PostRev 0.00002419 s 0.000013262079964988516 s 1.82
add_one / IDefOpt / cpu / BothRev 0.000024283 s 0.000013631260007969104 s 1.78
add_one / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000007327479988816776 s 1.23
add_one / Jax / cpu / Primal 0.000008 s 0.000007088460015438613 s 1.13
add_one / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.000011369959956937236 s 0.79
add_one / PartOpt / cpu / Primal 0.000008 s 0.0000071144199955597285 s 1.12
add_one / IPartOpt / cpu / Primal 0.000008 s 0.000007482559976779157 s 1.07
add_one / DefOpt / cpu / Primal 0.000008 s 0.000010894339966398548 s 0.73
add_one / IDefOpt / cpu / Primal 0.000008 s 0.00000700012003107986 s 1.14
add_one / JaXPipe / cpu / Forward 0.000011 s 0.00001131887995143188 s 0.97
add_one / Jax / cpu / Forward 0.000011 s 0.000011139899997942848 s 0.99
add_one / HLOOpt / cpu / Forward 0.000011 s 0.000016387120049330405 s 0.67
add_one / PartOpt / cpu / Forward 0.000011 s 0.00001628274001632235 s 0.68
add_one / IPartOpt / cpu / Forward 0.000011 s 0.000011611659992922796 s 0.95
add_one / DefOpt / cpu / Forward 0.000011 s 0.00001598849999027152 s 0.69
add_one / IDefOpt / cpu / Forward 0.000012 s 0.00001159851997726946 s 1.03
add_one / JaXPipe / cpu / PreRev 0.000013 s 0.000013755660020251526 s 0.95
add_one / JaXPipe / cpu / PostRev 0.000012 s 0.000013645239978359312 s 0.88
add_one / JaXPipe / cpu / BothRev 0.000013 s 0.000016904079984669806 s 0.77
add_one / Jax / cpu / BothRev 0.000013 s 0.000013961900031063123 s 0.93
add_one / HLOOpt / cpu / PreRev 0.000013 s 0.00001307442003962933 s 0.99
add_one / HLOOpt / cpu / PostRev 0.000013 s 0.000013742180017288777 s 0.95
add_one / HLOOpt / cpu / BothRev 0.000012 s 0.000014313519959614496 s 0.84
add_one / PartOpt / cpu / PreRev 0.000014 s 0.000013014620026297051 s 1.08
add_one / PartOpt / cpu / PostRev 0.000013 s 0.00001419920000444108 s 0.92
add_one / PartOpt / cpu / BothRev 0.000013 s 0.000013423580012386085 s 0.97
add_one / IPartOpt / cpu / PreRev 0.000013 s 0.00001846378003392601 s 0.70
add_one / IPartOpt / cpu / PostRev 0.000013 s 0.000012843100039390263 s 1.01
add_one / IPartOpt / cpu / BothRev 0.000013 s 0.000013335700014067698 s 0.97
add_one / DefOpt / cpu / PreRev 0.000013 s 0.0000128624799890531 s 1.01
add_one / DefOpt / cpu / PostRev 0.000013 s 0.000013225259999671837 s 0.98
add_one / DefOpt / cpu / BothRev 0.000014 s 0.000012803099998563994 s 1.09
add_one / IDefOpt / cpu / PreRev 0.000013 s 0.000013157600014892524 s 0.99
add_one / IDefOpt / cpu / PostRev 0.000013 s 0.000013262079964988516 s 0.98
add_one / IDefOpt / cpu / BothRev 0.000014 s 0.000013631260007969104 s 1.03
add_two / JaXPipe / cpu / Primal 0.000007852720036680694 s 0.000008326599991050898 s 0.94
add_two / Jax / cpu / Primal 0.000006589419999727397 s 0.0000076485999943543 s 0.86
add_two / HLOOpt / cpu / Primal 0.000011152920005770284 s 0.00001191600003039639 s 0.94
add_two / PartOpt / cpu / Primal 0.000007606579983985284 s 0.000007528160012952867 s 1.01
add_two / IPartOpt / cpu / Primal 0.0000072603800254000814 s 0.000007817199993951362 s 0.93
add_two / DefOpt / cpu / Primal 0.000012352119974821108 s 0.000011974039989581797 s 1.03
add_two / IDefOpt / cpu / Primal 0.000007265379990712973 s 0.000007639639970875578 s 0.95
add_two / JaXPipe / cpu / Forward 0.00001190249999126536 s 0.000011662280003292836 s 1.02
add_two / Jax / cpu / Forward 0.000011534380064404104 s 0.000011367079969204496 s 1.01
add_two / HLOOpt / cpu / Forward 0.000016189259968086843 s 0.00001602384004399937 s 1.01
add_two / PartOpt / cpu / Forward 0.000016443799968328676 s 0.000016002119991753716 s 1.03
add_two / IPartOpt / cpu / Forward 0.000011224940044485263 s 0.000011803220031652016 s 0.95
add_two / DefOpt / cpu / Forward 0.00001607964001777873 s 0.00001626469996153901 s 0.99
add_two / IDefOpt / cpu / Forward 0.000011490140004752902 s 0.000011975420002272585 s 0.96
add_two / JaXPipe / cpu / PreRev 0.000016233960004683467 s 0.000015861940019021857 s 1.02
add_two / JaXPipe / cpu / PostRev 0.000015849299988985876 s 0.00001538951994916715 s 1.03
add_two / JaXPipe / cpu / BothRev 0.000015228740003294662 s 0.000015698840006734826 s 0.97
add_two / Jax / cpu / BothRev 0.000014900480000505922 s 0.000015581679999741027 s 0.96
add_two / HLOOpt / cpu / PreRev 0.00001569932001075358 s 0.00001621366001018032 s 0.97
add_two / HLOOpt / cpu / PostRev 0.000015287839969460036 s 0.000015762840021125158 s 0.97
add_two / HLOOpt / cpu / BothRev 0.00001713612005005416 s 0.000016829940022944356 s 1.02
add_two / PartOpt / cpu / PreRev 0.000015181560065684608 s 0.000015711200003352132 s 0.97
add_two / PartOpt / cpu / PostRev 0.000016071919999376407 s 0.000015595500026392984 s 1.03
add_two / PartOpt / cpu / BothRev 0.000015994799987311126 s 0.000015256899996529682 s 1.05
add_two / IPartOpt / cpu / PreRev 0.000016314180002154898 s 0.00001616128002751793 s 1.01
add_two / IPartOpt / cpu / PostRev 0.000015953819965943693 s 0.000015450080018126754 s 1.03
add_two / IPartOpt / cpu / BothRev 0.000015345800020440946 s 0.00001545552000607131 s 0.99
add_two / DefOpt / cpu / PreRev 0.000015634419969501324 s 0.000015932999967844808 s 0.98
add_two / DefOpt / cpu / PostRev 0.000016071219970399398 s 0.00001568419996146986 s 1.02
add_two / DefOpt / cpu / BothRev 0.000015148040001804474 s 0.000015379220012619044 s 0.98
add_two / IDefOpt / cpu / PreRev 0.000015346979998867028 s 0.00001573410002492892 s 0.98
add_two / IDefOpt / cpu / PostRev 0.000015825900009076575 s 0.00001601969996954722 s 0.99
add_two / IDefOpt / cpu / BothRev 0.00001567481999700249 s 0.000015559479988951354 s 1.01
add_two / JaXPipe / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
add_two / Jax / cuda / Primal 0.000001919 s 0.0000019200000000000003 s 1.00
add_two / HLOOpt / cuda / Primal 0.000001919 s 0.000001951 s 0.98
add_two / PartOpt / cuda / Primal 0.000001919 s 0.0000019200000000000003 s 1.00
add_two / IPartOpt / cuda / Primal 0.000001919 s 0.000001951 s 0.98
add_two / DefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / IDefOpt / cuda / Primal 0.000001919 s 0.000001951 s 0.98
add_two / JaXPipe / cuda / Forward 0.00000992 s 0.000010463 s 0.95
add_two / Jax / cuda / Forward 0.000010016 s 0.000010336 s 0.97
add_two / HLOOpt / cuda / Forward 0.000010016 s 0.000010176 s 0.98
add_two / PartOpt / cuda / Forward 0.000010208 s 0.000010272 s 0.99
add_two / IPartOpt / cuda / Forward 0.000009888 s 0.00001008 s 0.98
add_two / DefOpt / cuda / Forward 0.000010176 s 0.000010144 s 1.00
add_two / IDefOpt / cuda / Forward 0.000010112 s 0.00000976 s 1.04
add_two / JaXPipe / cuda / PreRev 0.000032864 s 0.000031455 s 1.04
add_two / JaXPipe / cuda / PostRev 0.000032064 s 0.000031392 s 1.02
add_two / JaXPipe / cuda / BothRev 0.00003264 s 0.000031296 s 1.04
add_two / Jax / cuda / BothRev 0.00003232 s 0.000031456 s 1.03
add_two / HLOOpt / cuda / PreRev 0.000031839 s 0.000031936 s 1.00
add_two / HLOOpt / cuda / PostRev 0.000032448 s 0.000032032 s 1.01
add_two / HLOOpt / cuda / BothRev 0.000032672 s 0.000031487 s 1.04
add_two / PartOpt / cuda / PreRev 0.000032608 s 0.000031008 s 1.05
add_two / PartOpt / cuda / PostRev 0.000032512 s 0.000032832 s 0.99
add_two / PartOpt / cuda / BothRev 0.000032383000000000005 s 0.000031392 s 1.03
add_two / IPartOpt / cuda / PreRev 0.000032256 s 0.000031616 s 1.02
add_two / IPartOpt / cuda / PostRev 0.00003184 s 0.000031072 s 1.02
add_two / IPartOpt / cuda / BothRev 0.000032639000000000004 s 0.000032032 s 1.02
add_two / DefOpt / cuda / PreRev 0.000032832 s 0.000032384 s 1.01
add_two / DefOpt / cuda / PostRev 0.000033184 s 0.000032384 s 1.02
add_two / DefOpt / cuda / BothRev 0.000033216 s 0.00003264 s 1.02
add_two / IDefOpt / cuda / PreRev 0.000033088 s 0.000032352 s 1.02
add_two / IDefOpt / cuda / PostRev 0.00003264 s 0.000032606999999999995 s 1.00
add_two / IDefOpt / cuda / BothRev 0.000032832 s 0.000032576 s 1.01
add_two / JaXPipe / tpu / Primal 0.000001428375 s 0.000001430125 s 1.00
add_two / Jax / tpu / Primal 0.000001470225 s 0.000001472575 s 1.00
add_two / HLOOpt / tpu / Primal 0.000001435675 s 0.00000143395 s 1.00
add_two / PartOpt / tpu / Primal 0.000001477925 s 0.00000147685 s 1.00
add_two / IPartOpt / tpu / Primal 0.00000143505 s 0.000001430425 s 1.00
add_two / DefOpt / tpu / Primal 0.000001478575 s 0.000001474075 s 1.00
add_two / IDefOpt / tpu / Primal 0.000001434925 s 0.00000143795 s 1.00
add_two / JaXPipe / tpu / Forward 0.000001827875 s 0.00000182555 s 1.00
add_two / Jax / tpu / Forward 0.000001834575 s 0.000001829075 s 1.00
add_two / HLOOpt / tpu / Forward 0.000001825025 s 0.000001825125 s 1.00
add_two / PartOpt / tpu / Forward 0.000001836025 s 0.00000183185 s 1.00
add_two / IPartOpt / tpu / Forward 0.000001829625 s 0.00000182465 s 1.00
add_two / DefOpt / tpu / Forward 0.0000018252 s 0.0000018316 s 1.00
add_two / IDefOpt / tpu / Forward 0.00000183135 s 0.000001837675 s 1.00
add_two / JaXPipe / tpu / PreRev 0.000002858675 s 0.0000028472 s 1.00
add_two / JaXPipe / tpu / PostRev 0.0000027821 s 0.000002751275 s 1.01
add_two / JaXPipe / tpu / BothRev 0.0000028423250000000004 s 0.000002846025 s 1.00
add_two / Jax / tpu / BothRev 0.0000027801500000000004 s 0.0000027604250000000003 s 1.01
add_two / HLOOpt / tpu / PreRev 0.000002853625 s 0.000002848225 s 1.00
add_two / HLOOpt / tpu / PostRev 0.0000027637 s 0.0000027621499999999995 s 1.00
add_two / HLOOpt / tpu / BothRev 0.00000284345 s 0.00000284105 s 1.00
add_two / PartOpt / tpu / PreRev 0.0000027732250000000005 s 0.00000274965 s 1.01
add_two / PartOpt / tpu / PostRev 0.000002855325 s 0.0000028436 s 1.00
add_two / PartOpt / tpu / BothRev 0.00000276455 s 0.000002745875 s 1.01
add_two / IPartOpt / tpu / PreRev 0.00000284535 s 0.00000284685 s 1.00
add_two / IPartOpt / tpu / PostRev 0.0000027525500000000004 s 0.000002767475 s 0.99
add_two / IPartOpt / tpu / BothRev 0.000002851725 s 0.000002841675 s 1.00
add_two / DefOpt / tpu / PreRev 0.0000027628 s 0.000002756725 s 1.00
add_two / DefOpt / tpu / PostRev 0.0000028639750000000004 s 0.000002853375 s 1.00
add_two / DefOpt / tpu / BothRev 0.000002765725 s 0.000002743775 s 1.01
add_two / IDefOpt / tpu / PreRev 0.000002850475 s 0.000002857125 s 1.00
add_two / IDefOpt / tpu / PostRev 0.000002762825 s 0.000002761625 s 1.00
add_two / IDefOpt / tpu / BothRev 0.0000028603250000000004 s 0.000002848825 s 1.00
add_two / JaXPipe / cpu / Primal 0.000016179 s 0.000008326599991050898 s 1.94
add_two / Jax / cpu / Primal 0.000016587999999999998 s 0.0000076485999943543 s 2.17
add_two / HLOOpt / cpu / Primal 0.000016643000000000003 s 0.00001191600003039639 s 1.40
add_two / PartOpt / cpu / Primal 0.000016563 s 0.000007528160012952867 s 2.20
add_two / IPartOpt / cpu / Primal 0.000016528999999999997 s 0.000007817199993951362 s 2.11
add_two / DefOpt / cpu / Primal 0.000016927999999999998 s 0.000011974039989581797 s 1.41
add_two / IDefOpt / cpu / Primal 0.000016645 s 0.000007639639970875578 s 2.18
add_two / JaXPipe / cpu / Forward 0.000022781 s 0.000011662280003292836 s 1.95
add_two / Jax / cpu / Forward 0.000022365 s 0.000011367079969204496 s 1.97
add_two / HLOOpt / cpu / Forward 0.000022618 s 0.00001602384004399937 s 1.41
add_two / PartOpt / cpu / Forward 0.000022794 s 0.000016002119991753716 s 1.42
add_two / IPartOpt / cpu / Forward 0.00002224 s 0.000011803220031652016 s 1.88
add_two / DefOpt / cpu / Forward 0.000022578 s 0.00001626469996153901 s 1.39
add_two / IDefOpt / cpu / Forward 0.000022747 s 0.000011975420002272585 s 1.90
add_two / JaXPipe / cpu / PreRev 0.000029544 s 0.000015861940019021857 s 1.86
add_two / JaXPipe / cpu / PostRev 0.00002884 s 0.00001538951994916715 s 1.87
add_two / JaXPipe / cpu / BothRev 0.000028635 s 0.000015698840006734826 s 1.82
add_two / Jax / cpu / BothRev 0.000028567 s 0.000015581679999741027 s 1.83
add_two / HLOOpt / cpu / PreRev 0.000028663 s 0.00001621366001018032 s 1.77
add_two / HLOOpt / cpu / PostRev 0.000030394 s 0.000015762840021125158 s 1.93
add_two / HLOOpt / cpu / BothRev 0.000029633 s 0.000016829940022944356 s 1.76
add_two / PartOpt / cpu / PreRev 0.000029079 s 0.000015711200003352132 s 1.85
add_two / PartOpt / cpu / PostRev 0.000029176 s 0.000015595500026392984 s 1.87
add_two / PartOpt / cpu / BothRev 0.000030436 s 0.000015256899996529682 s 1.99
add_two / IPartOpt / cpu / PreRev 0.000028502 s 0.00001616128002751793 s 1.76
add_two / IPartOpt / cpu / PostRev 0.000029581 s 0.000015450080018126754 s 1.91
add_two / IPartOpt / cpu / BothRev 0.000029555 s 0.00001545552000607131 s 1.91
add_two / DefOpt / cpu / PreRev 0.000029571 s 0.000015932999967844808 s 1.86
add_two / DefOpt / cpu / PostRev 0.000029672 s 0.00001568419996146986 s 1.89
add_two / DefOpt / cpu / BothRev 0.000028876 s 0.000015379220012619044 s 1.88
add_two / IDefOpt / cpu / PreRev 0.000028851 s 0.00001573410002492892 s 1.83
add_two / IDefOpt / cpu / PostRev 0.000029508 s 0.00001601969996954722 s 1.84
add_two / IDefOpt / cpu / BothRev 0.000029287 s 0.000015559479988951354 s 1.88
add_two / JaXPipe / cpu / Primal 0.000008 s 0.000008326599991050898 s 0.96
add_two / Jax / cpu / Primal 0.000008999999999999999 s 0.0000076485999943543 s 1.18
add_two / HLOOpt / cpu / Primal 0.000008 s 0.00001191600003039639 s 0.67
add_two / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000007528160012952867 s 1.20
add_two / IPartOpt / cpu / Primal 0.000008 s 0.000007817199993951362 s 1.02
add_two / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000011974039989581797 s 0.75
add_two / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000007639639970875578 s 1.18
add_two / JaXPipe / cpu / Forward 0.000012 s 0.000011662280003292836 s 1.03
add_two / Jax / cpu / Forward 0.000012 s 0.000011367079969204496 s 1.06
add_two / HLOOpt / cpu / Forward 0.000011 s 0.00001602384004399937 s 0.69
add_two / PartOpt / cpu / Forward 0.000012 s 0.000016002119991753716 s 0.75
add_two / IPartOpt / cpu / Forward 0.000012 s 0.000011803220031652016 s 1.02
add_two / DefOpt / cpu / Forward 0.000012 s 0.00001626469996153901 s 0.74
add_two / IDefOpt / cpu / Forward 0.000012 s 0.000011975420002272585 s 1.00
add_two / JaXPipe / cpu / PreRev 0.000015 s 0.000015861940019021857 s 0.95
add_two / JaXPipe / cpu / PostRev 0.000016 s 0.00001538951994916715 s 1.04
add_two / JaXPipe / cpu / BothRev 0.000016 s 0.000015698840006734826 s 1.02
add_two / Jax / cpu / BothRev 0.000016 s 0.000015581679999741027 s 1.03
add_two / HLOOpt / cpu / PreRev 0.000015 s 0.00001621366001018032 s 0.93
add_two / HLOOpt / cpu / PostRev 0.000015 s 0.000015762840021125158 s 0.95
add_two / HLOOpt / cpu / BothRev 0.000015 s 0.000016829940022944356 s 0.89
add_two / PartOpt / cpu / PreRev 0.000016 s 0.000015711200003352132 s 1.02
add_two / PartOpt / cpu / PostRev 0.000016 s 0.000015595500026392984 s 1.03
add_two / PartOpt / cpu / BothRev 0.000016 s 0.000015256899996529682 s 1.05
add_two / IPartOpt / cpu / PreRev 0.000015 s 0.00001616128002751793 s 0.93
add_two / IPartOpt / cpu / PostRev 0.000016 s 0.000015450080018126754 s 1.04
add_two / IPartOpt / cpu / BothRev 0.000015 s 0.00001545552000607131 s 0.97
add_two / DefOpt / cpu / PreRev 0.000016 s 0.000015932999967844808 s 1.00
add_two / DefOpt / cpu / PostRev 0.000016 s 0.00001568419996146986 s 1.02
add_two / DefOpt / cpu / BothRev 0.000015 s 0.000015379220012619044 s 0.98
add_two / IDefOpt / cpu / PreRev 0.000015 s 0.00001573410002492892 s 0.95
add_two / IDefOpt / cpu / PostRev 0.000015 s 0.00001601969996954722 s 0.94
add_two / IDefOpt / cpu / BothRev 0.000016 s 0.000015559479988951354 s 1.03
cache / JaXPipe / cpu / Primal 0.00000692341996000323 s 0.000007605419968967908 s 0.91
cache / Jax / cpu / Primal 0.0000070851800410309805 s 0.000008150019993991009 s 0.87
cache / HLOOpt / cpu / Primal 0.000006749860003765206 s 0.0000069164399974397385 s 0.98
cache / PartOpt / cpu / Primal 0.00000769159996707458 s 0.000007421300024361699 s 1.04
cache / IPartOpt / cpu / Primal 0.000007338839977819589 s 0.000007143139991967474 s 1.03
cache / DefOpt / cpu / Primal 0.000006848239991086302 s 0.000006948359987291041 s 0.99
cache / IDefOpt / cpu / Primal 0.000006559359981110902 s 0.000007039800020720577 s 0.93
cache / JaXPipe / cpu / Forward 0.000016687740026100072 s 0.000014946399978725822 s 1.12
cache / Jax / cpu / Forward 0.000016540800033908453 s 0.000015384240014100216 s 1.08
cache / HLOOpt / cpu / Forward 0.00002181561999350379 s 0.000020600239995474112 s 1.06
cache / PartOpt / cpu / Forward 0.000021887399980187184 s 0.0000206277599590976 s 1.06
cache / IPartOpt / cpu / Forward 0.00001721180004096823 s 0.000015604360041834296 s 1.10
cache / DefOpt / cpu / Forward 0.00002132892003828601 s 0.00002073108001241053 s 1.03
cache / IDefOpt / cpu / Forward 0.000017391980009051622 s 0.000014873839982101344 s 1.17
cache / JaXPipe / cpu / PreRev 0.00001850552004725614 s 0.000016908659990804153 s 1.09
cache / JaXPipe / cpu / PostRev 0.00002364177998060768 s 0.000022480599991467898 s 1.05
cache / JaXPipe / cpu / BothRev 0.000022616100022787577 s 0.00001723479996144306 s 1.31
cache / Jax / cpu / BothRev 0.000022891379976499597 s 0.00002226266003162891 s 1.03
cache / HLOOpt / cpu / PreRev 0.00001849208002568048 s 0.000016768200030128355 s 1.10
cache / HLOOpt / cpu / PostRev 0.000022042259997760995 s 0.000020520200005194057 s 1.07
cache / HLOOpt / cpu / BothRev 0.00002081913999063545 s 0.000018931259992314152 s 1.10
cache / PartOpt / cpu / PreRev 0.000017064739986381027 s 0.00001635605997762468 s 1.04
cache / PartOpt / cpu / PostRev 0.00002245478000077128 s 0.00002118551997227769 s 1.06
cache / PartOpt / cpu / BothRev 0.000017673999964245012 s 0.000016416320049756906 s 1.08
cache / IPartOpt / cpu / PreRev 0.00002295090000188793 s 0.000021871360004297455 s 1.05
cache / IPartOpt / cpu / PostRev 0.000022771740004827736 s 0.00002011587999732001 s 1.13
cache / IPartOpt / cpu / BothRev 0.00001684472000306414 s 0.0000167720999706944 s 1.00
cache / DefOpt / cpu / PreRev 0.000017295919969910757 s 0.00001704424003037275 s 1.01
cache / DefOpt / cpu / PostRev 0.000017357919978167048 s 0.00001719764001791191 s 1.01
cache / DefOpt / cpu / BothRev 0.00001803462000680156 s 0.000020657419963754364 s 0.87
cache / IDefOpt / cpu / PreRev 0.000017849220012067234 s 0.000017212519987879206 s 1.04
cache / IDefOpt / cpu / PostRev 0.00001740026003062667 s 0.000016702299999451496 s 1.04
cache / IDefOpt / cpu / BothRev 0.00001811636000638828 s 0.000017091460003939575 s 1.06
cache / JaXPipe / cuda / Primal 0.000002272 s 0.000002304 s 0.99
cache / Jax / cuda / Primal 0.00000224 s 0.000002272 s 0.99
cache / HLOOpt / cuda / Primal 0.00000224 s 0.00000224 s 1
cache / PartOpt / cuda / Primal 0.00000224 s 0.00000224 s 1
cache / IPartOpt / cuda / Primal 0.000002208 s 0.00000224 s 0.99
cache / DefOpt / cuda / Primal 0.000002272 s 0.00000224 s 1.01
cache / IDefOpt / cuda / Primal 0.000002271 s 0.000002304 s 0.99
cache / JaXPipe / cuda / Forward 0.0000023050000000000004 s 0.000002335 s 0.99
cache / Jax / cuda / Forward 0.000002272 s 0.000002272 s 1
cache / HLOOpt / cuda / Forward 0.000002304 s 0.000002335 s 0.99
cache / PartOpt / cuda / Forward 0.000002303 s 0.000002335 s 0.99
cache / IPartOpt / cuda / Forward 0.00000224 s 0.000002303 s 0.97
cache / DefOpt / cuda / Forward 0.000002207 s 0.000002272 s 0.97
cache / IDefOpt / cuda / Forward 0.000002271 s 0.000002272 s 1.00
cache / JaXPipe / cuda / PreRev 0.000013408 s 0.00001328 s 1.01
cache / JaXPipe / cuda / PostRev 0.00001168 s 0.000011904 s 0.98
cache / JaXPipe / cuda / BothRev 0.000013408 s 0.00001328 s 1.01
cache / Jax / cuda / BothRev 0.000011648 s 0.000012096 s 0.96
cache / HLOOpt / cuda / PreRev 0.000013408 s 0.00001328 s 1.01
cache / HLOOpt / cuda / PostRev 0.000013376 s 0.000013248 s 1.01
cache / HLOOpt / cuda / BothRev 0.000013408 s 0.000013311 s 1.01
cache / PartOpt / cuda / PreRev 0.000013439 s 0.00001328 s 1.01
cache / PartOpt / cuda / PostRev 0.000011776 s 0.00001472 s 0.80
cache / PartOpt / cuda / BothRev 0.00001344 s 0.00001328 s 1.01
cache / IPartOpt / cuda / PreRev 0.000013439 s 0.00001328 s 1.01
cache / IPartOpt / cuda / PostRev 0.000011616 s 0.000011776 s 0.99
cache / IPartOpt / cuda / BothRev 0.000013407 s 0.00001328 s 1.01
cache / DefOpt / cuda / PreRev 0.00001344 s 0.000013247 s 1.01
cache / DefOpt / cuda / PostRev 0.000013344 s 0.000013247 s 1.01
cache / DefOpt / cuda / BothRev 0.000013439 s 0.00001328 s 1.01
cache / IDefOpt / cuda / PreRev 0.000013375 s 0.00001328 s 1.01
cache / IDefOpt / cuda / PostRev 0.000013376 s 0.000013248 s 1.01
cache / IDefOpt / cuda / BothRev 0.000013376 s 0.000013311 s 1.00
cache / JaXPipe / tpu / Primal 0.000002454625 s 0.000002476925 s 0.99
cache / Jax / tpu / Primal 0.00000245305 s 0.00000246545 s 0.99
cache / HLOOpt / tpu / Primal 0.00000245665 s 0.000002457825 s 1.00
cache / PartOpt / tpu / Primal 0.00000245845 s 0.00000245365 s 1.00
cache / IPartOpt / tpu / Primal 0.00000246365 s 0.00000247005 s 1.00
cache / DefOpt / tpu / Primal 0.0000024712750000000003 s 0.0000024523 s 1.01
cache / IDefOpt / tpu / Primal 0.0000024573 s 0.0000024655 s 1.00
cache / JaXPipe / tpu / Forward 0.000003549875 s 0.00000354875 s 1.00
cache / Jax / tpu / Forward 0.000003564275 s 0.00000353975 s 1.01
cache / HLOOpt / tpu / Forward 0.000003588075 s 0.00000355315 s 1.01
cache / PartOpt / tpu / Forward 0.0000035490500000000004 s 0.000003517825 s 1.01
cache / IPartOpt / tpu / Forward 0.000003579925 s 0.00000355065 s 1.01
cache / DefOpt / tpu / Forward 0.00000354035 s 0.0000035306 s 1.00
cache / IDefOpt / tpu / Forward 0.000003565275 s 0.000003535825 s 1.01
cache / JaXPipe / tpu / PreRev 0.00000392975 s 0.00000396645 s 0.99
cache / JaXPipe / tpu / PostRev 0.000004959324999999999 s 0.000005007249999999999 s 0.99
cache / JaXPipe / tpu / BothRev 0.000003933725 s 0.000003960725 s 0.99
cache / Jax / tpu / BothRev 0.000004981975000000001 s 0.0000050058250000000005 s 1.00
cache / HLOOpt / tpu / PreRev 0.00000393925 s 0.000003983075 s 0.99
cache / HLOOpt / tpu / PostRev 0.00000411015 s 0.00000414035 s 0.99
cache / HLOOpt / tpu / BothRev 0.000003928075 s 0.000003968725 s 0.99
cache / PartOpt / tpu / PreRev 0.00000411315 s 0.000004152 s 0.99
cache / PartOpt / tpu / PostRev 0.000004964675 s 0.0000050189 s 0.99
cache / PartOpt / tpu / BothRev 0.000004109350000000001 s 0.00000415485 s 0.99
cache / IPartOpt / tpu / PreRev 0.00000394035 s 0.0000039595 s 1.00
cache / IPartOpt / tpu / PostRev 0.000004970550000000001 s 0.00000500095 s 0.99
cache / IPartOpt / tpu / BothRev 0.000003933175 s 0.000003965625 s 0.99
cache / DefOpt / tpu / PreRev 0.0000041445 s 0.00000415915 s 1.00
cache / DefOpt / tpu / PostRev 0.000003940725 s 0.0000039627 s 0.99
cache / DefOpt / tpu / BothRev 0.000004121125 s 0.00000414625 s 0.99
cache / IDefOpt / tpu / PreRev 0.0000039437 s 0.000003966575 s 0.99
cache / IDefOpt / tpu / PostRev 0.000004132925 s 0.000004139999999999999 s 1.00
cache / IDefOpt / tpu / BothRev 0.000003938875 s 0.000003985 s 0.99
cache / JaXPipe / cpu / Primal 0.000018933 s 0.000007605419968967908 s 2.49
cache / Jax / cpu / Primal 0.00002541 s 0.000008150019993991009 s 3.12
cache / HLOOpt / cpu / Primal 0.000018971 s 0.0000069164399974397385 s 2.74
cache / PartOpt / cpu / Primal 0.000018258 s 0.000007421300024361699 s 2.46
cache / IPartOpt / cpu / Primal 0.000018603 s 0.000007143139991967474 s 2.60
cache / DefOpt / cpu / Primal 0.000018906 s 0.000006948359987291041 s 2.72
cache / IDefOpt / cpu / Primal 0.000018651 s 0.000007039800020720577 s 2.65
cache / JaXPipe / cpu / Forward 0.00002153 s 0.000014946399978725822 s 1.44
cache / Jax / cpu / Forward 0.000021489 s 0.000015384240014100216 s 1.40
cache / HLOOpt / cpu / Forward 0.000021524000000000003 s 0.000020600239995474112 s 1.04
cache / PartOpt / cpu / Forward 0.000022089 s 0.0000206277599590976 s 1.07
cache / IPartOpt / cpu / Forward 0.000025344 s 0.000015604360041834296 s 1.62
cache / DefOpt / cpu / Forward 0.000030849 s 0.00002073108001241053 s 1.49
cache / IDefOpt / cpu / Forward 0.000033888 s 0.000014873839982101344 s 2.28
cache / JaXPipe / cpu / PreRev 0.000032915 s 0.000016908659990804153 s 1.95
cache / JaXPipe / cpu / PostRev 0.000039224 s 0.000022480599991467898 s 1.74
cache / JaXPipe / cpu / BothRev 0.000031529 s 0.00001723479996144306 s 1.83
cache / Jax / cpu / BothRev 0.000043712 s 0.00002226266003162891 s 1.96
cache / HLOOpt / cpu / PreRev 0.000032916 s 0.000016768200030128355 s 1.96
cache / HLOOpt / cpu / PostRev 0.00002403 s 0.000020520200005194057 s 1.17
cache / HLOOpt / cpu / BothRev 0.000021705 s 0.000018931259992314152 s 1.15
cache / PartOpt / cpu / PreRev 0.000021704 s 0.00001635605997762468 s 1.33
cache / PartOpt / cpu / PostRev 0.000031679 s 0.00002118551997227769 s 1.50
cache / PartOpt / cpu / BothRev 0.000022615 s 0.000016416320049756906 s 1.38
cache / IPartOpt / cpu / PreRev 0.000021895 s 0.000021871360004297455 s 1.00
cache / IPartOpt / cpu / PostRev 0.000024728 s 0.00002011587999732001 s 1.23
cache / IPartOpt / cpu / BothRev 0.000022363000000000003 s 0.0000167720999706944 s 1.33
cache / DefOpt / cpu / PreRev 0.000021802 s 0.00001704424003037275 s 1.28
cache / DefOpt / cpu / PostRev 0.000035375999999999995 s 0.00001719764001791191 s 2.06
cache / DefOpt / cpu / BothRev 0.000040293 s 0.000020657419963754364 s 1.95
cache / IDefOpt / cpu / PreRev 0.000030074 s 0.000017212519987879206 s 1.75
cache / IDefOpt / cpu / PostRev 0.000022089 s 0.000016702299999451496 s 1.32
cache / IDefOpt / cpu / BothRev 0.000022222 s 0.000017091460003939575 s 1.30
cache / JaXPipe / cpu / Primal 0.000008 s 0.000007605419968967908 s 1.05
cache / Jax / cpu / Primal 0.000008 s 0.000008150019993991009 s 0.98
cache / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.0000069164399974397385 s 1.30
cache / PartOpt / cpu / Primal 0.000008 s 0.000007421300024361699 s 1.08
cache / IPartOpt / cpu / Primal 0.000008 s 0.000007143139991967474 s 1.12
cache / DefOpt / cpu / Primal 0.000008 s 0.000006948359987291041 s 1.15
cache / IDefOpt / cpu / Primal 0.000008 s 0.000007039800020720577 s 1.14
cache / JaXPipe / cpu / Forward 0.000011 s 0.000014946399978725822 s 0.74
cache / Jax / cpu / Forward 0.00001 s 0.000015384240014100216 s 0.65
cache / HLOOpt / cpu / Forward 0.00001 s 0.000020600239995474112 s 0.49
cache / PartOpt / cpu / Forward 0.00001 s 0.0000206277599590976 s 0.48
cache / IPartOpt / cpu / Forward 0.00001 s 0.000015604360041834296 s 0.64
cache / DefOpt / cpu / Forward 0.00001 s 0.00002073108001241053 s 0.48
cache / IDefOpt / cpu / Forward 0.00001 s 0.000014873839982101344 s 0.67
cache / JaXPipe / cpu / PreRev 0.00001 s 0.000016908659990804153 s 0.59
cache / JaXPipe / cpu / PostRev 0.000011 s 0.000022480599991467898 s 0.49
cache / JaXPipe / cpu / BothRev 0.00001 s 0.00001723479996144306 s 0.58
cache / Jax / cpu / BothRev 0.000011 s 0.00002226266003162891 s 0.49
cache / HLOOpt / cpu / PreRev 0.000011 s 0.000016768200030128355 s 0.66
cache / HLOOpt / cpu / PostRev 0.00001 s 0.000020520200005194057 s 0.49
cache / HLOOpt / cpu / BothRev 0.000011 s 0.000018931259992314152 s 0.58
cache / PartOpt / cpu / PreRev 0.00001 s 0.00001635605997762468 s 0.61
cache / PartOpt / cpu / PostRev 0.000011 s 0.00002118551997227769 s 0.52
cache / PartOpt / cpu / BothRev 0.000011 s 0.000016416320049756906 s 0.67
cache / IPartOpt / cpu / PreRev 0.00001 s 0.000021871360004297455 s 0.46
cache / IPartOpt / cpu / PostRev 0.000011 s 0.00002011587999732001 s 0.55
cache / IPartOpt / cpu / BothRev 0.000011 s 0.0000167720999706944 s 0.66
cache / DefOpt / cpu / PreRev 0.000011 s 0.00001704424003037275 s 0.65
cache / DefOpt / cpu / PostRev 0.00001 s 0.00001719764001791191 s 0.58
cache / DefOpt / cpu / BothRev 0.00001 s 0.000020657419963754364 s 0.48
cache / IDefOpt / cpu / PreRev 0.000011 s 0.000017212519987879206 s 0.64
cache / IDefOpt / cpu / PostRev 0.00001 s 0.000016702299999451496 s 0.60
cache / IDefOpt / cpu / BothRev 0.00001 s 0.000017091460003939575 s 0.59
Concat / JaXPipe / cpu / Primal 0.000008478199979435886 s 0.000008160360048350412 s 1.04
Concat / Jax / cpu / Primal 0.000007790600029693451 s 0.000007911260008768294 s 0.98
Concat / HLOOpt / cpu / Primal 0.000011056660014219233 s 0.00001136832001066068 s 0.97
Concat / PartOpt / cpu / Primal 0.000007185379963630112 s 0.000007840580010451959 s 0.92
Concat / IPartOpt / cpu / Primal 0.000007124620024114847 s 0.000007454679998772917 s 0.96
Concat / DefOpt / cpu / Primal 0.000010920680015260588 s 0.000009908619949783316 s 1.10
Concat / IDefOpt / cpu / Primal 0.000007460100068783503 s 0.000007624559993928415 s 0.98
Concat / JaXPipe / cpu / Forward 0.000011037499989470234 s 0.000011352860019542277 s 0.97
Concat / Jax / cpu / Forward 0.000010927620014626882 s 0.000011229959991396754 s 0.97
Concat / HLOOpt / cpu / Forward 0.000015735500019218305 s 0.0000151267399542121 s 1.04
Concat / PartOpt / cpu / Forward 0.00001624696001272241 s 0.000015686679971622653 s 1.04
Concat / IPartOpt / cpu / Forward 0.000010885559977396042 s 0.000011019880039384589 s 0.99
Concat / DefOpt / cpu / Forward 0.000015319099975386053 s 0.00001622339999812539 s 0.94
Concat / IDefOpt / cpu / Forward 0.000011319820032440475 s 0.000010681559979275337 s 1.06
Concat / JaXPipe / cpu / PreRev 0.000013405940026132157 s 0.000012957980043211136 s 1.03
Concat / JaXPipe / cpu / PostRev 0.000012147800007369369 s 0.000012530779968074057 s 0.97
Concat / JaXPipe / cpu / BothRev 0.000016764399961175515 s 0.000012426180019247113 s 1.35
Concat / Jax / cpu / BothRev 0.000012889119980172835 s 0.000013003280009797893 s 0.99
Concat / HLOOpt / cpu / PreRev 0.000012573820031320793 s 0.000013130240004102234 s 0.96
Concat / HLOOpt / cpu / PostRev 0.00001250731998879928 s 0.000016995599989968467 s 0.74
Concat / HLOOpt / cpu / BothRev 0.00001464240000132122 s 0.00001541774000543228 s 0.95
Concat / PartOpt / cpu / PreRev 0.00001228750001246226 s 0.000012957440012542066 s 0.95
Concat / PartOpt / cpu / PostRev 0.000012885760006611236 s 0.00001301497998611012 s 0.99
Concat / PartOpt / cpu / BothRev 0.000016212379987337043 s 0.000011962740018134356 s 1.36
Concat / IPartOpt / cpu / PreRev 0.00001302390000091691 s 0.000013326339985724187 s 0.98
Concat / IPartOpt / cpu / PostRev 0.000012785020016963244 s 0.000012175160009064712 s 1.05
Concat / IPartOpt / cpu / BothRev 0.00001297214002988767 s 0.0000123994800287619 s 1.05
Concat / DefOpt / cpu / PreRev 0.000012435380040187736 s 0.000012641379962587962 s 0.98
Concat / DefOpt / cpu / PostRev 0.000013222720017438404 s 0.000012612659947990325 s 1.05
Concat / DefOpt / cpu / BothRev 0.000012623219954548405 s 0.000012942380035383394 s 0.98
Concat / IDefOpt / cpu / PreRev 0.0000129033199755213 s 0.00001297918001000653 s 0.99
Concat / IDefOpt / cpu / PostRev 0.000013020059986956766 s 0.000013093399993522325 s 0.99
Concat / IDefOpt / cpu / BothRev 0.000012741880000248785 s 0.000012803519994122323 s 1.00
Concat / JaXPipe / cuda / Primal 0.000001919 s 0.000001951 s 0.98
Concat / Jax / cuda / Primal 0.000001919 s 0.000001951 s 0.98
Concat / HLOOpt / cuda / Primal 0.000001919 s 0.000001951 s 0.98
Concat / PartOpt / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
Concat / IPartOpt / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
Concat / DefOpt / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
Concat / IDefOpt / cuda / Primal 0.0000019200000000000003 s 0.000001951 s 0.98
Concat / JaXPipe / cuda / Forward 0.000010145 s 0.00000992 s 1.02
Concat / Jax / cuda / Forward 0.000010368 s 0.000015360000000000002 s 0.67
Concat / HLOOpt / cuda / Forward 0.00001008 s 0.000009953 s 1.01
Concat / PartOpt / cuda / Forward 0.000010176 s 0.00000992 s 1.03
Concat / IPartOpt / cuda / Forward 0.000010048 s 0.00001008 s 1.00
Concat / DefOpt / cuda / Forward 0.000009888 s 0.000010047 s 0.98
Concat / IDefOpt / cuda / Forward 0.000009792 s 0.000010016 s 0.98
Concat / JaXPipe / cuda / PreRev 0.000016768000000000003 s 0.000016288 s 1.03
Concat / JaXPipe / cuda / PostRev 0.000016448000000000002 s 0.000017088 s 0.96
Concat / JaXPipe / cuda / BothRev 0.00001648 s 0.000016608 s 0.99
Concat / Jax / cuda / BothRev 0.000016832 s 0.000016672 s 1.01
Concat / HLOOpt / cuda / PreRev 0.000016703 s 0.000016608 s 1.01
Concat / HLOOpt / cuda / PostRev 0.000016736 s 0.000016544 s 1.01
Concat / HLOOpt / cuda / BothRev 0.000016352 s 0.000016896000000000002 s 0.97
Concat / PartOpt / cuda / PreRev 0.00001712 s 0.000016576000000000002 s 1.03
Concat / PartOpt / cuda / PostRev 0.000016704 s 0.000016383000000000002 s 1.02
Concat / PartOpt / cuda / BothRev 0.000015968 s 0.000016832 s 0.95
Concat / IPartOpt / cuda / PreRev 0.0000168 s 0.000016864 s 1.00
Concat / IPartOpt / cuda / PostRev 0.00001728 s 0.000016768000000000003 s 1.03
Concat / IPartOpt / cuda / BothRev 0.000016768999999999998 s 0.000016673 s 1.01
Concat / DefOpt / cuda / PreRev 0.000016927999999999998 s 0.00001712 s 0.99
Concat / DefOpt / cuda / PostRev 0.000016768999999999998 s 0.000016512 s 1.02
Concat / DefOpt / cuda / BothRev 0.000016768000000000003 s 0.000016672 s 1.01
Concat / IDefOpt / cuda / PreRev 0.000016927999999999998 s 0.000016735 s 1.01
Concat / IDefOpt / cuda / PostRev 0.000017536 s 0.000016704 s 1.05
Concat / IDefOpt / cuda / BothRev 0.0000168 s 0.000016609 s 1.01
Concat / JaXPipe / tpu / Primal 0.0000015448 s 0.00000152705 s 1.01
Concat / Jax / tpu / Primal 0.000001528675 s 0.0000015348 s 1.00
Concat / HLOOpt / tpu / Primal 0.0000015353 s 0.0000015368 s 1.00
Concat / PartOpt / tpu / Primal 0.0000015388 s 0.0000015253749999999998 s 1.01
Concat / IPartOpt / tpu / Primal 0.000001529575 s 0.00000153285 s 1.00
Concat / DefOpt / tpu / Primal 0.00000153715 s 0.0000015403 s 1.00
Concat / IDefOpt / tpu / Primal 0.0000015339500000000002 s 0.0000015354499999999998 s 1.00
Concat / JaXPipe / tpu / Forward 0.0000015749 s 0.0000015934 s 0.99
Concat / Jax / tpu / Forward 0.0000015528249999999998 s 0.00000156025 s 1.00
Concat / HLOOpt / tpu / Forward 0.0000015706 s 0.0000015912 s 0.99
Concat / PartOpt / tpu / Forward 0.0000015479 s 0.0000015598500000000002 s 0.99
Concat / IPartOpt / tpu / Forward 0.000001572125 s 0.00000157875 s 1.00
Concat / DefOpt / tpu / Forward 0.0000015518499999999998 s 0.0000015705749999999998 s 0.99
Concat / IDefOpt / tpu / Forward 0.000001568875 s 0.0000015839 s 0.99
Concat / JaXPipe / tpu / PreRev 0.0000020067 s 0.0000020050000000000003 s 1.00
Concat / JaXPipe / tpu / PostRev 0.000002087225 s 0.000002061325 s 1.01
Concat / JaXPipe / tpu / BothRev 0.0000020171 s 0.000002009225 s 1.00
Concat / Jax / tpu / BothRev 0.000002069675 s 0.000002066125 s 1.00
Concat / HLOOpt / tpu / PreRev 0.0000020091000000000004 s 0.000002008225 s 1.00
Concat / HLOOpt / tpu / PostRev 0.00000208075 s 0.000002081375 s 1.00
Concat / HLOOpt / tpu / BothRev 0.000002012125 s 0.00000202095 s 1.00
Concat / PartOpt / tpu / PreRev 0.0000020653250000000004 s 0.000002062525 s 1.00
Concat / PartOpt / tpu / PostRev 0.00000201125 s 0.0000020108500000000004 s 1.00
Concat / PartOpt / tpu / BothRev 0.0000020712 s 0.0000020662 s 1.00
Concat / IPartOpt / tpu / PreRev 0.000002005275 s 0.000002021875 s 0.99
Concat / IPartOpt / tpu / PostRev 0.0000020776 s 0.000002069275 s 1.00
Concat / IPartOpt / tpu / BothRev 0.00000200335 s 0.000002003525 s 1.00
Concat / DefOpt / tpu / PreRev 0.000002080125 s 0.0000020617 s 1.01
Concat / DefOpt / tpu / PostRev 0.0000020094500000000003 s 0.000002004525 s 1.00
Concat / DefOpt / tpu / BothRev 0.000002079675 s 0.00000205825 s 1.01
Concat / IDefOpt / tpu / PreRev 0.000002000725 s 0.000002016325 s 0.99
Concat / IDefOpt / tpu / PostRev 0.000002071875 s 0.00000206785 s 1.00
Concat / IDefOpt / tpu / BothRev 0.0000020007 s 0.000002006025 s 1.00
Concat / JaXPipe / cpu / Primal 0.000015808 s 0.000008160360048350412 s 1.94
Concat / Jax / cpu / Primal 0.000016425999999999998 s 0.000007911260008768294 s 2.08
Concat / HLOOpt / cpu / Primal 0.000016182 s 0.00001136832001066068 s 1.42
Concat / PartOpt / cpu / Primal 0.000016046999999999997 s 0.000007840580010451959 s 2.05
Concat / IPartOpt / cpu / Primal 0.000016117 s 0.000007454679998772917 s 2.16
Concat / DefOpt / cpu / Primal 0.000016306999999999998 s 0.000009908619949783316 s 1.65
Concat / IDefOpt / cpu / Primal 0.000016539000000000002 s 0.000007624559993928415 s 2.17
Concat / JaXPipe / cpu / Forward 0.00002242 s 0.000011352860019542277 s 1.97
Concat / Jax / cpu / Forward 0.000022188 s 0.000011229959991396754 s 1.98
Concat / HLOOpt / cpu / Forward 0.000022354 s 0.0000151267399542121 s 1.48
Concat / PartOpt / cpu / Forward 0.000021775 s 0.000015686679971622653 s 1.39
Concat / IPartOpt / cpu / Forward 0.000022016 s 0.000011019880039384589 s 2.00
Concat / DefOpt / cpu / Forward 0.00002188 s 0.00001622339999812539 s 1.35
Concat / IDefOpt / cpu / Forward 0.000022195 s 0.000010681559979275337 s 2.08
Concat / JaXPipe / cpu / PreRev 0.000025236 s 0.000012957980043211136 s 1.95
Concat / JaXPipe / cpu / PostRev 0.000025003 s 0.000012530779968074057 s 2.00
Concat / JaXPipe / cpu / BothRev 0.000025029 s 0.000012426180019247113 s 2.01
Concat / Jax / cpu / BothRev 0.000024168 s 0.000013003280009797893 s 1.86
Concat / HLOOpt / cpu / PreRev 0.000024811 s 0.000013130240004102234 s 1.89
Concat / HLOOpt / cpu / PostRev 0.000024844 s 0.000016995599989968467 s 1.46
Concat / HLOOpt / cpu / BothRev 0.000024413 s 0.00001541774000543228 s 1.58
Concat / PartOpt / cpu / PreRev 0.000025257 s 0.000012957440012542066 s 1.95
Concat / PartOpt / cpu / PostRev 0.000025294 s 0.00001301497998611012 s 1.94
Concat / PartOpt / cpu / BothRev 0.000023995 s 0.000011962740018134356 s 2.01
Concat / IPartOpt / cpu / PreRev 0.000025651 s 0.000013326339985724187 s 1.92
Concat / IPartOpt / cpu / PostRev 0.000024446 s 0.000012175160009064712 s 2.01
Concat / IPartOpt / cpu / BothRev 0.000024617 s 0.0000123994800287619 s 1.99
Concat / DefOpt / cpu / PreRev 0.000024426 s 0.000012641379962587962 s 1.93
Concat / DefOpt / cpu / PostRev 0.000030613 s 0.000012612659947990325 s 2.43
Concat / DefOpt / cpu / BothRev 0.000024778 s 0.000012942380035383394 s 1.91
Concat / IDefOpt / cpu / PreRev 0.000024316 s 0.00001297918001000653 s 1.87
Concat / IDefOpt / cpu / PostRev 0.000024545 s 0.000013093399993522325 s 1.87
Concat / IDefOpt / cpu / BothRev 0.000024023 s 0.000012803519994122323 s 1.88
Concat / JaXPipe / cpu / Primal 0.000008 s 0.000008160360048350412 s 0.98
Concat / Jax / cpu / Primal 0.000008 s 0.000007911260008768294 s 1.01
Concat / HLOOpt / cpu / Primal 0.000008 s 0.00001136832001066068 s 0.70
Concat / PartOpt / cpu / Primal 0.000008 s 0.000007840580010451959 s 1.02
Concat / IPartOpt / cpu / Primal 0.000008 s 0.000007454679998772917 s 1.07
Concat / DefOpt / cpu / Primal 0.000008 s 0.000009908619949783316 s 0.81
Concat / IDefOpt / cpu / Primal 0.000008 s 0.000007624559993928415 s 1.05
Concat / JaXPipe / cpu / Forward 0.000011 s 0.000011352860019542277 s 0.97
Concat / Jax / cpu / Forward 0.000012 s 0.000011229959991396754 s 1.07
Concat / HLOOpt / cpu / Forward 0.000012 s 0.0000151267399542121 s 0.79
Concat / PartOpt / cpu / Forward 0.000011 s 0.000015686679971622653 s 0.70
Concat / IPartOpt / cpu / Forward 0.000011 s 0.000011019880039384589 s 1.00
Concat / DefOpt / cpu / Forward 0.000011 s 0.00001622339999812539 s 0.68
Concat / IDefOpt / cpu / Forward 0.000012 s 0.000010681559979275337 s 1.12
Concat / JaXPipe / cpu / PreRev 0.000013 s 0.000012957980043211136 s 1.00
Concat / JaXPipe / cpu / PostRev 0.000013 s 0.000012530779968074057 s 1.04
Concat / JaXPipe / cpu / BothRev 0.000013 s 0.000012426180019247113 s 1.05
Concat / Jax / cpu / BothRev 0.000013 s 0.000013003280009797893 s 1.00
Concat / HLOOpt / cpu / PreRev 0.000013 s 0.000013130240004102234 s 0.99
Concat / HLOOpt / cpu / PostRev 0.000013 s 0.000016995599989968467 s 0.76
Concat / HLOOpt / cpu / BothRev 0.000013 s 0.00001541774000543228 s 0.84
Concat / PartOpt / cpu / PreRev 0.000013 s 0.000012957440012542066 s 1.00
Concat / PartOpt / cpu / PostRev 0.000014 s 0.00001301497998611012 s 1.08
Concat / PartOpt / cpu / BothRev 0.000014 s 0.000011962740018134356 s 1.17
Concat / IPartOpt / cpu / PreRev 0.000014 s 0.000013326339985724187 s 1.05
Concat / IPartOpt / cpu / PostRev 0.000013 s 0.000012175160009064712 s 1.07
Concat / IPartOpt / cpu / BothRev 0.000013 s 0.0000123994800287619 s 1.05
Concat / DefOpt / cpu / PreRev 0.000013 s 0.000012641379962587962 s 1.03
Concat / DefOpt / cpu / PostRev 0.000013 s 0.000012612659947990325 s 1.03
Concat / DefOpt / cpu / BothRev 0.000013 s 0.000012942380035383394 s 1.00
Concat / IDefOpt / cpu / PreRev 0.000014 s 0.00001297918001000653 s 1.08
Concat / IDefOpt / cpu / PostRev 0.000013 s 0.000013093399993522325 s 0.99
Concat / IDefOpt / cpu / BothRev 0.000013 s 0.000012803519994122323 s 1.02
const_scatter / JaXPipe / cpu / Primal 0.000007141699989006156 s 0.000007573320035589859 s 0.94
const_scatter / Jax / cpu / Primal 0.00000694230005137797 s 0.000007523460035372409 s 0.92
const_scatter / HLOOpt / cpu / Primal 0.0000074823199884122 s 0.000007548280009359587 s 0.99
const_scatter / PartOpt / cpu / Primal 0.000007247040039146668 s 0.000007051299999147886 s 1.03
const_scatter / IPartOpt / cpu / Primal 0.0000068887600173184185 s 0.000007277360009538825 s 0.95
const_scatter / DefOpt / cpu / Primal 0.000011445499967521756 s 0.000011001980001310585 s 1.04
const_scatter / IDefOpt / cpu / Primal 0.000007529359963882598 s 0.0000073897600213967965 s 1.02
const_scatter / JaXPipe / cpu / Forward 0.00001097836003282282 s 0.000010801600001286716 s 1.02
const_scatter / Jax / cpu / Forward 0.00001139003996286192 s 0.000012025640007777838 s 0.95
const_scatter / HLOOpt / cpu / Forward 0.000014899380039423704 s 0.000015026320006654716 s 0.99
const_scatter / PartOpt / cpu / Forward 0.000014998199976616888 s 0.00001506768003309844 s 1.00
const_scatter / IPartOpt / cpu / Forward 0.000010500219996174563 s 0.000011029080005755532 s 0.95
const_scatter / DefOpt / cpu / Forward 0.000015119660056370775 s 0.00001408544005244039 s 1.07
const_scatter / IDefOpt / cpu / Forward 0.000010143900008188211 s 0.000010857140014195463 s 0.93
const_scatter / JaXPipe / cpu / PreRev 0.000305434380034 s 0.0002981160800482 s 1.02
const_scatter / JaXPipe / cpu / PostRev 0.0002917904799414 s 0.0002890498800024 s 1.01
const_scatter / JaXPipe / cpu / BothRev 0.0002839524999853 s 0.000284081560012 s 1.00
const_scatter / Jax / cpu / BothRev 0.0002842050599701 s 0.0002841971000088 s 1.00
const_scatter / HLOOpt / cpu / PreRev 0.0002900995600339 s 0.000284310520019 s 1.02
const_scatter / HLOOpt / cpu / PostRev 0.0002851976799956 s 0.0002856364000217 s 1.00
const_scatter / HLOOpt / cpu / BothRev 0.0002866028600237 s 0.0002858477599784 s 1.00
const_scatter / PartOpt / cpu / PreRev 0.0002848960799747 s 0.0002857958599724 s 1.00
const_scatter / PartOpt / cpu / PostRev 0.0002867019199766 s 0.0002826219000144 s 1.01
const_scatter / PartOpt / cpu / BothRev 0.0002901777799979 s 0.0002843083599873 s 1.02
const_scatter / IPartOpt / cpu / PreRev 0.000283422420025 s 0.0002835907000189 s 1.00
const_scatter / IPartOpt / cpu / PostRev 0.0002920498599542 s 0.0002907145999961 s 1.00
const_scatter / IPartOpt / cpu / BothRev 0.0002891780800018 s 0.0002838761000202 s 1.02
const_scatter / DefOpt / cpu / PreRev 0.0002859942000213 s 0.0002872262800246 s 1.00
const_scatter / DefOpt / cpu / PostRev 0.0002899448000334 s 0.000290691500013 s 1.00
const_scatter / DefOpt / cpu / BothRev 0.0002914633800355 s 0.0002843750600095 s 1.02
const_scatter / IDefOpt / cpu / PreRev 0.000286188480004 s 0.0002891831799752 s 0.99
const_scatter / IDefOpt / cpu / PostRev 0.0002913612600059 s 0.000292735439998 s 1.00
const_scatter / IDefOpt / cpu / BothRev 0.0002906637799787 s 0.0003014979000181 s 0.96
const_scatter / JaXPipe / cuda / Primal 0.000001887 s 0.0000019200000000000003 s 0.98
const_scatter / Jax / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / HLOOpt / cuda / Primal 0.000001887 s 0.0000019200000000000003 s 0.98
const_scatter / PartOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / IPartOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
const_scatter / DefOpt / cuda / Primal 0.000001887 s 0.000001889 s 1.00
const_scatter / IDefOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
const_scatter / JaXPipe / cuda / Forward 0.000009824 s 0.000009984 s 0.98
const_scatter / Jax / cuda / Forward 0.00000976 s 0.000009919 s 0.98
const_scatter / HLOOpt / cuda / Forward 0.000009696 s 0.000009888 s 0.98
const_scatter / PartOpt / cuda / Forward 0.000009984 s 0.000009728 s 1.03
const_scatter / IPartOpt / cuda / Forward 0.000009823 s 0.000009984 s 0.98
const_scatter / DefOpt / cuda / Forward 0.000010528 s 0.000012193 s 0.86
const_scatter / IDefOpt / cuda / Forward 0.000009952 s 0.000012224 s 0.81
const_scatter / JaXPipe / cuda / PreRev 0.000013376 s 0.000013088 s 1.02
const_scatter / JaXPipe / cuda / PostRev 0.00001664 s 0.000016351 s 1.02
const_scatter / JaXPipe / cuda / BothRev 0.000012833 s 0.000012736 s 1.01
const_scatter / Jax / cuda / BothRev 0.000016544 s 0.000016927000000000002 s 0.98
const_scatter / HLOOpt / cuda / PreRev 0.000012929 s 0.000012736 s 1.02
const_scatter / HLOOpt / cuda / PostRev 0.000013248 s 0.000015584000000000002 s 0.85
const_scatter / HLOOpt / cuda / BothRev 0.000012864 s 0.000012448 s 1.03
const_scatter / PartOpt / cuda / PreRev 0.000012992 s 0.000012992 s 1
const_scatter / PartOpt / cuda / PostRev 0.000016768999999999998 s 0.000017024 s 0.99
const_scatter / PartOpt / cuda / BothRev 0.00001296 s 0.000012992 s 1.00
const_scatter / IPartOpt / cuda / PreRev 0.00001312 s 0.00001312 s 1
const_scatter / IPartOpt / cuda / PostRev 0.000016992 s 0.000016512 s 1.03
const_scatter / IPartOpt / cuda / BothRev 0.000012896 s 0.000012961 s 0.99
const_scatter / DefOpt / cuda / PreRev 0.000012864 s 0.00001264 s 1.02
const_scatter / DefOpt / cuda / PostRev 0.000012832 s 0.000013248 s 0.97
const_scatter / DefOpt / cuda / BothRev 0.000013024 s 0.000012544 s 1.04
const_scatter / IDefOpt / cuda / PreRev 0.000012704 s 0.000012928 s 0.98
const_scatter / IDefOpt / cuda / PostRev 0.000012864 s 0.00001264 s 1.02
const_scatter / IDefOpt / cuda / BothRev 0.00001312 s 0.000012832 s 1.02
const_scatter / JaXPipe / tpu / Primal 0.000003813725 s 0.00000379155 s 1.01
const_scatter / Jax / tpu / Primal 0.000003819625 s 0.0000038203 s 1.00
const_scatter / HLOOpt / tpu / Primal 9.586e-7 s 9.5665e-7 s 1.00
const_scatter / PartOpt / tpu / Primal 0.000003814175 s 0.0000038067 s 1.00
const_scatter / IPartOpt / tpu / Primal 0.000003803 s 0.0000037985 s 1.00
const_scatter / DefOpt / tpu / Primal 9.66025e-7 s 9.773e-7 s 0.99
const_scatter / IDefOpt / tpu / Primal 9.4975e-7 s 9.63175e-7 s 0.99
const_scatter / JaXPipe / tpu / Forward 0.000001941425 s 0.000001931875 s 1.00
const_scatter / Jax / tpu / Forward 0.000006483850000000001 s 0.000006469125 s 1.00
const_scatter / HLOOpt / tpu / Forward 0.00000190735 s 0.00000192795 s 0.99
const_scatter / PartOpt / tpu / Forward 0.0000019404750000000003 s 0.0000019590750000000004 s 0.99
const_scatter / IPartOpt / tpu / Forward 0.0000019162 s 0.00000192115 s 1.00
const_scatter / DefOpt / tpu / Forward 0.00000195335 s 0.0000019458 s 1.00
const_scatter / IDefOpt / tpu / Forward 0.000001947675 s 0.0000019572 s 1.00
const_scatter / JaXPipe / tpu / PreRev 0.000004309875 s 0.0000043185 s 1.00
const_scatter / JaXPipe / tpu / PostRev 0.000006613474999999999 s 0.00000659855 s 1.00
const_scatter / JaXPipe / tpu / BothRev 0.00000432155 s 0.0000042997 s 1.01
const_scatter / Jax / tpu / BothRev 0.000006627175 s 0.00000659215 s 1.01
const_scatter / HLOOpt / tpu / PreRev 0.0000042988 s 0.000004292725 s 1.00
const_scatter / HLOOpt / tpu / PostRev 0.000004292050000000001 s 0.0000043021 s 1.00
const_scatter / HLOOpt / tpu / BothRev 0.000004300649999999999 s 0.00000432075 s 1.00
const_scatter / PartOpt / tpu / PreRev 0.00000429445 s 0.000004310975 s 1.00
const_scatter / PartOpt / tpu / PostRev 0.000006607725 s 0.000006599425 s 1.00
const_scatter / PartOpt / tpu / BothRev 0.00000429945 s 0.00000431505 s 1.00
const_scatter / IPartOpt / tpu / PreRev 0.00000430695 s 0.0000043056 s 1.00
const_scatter / IPartOpt / tpu / PostRev 0.000006610975 s 0.000006592400000000001 s 1.00
const_scatter / IPartOpt / tpu / BothRev 0.00000431085 s 0.0000043061 s 1.00
const_scatter / DefOpt / tpu / PreRev 0.000004310275 s 0.000004306125 s 1.00
const_scatter / DefOpt / tpu / PostRev 0.000004291475 s 0.0000042977500000000005 s 1.00
const_scatter / DefOpt / tpu / BothRev 0.000004315025 s 0.000004307875 s 1.00
const_scatter / IDefOpt / tpu / PreRev 0.000004306375 s 0.00000430395 s 1.00
const_scatter / IDefOpt / tpu / PostRev 0.0000043113000000000005 s 0.00000429625 s 1.00
const_scatter / IDefOpt / tpu / BothRev 0.00000431225 s 0.0000043191 s 1.00
const_scatter / JaXPipe / cpu / Primal 0.000016678 s 0.000007573320035589859 s 2.20
const_scatter / Jax / cpu / Primal 0.00001614 s 0.000007523460035372409 s 2.15
const_scatter / HLOOpt / cpu / Primal 0.000016151 s 0.000007548280009359587 s 2.14
const_scatter / PartOpt / cpu / Primal 0.000016274 s 0.000007051299999147886 s 2.31
const_scatter / IPartOpt / cpu / Primal 0.000016484 s 0.000007277360009538825 s 2.27
const_scatter / DefOpt / cpu / Primal 0.000015846 s 0.000011001980001310585 s 1.44
const_scatter / IDefOpt / cpu / Primal 0.000016294 s 0.0000073897600213967965 s 2.20
const_scatter / JaXPipe / cpu / Forward 0.000021637 s 0.000010801600001286716 s 2.00
const_scatter / Jax / cpu / Forward 0.000021009 s 0.000012025640007777838 s 1.75
const_scatter / HLOOpt / cpu / Forward 0.000021436 s 0.000015026320006654716 s 1.43
const_scatter / PartOpt / cpu / Forward 0.000021648 s 0.00001506768003309844 s 1.44
const_scatter / IPartOpt / cpu / Forward 0.000021095 s 0.000011029080005755532 s 1.91
const_scatter / DefOpt / cpu / Forward 0.000021012 s 0.00001408544005244039 s 1.49
const_scatter / IDefOpt / cpu / Forward 0.000021578 s 0.000010857140014195463 s 1.99
const_scatter / JaXPipe / cpu / PreRev 0.000532784 s 0.0002981160800482 s 1.79
const_scatter / JaXPipe / cpu / PostRev 0.0005369699999999 s 0.0002890498800024 s 1.86
const_scatter / JaXPipe / cpu / BothRev 0.00053263 s 0.000284081560012 s 1.87
const_scatter / Jax / cpu / BothRev 0.000542574 s 0.0002841971000088 s 1.91
const_scatter / HLOOpt / cpu / PreRev 0.000538908 s 0.000284310520019 s 1.90
const_scatter / HLOOpt / cpu / PostRev 0.000548785 s 0.0002856364000217 s 1.92
const_scatter / HLOOpt / cpu / BothRev 0.0005345609999999 s 0.0002858477599784 s 1.87
const_scatter / PartOpt / cpu / PreRev 0.000531448 s 0.0002857958599724 s 1.86
const_scatter / PartOpt / cpu / PostRev 0.000553349 s 0.0002826219000144 s 1.96
const_scatter / PartOpt / cpu / BothRev 0.00053338 s 0.0002843083599873 s 1.88
const_scatter / IPartOpt / cpu / PreRev 0.00052137 s 0.0002835907000189 s 1.84
const_scatter / IPartOpt / cpu / PostRev 0.000546307 s 0.0002907145999961 s 1.88
const_scatter / IPartOpt / cpu / BothRev 0.000534161 s 0.0002838761000202 s 1.88
const_scatter / DefOpt / cpu / PreRev 0.000530668 s 0.0002872262800246 s 1.85
const_scatter / DefOpt / cpu / PostRev 0.000537797 s 0.000290691500013 s 1.85
const_scatter / DefOpt / cpu / BothRev 0.0005306229999999 s 0.0002843750600095 s 1.87
const_scatter / IDefOpt / cpu / PreRev 0.00054104 s 0.0002891831799752 s 1.87
const_scatter / IDefOpt / cpu / PostRev 0.000536321 s 0.000292735439998 s 1.83
const_scatter / IDefOpt / cpu / BothRev 0.000570087 s 0.0003014979000181 s 1.89
const_scatter / JaXPipe / cpu / Primal 0.000008 s 0.000007573320035589859 s 1.06
const_scatter / Jax / cpu / Primal 0.000008 s 0.000007523460035372409 s 1.06
const_scatter / HLOOpt / cpu / Primal 0.000008 s 0.000007548280009359587 s 1.06
const_scatter / PartOpt / cpu / Primal 0.000008 s 0.000007051299999147886 s 1.13
const_scatter / IPartOpt / cpu / Primal 0.000008 s 0.000007277360009538825 s 1.10
const_scatter / DefOpt / cpu / Primal 0.000008 s 0.000011001980001310585 s 0.73
const_scatter / IDefOpt / cpu / Primal 0.000008 s 0.0000073897600213967965 s 1.08
const_scatter / JaXPipe / cpu / Forward 0.000011 s 0.000010801600001286716 s 1.02
const_scatter / Jax / cpu / Forward 0.000011 s 0.000012025640007777838 s 0.91
const_scatter / HLOOpt / cpu / Forward 0.000011 s 0.000015026320006654716 s 0.73
const_scatter / PartOpt / cpu / Forward 0.000012 s 0.00001506768003309844 s 0.80
const_scatter / IPartOpt / cpu / Forward 0.000011 s 0.000011029080005755532 s 1.00
const_scatter / DefOpt / cpu / Forward 0.000011 s 0.00001408544005244039 s 0.78
const_scatter / IDefOpt / cpu / Forward 0.000011 s 0.000010857140014195463 s 1.01
const_scatter / JaXPipe / cpu / PreRev 0.000317 s 0.0002981160800482 s 1.06
const_scatter / JaXPipe / cpu / PostRev 0.0003459999999999 s 0.0002890498800024 s 1.20
const_scatter / JaXPipe / cpu / BothRev 0.000317 s 0.000284081560012 s 1.12
const_scatter / Jax / cpu / BothRev 0.000318 s 0.0002841971000088 s 1.12
const_scatter / HLOOpt / cpu / PreRev 0.000323 s 0.000284310520019 s 1.14
const_scatter / HLOOpt / cpu / PostRev 0.000319 s 0.0002856364000217 s 1.12
const_scatter / HLOOpt / cpu / BothRev 0.000333 s 0.0002858477599784 s 1.16
const_scatter / PartOpt / cpu / PreRev 0.000319 s 0.0002857958599724 s 1.12
const_scatter / PartOpt / cpu / PostRev 0.000347 s 0.0002826219000144 s 1.23
const_scatter / PartOpt / cpu / BothRev 0.000326 s 0.0002843083599873 s 1.15
const_scatter / IPartOpt / cpu / PreRev 0.000323 s 0.0002835907000189 s 1.14
const_scatter / IPartOpt / cpu / PostRev 0.000338 s 0.0002907145999961 s 1.16
const_scatter / IPartOpt / cpu / BothRev 0.000336 s 0.0002838761000202 s 1.18
const_scatter / DefOpt / cpu / PreRev 0.000316 s 0.0002872262800246 s 1.10
const_scatter / DefOpt / cpu / PostRev 0.000319 s 0.000290691500013 s 1.10
const_scatter / DefOpt / cpu / BothRev 0.00032 s 0.0002843750600095 s 1.13
const_scatter / IDefOpt / cpu / PreRev 0.00032 s 0.0002891831799752 s 1.11
const_scatter / IDefOpt / cpu / PostRev 0.000363 s 0.000292735439998 s 1.24
const_scatter / IDefOpt / cpu / BothRev 0.0003509999999999 s 0.0003014979000181 s 1.16
GenDot / JaXPipe / cpu / Primal 0.000009508599978289569 s 0.000009190540004055948 s 1.03
GenDot / Jax / cpu / Primal 0.000008288700018965756 s 0.00000810308001746307 s 1.02
GenDot / HLOOpt / cpu / Primal 0.00001167996003459848 s 0.000011699080032485651 s 1.00
GenDot / PartOpt / cpu / Primal 0.00000778105995777878 s 0.000008187180010281735 s 0.95
GenDot / IPartOpt / cpu / Primal 0.000007946920004542334 s 0.000009089939994737508 s 0.87
GenDot / DefOpt / cpu / Primal 0.000008645779953440069 s 0.00001104285997826082 s 0.78
GenDot / IDefOpt / cpu / Primal 0.00000836214000628388 s 0.000007813940001142328 s 1.07
GenDot / JaXPipe / cpu / Forward 0.000012009800020678083 s 0.000011676719977913308 s 1.03
GenDot / Jax / cpu / Forward 0.000011514360012370162 s 0.000010420540029372205 s 1.10
GenDot / HLOOpt / cpu / Forward 0.000016572440008530974 s 0.000016459399985251368 s 1.01
GenDot / PartOpt / cpu / Forward 0.000016553639998164727 s 0.000016939759998422232 s 0.98
GenDot / IPartOpt / cpu / Forward 0.00001198662002934725 s 0.000010930179987553857 s 1.10
GenDot / DefOpt / cpu / Forward 0.000016897819987207184 s 0.00001704338001218275 s 0.99
GenDot / IDefOpt / cpu / Forward 0.00001185868001812196 s 0.000011517780021677029 s 1.03
GenDot / JaXPipe / cpu / PreRev 0.000012655560021812562 s 0.000012354499976936496 s 1.02
GenDot / JaXPipe / cpu / PostRev 0.00001141107998591906 s 0.000011731519962268066 s 0.97
GenDot / JaXPipe / cpu / BothRev 0.000012110779998693031 s 0.000011670880021483751 s 1.04
GenDot / Jax / cpu / BothRev 0.00001224362002176349 s 0.000012271280011191266 s 1.00
GenDot / HLOOpt / cpu / PreRev 0.000011825919991679256 s 0.000011541980029505794 s 1.02
GenDot / HLOOpt / cpu / PostRev 0.000016077440004664823 s 0.000015967959989211522 s 1.01
GenDot / HLOOpt / cpu / BothRev 0.000013656219953190884 s 0.000013849659962943406 s 0.99
GenDot / PartOpt / cpu / PreRev 0.000011642040026345058 s 0.000011940159974983544 s 0.98
GenDot / PartOpt / cpu / PostRev 0.000010581360011201468 s 0.000011580260015762176 s 0.91
GenDot / PartOpt / cpu / BothRev 0.000012204200020278222 s 0.000012290200002098571 s 0.99
GenDot / IPartOpt / cpu / PreRev 0.000011700840022967896 s 0.00001575883999976213 s 0.74
GenDot / IPartOpt / cpu / PostRev 0.000010942740000245976 s 0.000011689839984683204 s 0.94
GenDot / IPartOpt / cpu / BothRev 0.000011756859994420663 s 0.00001164310002423008 s 1.01
GenDot / DefOpt / cpu / PreRev 0.000011950959997193424 s 0.00001241987998582772 s 0.96
GenDot / DefOpt / cpu / PostRev 0.000011592180007937714 s 0.000011746400014089886 s 0.99
GenDot / DefOpt / cpu / BothRev 0.00001205503996061452 s 0.000012488020001910626 s 0.97
GenDot / IDefOpt / cpu / PreRev 0.000011055380009565853 s 0.000011414039981900714 s 0.97
GenDot / IDefOpt / cpu / PostRev 0.0000116665799941984 s 0.000012164240006313776 s 0.96
GenDot / IDefOpt / cpu / BothRev 0.000011978959992120509 s 0.000012146600020059853 s 0.99
GenDot / JaXPipe / cuda / Primal 0.000002015 s 0.000002016 s 1.00
GenDot / Jax / cuda / Primal 0.000002015 s 0.000002016 s 1.00
GenDot / HLOOpt / cuda / Primal 0.000002016 s 0.000001984 s 1.02
GenDot / PartOpt / cuda / Primal 0.000002015 s 0.000002016 s 1.00
GenDot / IPartOpt / cuda / Primal 0.000002015 s 0.000002015 s 1
GenDot / DefOpt / cuda / Primal 0.000002015 s 0.000002016 s 1.00
GenDot / IDefOpt / cuda / Primal 0.000002015 s 0.000002015 s 1
GenDot / JaXPipe / cuda / Forward 0.000010336 s 0.000010752 s 0.96
GenDot / Jax / cuda / Forward 0.000010848 s 0.000010496 s 1.03
GenDot / HLOOpt / cuda / Forward 0.000010496 s 0.000010848 s 0.97
GenDot / PartOpt / cuda / Forward 0.000010048 s 0.000010816 s 0.93
GenDot / IPartOpt / cuda / Forward 0.000009728 s 0.00001008 s 0.97
GenDot / DefOpt / cuda / Forward 0.000010176 s 0.000010209 s 1.00
GenDot / IDefOpt / cuda / Forward 0.00001008 s 0.000010208 s 0.99
GenDot / JaXPipe / cuda / PreRev 0.00001008 s 0.00001024 s 0.98
GenDot / JaXPipe / cuda / PostRev 0.000010272 s 0.000010144 s 1.01
GenDot / JaXPipe / cuda / BothRev 0.000010175 s 0.000009888 s 1.03
GenDot / Jax / cuda / BothRev 0.000010176 s 0.000009408 s 1.08
GenDot / HLOOpt / cuda / PreRev 0.00000992 s 0.000010464 s 0.95
GenDot / HLOOpt / cuda / PostRev 0.000010143 s 0.00001024 s 0.99
GenDot / HLOOpt / cuda / BothRev 0.000010239 s 0.000012512 s 0.82
GenDot / PartOpt / cuda / PreRev 0.000010208 s 0.000010368 s 0.98
GenDot / PartOpt / cuda / PostRev 0.0000096 s 0.000010336 s 0.93
GenDot / PartOpt / cuda / BothRev 0.000010048 s 0.0000112 s 0.90
GenDot / IPartOpt / cuda / PreRev 0.000010176 s 0.000009952 s 1.02
GenDot / IPartOpt / cuda / PostRev 0.000010272 s 0.00001024 s 1.00
GenDot / IPartOpt / cuda / BothRev 0.000010016 s 0.000010369 s 0.97
GenDot / DefOpt / cuda / PreRev 0.000010751 s 0.000009664 s 1.11
GenDot / DefOpt / cuda / PostRev 0.000009856 s 0.000010496 s 0.94
GenDot / DefOpt / cuda / BothRev 0.000010304 s 0.000010208 s 1.01
GenDot / IDefOpt / cuda / PreRev 0.000010176 s 0.000010336 s 0.98
GenDot / IDefOpt / cuda / PostRev 0.000009984 s 0.000010016 s 1.00
GenDot / IDefOpt / cuda / BothRev 0.000009952 s 0.000010144 s 0.98
GenDot / JaXPipe / tpu / Primal 9.29875e-7 s 9.26225e-7 s 1.00
GenDot / Jax / tpu / Primal 9.35725e-7 s 9.359e-7 s 1.00
GenDot / HLOOpt / tpu / Primal 0.000001578525 s 0.00000155945 s 1.01
GenDot / PartOpt / tpu / Primal 9.3595e-7 s 9.35975e-7 s 1.00
GenDot / IPartOpt / tpu / Primal 9.398e-7 s 9.35525e-7 s 1.00
GenDot / DefOpt / tpu / Primal 0.00000149035 s 0.00000148685 s 1.00
GenDot / IDefOpt / tpu / Primal 0.0000015889000000000002 s 0.0000015564 s 1.02
GenDot / JaXPipe / tpu / Forward 0.000003169925 s 0.000003165 s 1.00
GenDot / Jax / tpu / Forward 0.00000232775 s 0.000002322725 s 1.00
GenDot / HLOOpt / tpu / Forward 0.000003121175 s 0.000003111825 s 1.00
GenDot / PartOpt / tpu / Forward 0.0000032152500000000003 s 0.00000320915 s 1.00
GenDot / IPartOpt / tpu / Forward 0.0000031147 s 0.000003112175 s 1.00
GenDot / DefOpt / tpu / Forward 0.00000323245 s 0.0000032118000000000003 s 1.01
GenDot / IDefOpt / tpu / Forward 0.000003116575 s 0.000003113175 s 1.00
GenDot / JaXPipe / tpu / PreRev 0.0000029791250000000005 s 0.000002946625 s 1.01
GenDot / JaXPipe / tpu / PostRev 0.0000024071 s 0.000002412075 s 1.00
GenDot / JaXPipe / tpu / BothRev 0.0000029604999999999994 s 0.0000029560750000000003 s 1.00
GenDot / Jax / tpu / BothRev 0.000002404925 s 0.000002409425 s 1.00
GenDot / HLOOpt / tpu / PreRev 0.0000029634500000000004 s 0.00000295045 s 1.00
GenDot / HLOOpt / tpu / PostRev 0.0000029373250000000003 s 0.000002927325 s 1.00
GenDot / HLOOpt / tpu / BothRev 0.000002959 s 0.00000294865 s 1.00
GenDot / PartOpt / tpu / PreRev 0.0000029435250000000003 s 0.00000292505 s 1.01
GenDot / PartOpt / tpu / PostRev 0.0000023957500000000003 s 0.0000023912 s 1.00
GenDot / PartOpt / tpu / BothRev 0.000002936825 s 0.0000029281000000000004 s 1.00
GenDot / IPartOpt / tpu / PreRev 0.0000029734500000000003 s 0.0000029494749999999994 s 1.01
GenDot / IPartOpt / tpu / PostRev 0.000002409025 s 0.00000240535 s 1.00
GenDot / IPartOpt / tpu / BothRev 0.000002961175 s 0.0000029496 s 1.00
GenDot / DefOpt / tpu / PreRev 0.0000029414249999999995 s 0.000002947675 s 1.00
GenDot / DefOpt / tpu / PostRev 0.000002972975 s 0.000002950175 s 1.01
GenDot / DefOpt / tpu / BothRev 0.000002934325 s 0.00000292765 s 1.00
GenDot / IDefOpt / tpu / PreRev 0.00000296485 s 0.0000029554750000000003 s 1.00
GenDot / IDefOpt / tpu / PostRev 0.000002935075 s 0.0000029356 s 1.00
GenDot / IDefOpt / tpu / BothRev 0.00000296545 s 0.000002954825 s 1.00
GenDot / JaXPipe / cpu / Primal 0.000018739 s 0.000009190540004055948 s 2.04
GenDot / Jax / cpu / Primal 0.000018857 s 0.00000810308001746307 s 2.33
GenDot / HLOOpt / cpu / Primal 0.000017616 s 0.000011699080032485651 s 1.51
GenDot / PartOpt / cpu / Primal 0.00001909 s 0.000008187180010281735 s 2.33
GenDot / IPartOpt / cpu / Primal 0.000017612 s 0.000009089939994737508 s 1.94
GenDot / DefOpt / cpu / Primal 0.000017541 s 0.00001104285997826082 s 1.59
GenDot / IDefOpt / cpu / Primal 0.000017628 s 0.000007813940001142328 s 2.26
GenDot / JaXPipe / cpu / Forward 0.000024619 s 0.000011676719977913308 s 2.11
GenDot / Jax / cpu / Forward 0.000025188 s 0.000010420540029372205 s 2.42
GenDot / HLOOpt / cpu / Forward 0.000024207 s 0.000016459399985251368 s 1.47
GenDot / PartOpt / cpu / Forward 0.000024117 s 0.000016939759998422232 s 1.42
GenDot / IPartOpt / cpu / Forward 0.000023921 s 0.000010930179987553857 s 2.19
GenDot / DefOpt / cpu / Forward 0.000024457 s 0.00001704338001218275 s 1.43
GenDot / IDefOpt / cpu / Forward 0.000024305 s 0.000011517780021677029 s 2.11
GenDot / JaXPipe / cpu / PreRev 0.000024753 s 0.000012354499976936496 s 2.00
GenDot / JaXPipe / cpu / PostRev 0.000026256 s 0.000011731519962268066 s 2.24
GenDot / JaXPipe / cpu / BothRev 0.000025273 s 0.000011670880021483751 s 2.17
GenDot / Jax / cpu / BothRev 0.00002605 s 0.000012271280011191266 s 2.12
GenDot / HLOOpt / cpu / PreRev 0.00002447 s 0.000011541980029505794 s 2.12
GenDot / HLOOpt / cpu / PostRev 0.000024674 s 0.000015967959989211522 s 1.55
GenDot / HLOOpt / cpu / BothRev 0.000024659 s 0.000013849659962943406 s 1.78
GenDot / PartOpt / cpu / PreRev 0.000024228 s 0.000011940159974983544 s 2.03
GenDot / PartOpt / cpu / PostRev 0.000025165 s 0.000011580260015762176 s 2.17
GenDot / PartOpt / cpu / BothRev 0.000024134 s 0.000012290200002098571 s 1.96
GenDot / IPartOpt / cpu / PreRev 0.00002427 s 0.00001575883999976213 s 1.54
GenDot / IPartOpt / cpu / PostRev 0.000024603 s 0.000011689839984683204 s 2.10
GenDot / IPartOpt / cpu / BothRev 0.000025241 s 0.00001164310002423008 s 2.17
GenDot / DefOpt / cpu / PreRev 0.000023738 s 0.00001241987998582772 s 1.91
GenDot / DefOpt / cpu / PostRev 0.000024923 s 0.000011746400014089886 s 2.12
GenDot / DefOpt / cpu / BothRev 0.000023334 s 0.000012488020001910626 s 1.87
GenDot / IDefOpt / cpu / PreRev 0.000024452 s 0.000011414039981900714 s 2.14
GenDot / IDefOpt / cpu / PostRev 0.000023997 s 0.000012164240006313776 s 1.97
GenDot / IDefOpt / cpu / BothRev 0.000024693 s 0.000012146600020059853 s 2.03
GenDot / JaXPipe / cpu / Primal 0.00001 s 0.000009190540004055948 s 1.09
GenDot / Jax / cpu / Primal 0.000008999999999999999 s 0.00000810308001746307 s 1.11
GenDot / HLOOpt / cpu / Primal 0.00001 s 0.000011699080032485651 s 0.85
GenDot / PartOpt / cpu / Primal 0.00001 s 0.000008187180010281735 s 1.22
GenDot / IPartOpt / cpu / Primal 0.00001 s 0.000009089939994737508 s 1.10
GenDot / DefOpt / cpu / Primal 0.000008999999999999999 s 0.00001104285997826082 s 0.82
GenDot / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000007813940001142328 s 1.15
GenDot / JaXPipe / cpu / Forward 0.000013 s 0.000011676719977913308 s 1.11
GenDot / Jax / cpu / Forward 0.000013 s 0.000010420540029372205 s 1.25
GenDot / HLOOpt / cpu / Forward 0.000013 s 0.000016459399985251368 s 0.79
GenDot / PartOpt / cpu / Forward 0.000013 s 0.000016939759998422232 s 0.77
GenDot / IPartOpt / cpu / Forward 0.000013 s 0.000010930179987553857 s 1.19
GenDot / DefOpt / cpu / Forward 0.000013 s 0.00001704338001218275 s 0.76
GenDot / IDefOpt / cpu / Forward 0.000013 s 0.000011517780021677029 s 1.13
GenDot / JaXPipe / cpu / PreRev 0.000013 s 0.000012354499976936496 s 1.05
GenDot / JaXPipe / cpu / PostRev 0.000014 s 0.000011731519962268066 s 1.19
GenDot / JaXPipe / cpu / BothRev 0.000013 s 0.000011670880021483751 s 1.11
GenDot / Jax / cpu / BothRev 0.000013 s 0.000012271280011191266 s 1.06
GenDot / HLOOpt / cpu / PreRev 0.000013 s 0.000011541980029505794 s 1.13
GenDot / HLOOpt / cpu / PostRev 0.000014 s 0.000015967959989211522 s 0.88
GenDot / HLOOpt / cpu / BothRev 0.000014 s 0.000013849659962943406 s 1.01
GenDot / PartOpt / cpu / PreRev 0.000013 s 0.000011940159974983544 s 1.09
GenDot / PartOpt / cpu / PostRev 0.000014 s 0.000011580260015762176 s 1.21
GenDot / PartOpt / cpu / BothRev 0.000013 s 0.000012290200002098571 s 1.06
GenDot / IPartOpt / cpu / PreRev 0.000013 s 0.00001575883999976213 s 0.82
GenDot / IPartOpt / cpu / PostRev 0.000014 s 0.000011689839984683204 s 1.20
GenDot / IPartOpt / cpu / BothRev 0.000014 s 0.00001164310002423008 s 1.20
GenDot / DefOpt / cpu / PreRev 0.000013 s 0.00001241987998582772 s 1.05
GenDot / DefOpt / cpu / PostRev 0.000014 s 0.000011746400014089886 s 1.19
GenDot / DefOpt / cpu / BothRev 0.000014 s 0.000012488020001910626 s 1.12
GenDot / IDefOpt / cpu / PreRev 0.000013 s 0.000011414039981900714 s 1.14
GenDot / IDefOpt / cpu / PostRev 0.000014 s 0.000012164240006313776 s 1.15
GenDot / IDefOpt / cpu / BothRev 0.000013 s 0.000012146600020059853 s 1.07
hlo_ffi / JaXPipe / cpu / Primal 0.000012626280022232094 s 0.000010819560029631248 s 1.17
hlo_ffi / Jax / cpu / Primal 0.000010932599998341177 s 0.000009936580036082886 s 1.10
hlo_ffi / HLOOpt / cpu / Primal 0.000014818019953963813 s 0.000013691060021301382 s 1.08
hlo_ffi / PartOpt / cpu / Primal 0.00001095172005079803 s 0.000009589139981471815 s 1.14
hlo_ffi / IPartOpt / cpu / Primal 0.000010831580002559347 s 0.000009444800007258893 s 1.15
hlo_ffi / DefOpt / cpu / Primal 0.000015067760014062514 s 0.000014216780018614372 s 1.06
hlo_ffi / IDefOpt / cpu / Primal 0.00001085531998796796 s 0.000009802700005820951 s 1.11
hlo_ffi / JaXPipe / cpu / Forward 0.000016172660043594078 s 0.000015186179989541416 s 1.06
hlo_ffi / Jax / cpu / Forward 0.00001623877999918477 s 0.000014690119996885189 s 1.11
hlo_ffi / HLOOpt / cpu / Forward 0.000016390839982705073 s 0.00001446758000383852 s 1.13
hlo_ffi / PartOpt / cpu / Forward 0.000016610080028840456 s 0.000014529620002576847 s 1.14
hlo_ffi / IPartOpt / cpu / Forward 0.000015600099986841086 s 0.000014481360003628652 s 1.08
hlo_ffi / DefOpt / cpu / Forward 0.000016074259983724916 s 0.00001496964000580192 s 1.07
hlo_ffi / IDefOpt / cpu / Forward 0.000015632760059816065 s 0.000014544020041284968 s 1.07
hlo_ffi / JaXPipe / cpu / PreRev 0.000016531880019101662 s 0.000014137960015432328 s 1.17
hlo_ffi / JaXPipe / cpu / PostRev 0.00001567073994920065 s 0.000014079540051170625 s 1.11
hlo_ffi / JaXPipe / cpu / BothRev 0.000015622519986209226 s 0.000014294059974417906 s 1.09
hlo_ffi / Jax / cpu / BothRev 0.000017453780001233098 s 0.0000134094399982132 s 1.30
hlo_ffi / HLOOpt / cpu / PreRev 0.00001636621996112808 s 0.000014196640004229266 s 1.15
hlo_ffi / HLOOpt / cpu / PostRev 0.000015991699983715078 s 0.000013641719942825147 s 1.17
hlo_ffi / HLOOpt / cpu / BothRev 0.000017535420038257144 s 0.00001572217998727865 s 1.12
hlo_ffi / PartOpt / cpu / PreRev 0.00001608049999049399 s 0.000014185019972501325 s 1.13
hlo_ffi / PartOpt / cpu / PostRev 0.00001623875999939628 s 0.000013708520000363932 s 1.18
hlo_ffi / PartOpt / cpu / BothRev 0.00001634617996387533 s 0.000014153360007185256 s 1.15
hlo_ffi / IPartOpt / cpu / PreRev 0.000016015919973142444 s 0.000014302419958767133 s 1.12
hlo_ffi / IPartOpt / cpu / PostRev 0.00001603484000042954 s 0.000014463979950960492 s 1.11
hlo_ffi / IPartOpt / cpu / BothRev 0.00001600030002919084 s 0.00001367498001854983 s 1.17
hlo_ffi / DefOpt / cpu / PreRev 0.000016398680018028244 s 0.000014264779993027331 s 1.15
hlo_ffi / DefOpt / cpu / PostRev 0.00001659312000811042 s 0.00001391776000673417 s 1.19
hlo_ffi / DefOpt / cpu / BothRev 0.00001602299998012313 s 0.00001432760001080169 s 1.12
hlo_ffi / IDefOpt / cpu / PreRev 0.000016741400004320893 s 0.000014373059966601432 s 1.16
hlo_ffi / IDefOpt / cpu / PostRev 0.00001608756001587608 s 0.000013848860035068356 s 1.16
hlo_ffi / IDefOpt / cpu / BothRev 0.000017862079985206947 s 0.000014241379985833192 s 1.25
hlo_ffi / JaXPipe / cuda / Primal 0.000001984 s 0.000001984 s 1
hlo_ffi / Jax / cuda / Primal 0.000001984 s 0.000001984 s 1
hlo_ffi / HLOOpt / cuda / Primal 0.000001983 s 0.000001984 s 1.00
hlo_ffi / PartOpt / cuda / Primal 0.000001983 s 0.000001984 s 1.00
hlo_ffi / IPartOpt / cuda / Primal 0.000001983 s 0.000002015 s 0.98
hlo_ffi / DefOpt / cuda / Primal 0.000001984 s 0.000001983 s 1.00
hlo_ffi / IDefOpt / cuda / Primal 0.000001983 s 0.000001984 s 1.00
hlo_ffi / JaXPipe / cuda / Forward 0.000002079 s 0.00000208 s 1.00
hlo_ffi / Jax / cuda / Forward 0.00000208 s 0.00000208 s 1
hlo_ffi / HLOOpt / cuda / Forward 0.000002079 s 0.00000208 s 1.00
hlo_ffi / PartOpt / cuda / Forward 0.000002079 s 0.00000208 s 1.00
hlo_ffi / IPartOpt / cuda / Forward 0.00000208 s 0.000002079 s 1.00
hlo_ffi / DefOpt / cuda / Forward 0.000002079 s 0.00000208 s 1.00
hlo_ffi / IDefOpt / cuda / Forward 0.000002079 s 0.000002079 s 1
hlo_ffi / JaXPipe / cuda / PreRev 0.000002048 s 0.000002048 s 1
hlo_ffi / JaXPipe / cuda / PostRev 0.000002047 s 0.000002047 s 1
hlo_ffi / JaXPipe / cuda / BothRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / Jax / cuda / BothRev 0.000002048 s 0.000002047 s 1.00
hlo_ffi / HLOOpt / cuda / PreRev 0.000002048 s 0.000002048 s 1
hlo_ffi / HLOOpt / cuda / PostRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / HLOOpt / cuda / BothRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / PartOpt / cuda / PreRev 0.000002048 s 0.000002048 s 1
hlo_ffi / PartOpt / cuda / PostRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / PartOpt / cuda / BothRev 0.000002048 s 0.000002048 s 1
hlo_ffi / IPartOpt / cuda / PreRev 0.000002048 s 0.000002048 s 1
hlo_ffi / IPartOpt / cuda / PostRev 0.000002048 s 0.000002047 s 1.00
hlo_ffi / IPartOpt / cuda / BothRev 0.000002048 s 0.000002048 s 1
hlo_ffi / DefOpt / cuda / PreRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / DefOpt / cuda / PostRev 0.000002048 s 0.000002048 s 1
hlo_ffi / DefOpt / cuda / BothRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / IDefOpt / cuda / PreRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / IDefOpt / cuda / PostRev 0.000002049 s 0.000002047 s 1.00
hlo_ffi / IDefOpt / cuda / BothRev 0.000002048 s 0.000002047 s 1.00
hlo_ffi / JaXPipe / tpu / Primal 9.20075e-7 s 9.26775e-7 s 0.99
hlo_ffi / Jax / tpu / Primal 9.504e-7 s 9.529e-7 s 1.00
hlo_ffi / HLOOpt / tpu / Primal 8.98825e-7 s 9.06e-7 s 0.99
hlo_ffi / PartOpt / tpu / Primal 9.51425e-7 s 9.48675e-7 s 1.00
hlo_ffi / IPartOpt / tpu / Primal 8.981250000000001e-7 s 9.07e-7 s 0.99
hlo_ffi / DefOpt / tpu / Primal 9.515e-7 s 9.516e-7 s 1.00
hlo_ffi / IDefOpt / tpu / Primal 8.980499999999999e-7 s 9.047e-7 s 0.99
hlo_ffi / JaXPipe / tpu / Forward 9.5e-7 s 9.48825e-7 s 1.00
hlo_ffi / Jax / tpu / Forward 9.819500000000002e-7 s 9.81775e-7 s 1.00
hlo_ffi / HLOOpt / tpu / Forward 9.74725e-7 s 9.737e-7 s 1.00
hlo_ffi / PartOpt / tpu / Forward 9.3425e-7 s 9.3345e-7 s 1.00
hlo_ffi / IPartOpt / tpu / Forward 9.7455e-7 s 9.7355e-7 s 1.00
hlo_ffi / DefOpt / tpu / Forward 9.3465e-7 s 9.3395e-7 s 1.00
hlo_ffi / IDefOpt / tpu / Forward 9.74525e-7 s 9.734e-7 s 1.00
hlo_ffi / JaXPipe / tpu / PreRev 9.32275e-7 s 9.39075e-7 s 0.99
hlo_ffi / JaXPipe / tpu / PostRev 9.65275e-7 s 9.6425e-7 s 1.00
hlo_ffi / JaXPipe / tpu / BothRev 9.6025e-7 s 9.59975e-7 s 1.00
hlo_ffi / Jax / tpu / BothRev 9.654749999999998e-7 s 9.64475e-7 s 1.00
hlo_ffi / HLOOpt / tpu / PreRev 9.60225e-7 s 9.6005e-7 s 1.00
hlo_ffi / HLOOpt / tpu / PostRev 9.6525e-7 s 9.6475e-7 s 1.00
hlo_ffi / HLOOpt / tpu / BothRev 9.60325e-7 s 9.5975e-7 s 1.00
hlo_ffi / PartOpt / tpu / PreRev 9.65325e-7 s 9.64175e-7 s 1.00
hlo_ffi / PartOpt / tpu / PostRev 9.60575e-7 s 9.59875e-7 s 1.00
hlo_ffi / PartOpt / tpu / BothRev 9.65625e-7 s 9.64525e-7 s 1.00
hlo_ffi / IPartOpt / tpu / PreRev 9.6055e-7 s 9.59925e-7 s 1.00
hlo_ffi / IPartOpt / tpu / PostRev 9.65825e-7 s 9.648e-7 s 1.00
hlo_ffi / IPartOpt / tpu / BothRev 9.60175e-7 s 9.59825e-7 s 1.00
hlo_ffi / DefOpt / tpu / PreRev 9.65775e-7 s 9.64425e-7 s 1.00
hlo_ffi / DefOpt / tpu / PostRev 9.60325e-7 s 9.59825e-7 s 1.00
hlo_ffi / DefOpt / tpu / BothRev 9.6555e-7 s 9.64375e-7 s 1.00
hlo_ffi / IDefOpt / tpu / PreRev 9.60325e-7 s 9.599e-7 s 1.00
hlo_ffi / IDefOpt / tpu / PostRev 9.6555e-7 s 9.64825e-7 s 1.00
hlo_ffi / IDefOpt / tpu / BothRev 9.6035e-7 s 9.5975e-7 s 1.00
hlo_ffi / JaXPipe / cpu / Primal 0.000021987 s 0.000010819560029631248 s 2.03
hlo_ffi / Jax / cpu / Primal 0.000021757 s 0.000009936580036082886 s 2.19
hlo_ffi / HLOOpt / cpu / Primal 0.000022285 s 0.000013691060021301382 s 1.63
hlo_ffi / PartOpt / cpu / Primal 0.000022083 s 0.000009589139981471815 s 2.30
hlo_ffi / IPartOpt / cpu / Primal 0.000022137 s 0.000009444800007258893 s 2.34
hlo_ffi / DefOpt / cpu / Primal 0.000022002 s 0.000014216780018614372 s 1.55
hlo_ffi / IDefOpt / cpu / Primal 0.000022188 s 0.000009802700005820951 s 2.26
hlo_ffi / JaXPipe / cpu / Forward 0.000036571 s 0.000015186179989541416 s 2.41
hlo_ffi / Jax / cpu / Forward 0.000030222 s 0.000014690119996885189 s 2.06
hlo_ffi / HLOOpt / cpu / Forward 0.00002986 s 0.00001446758000383852 s 2.06
hlo_ffi / PartOpt / cpu / Forward 0.000029953 s 0.000014529620002576847 s 2.06
hlo_ffi / IPartOpt / cpu / Forward 0.00003 s 0.000014481360003628652 s 2.07
hlo_ffi / DefOpt / cpu / Forward 0.000030321 s 0.00001496964000580192 s 2.03
hlo_ffi / IDefOpt / cpu / Forward 0.000029868 s 0.000014544020041284968 s 2.05
hlo_ffi / JaXPipe / cpu / PreRev 0.000030398000000000003 s 0.000014137960015432328 s 2.15
hlo_ffi / JaXPipe / cpu / PostRev 0.000031362 s 0.000014079540051170625 s 2.23
hlo_ffi / JaXPipe / cpu / BothRev 0.000031526 s 0.000014294059974417906 s 2.21
hlo_ffi / Jax / cpu / BothRev 0.000030602 s 0.0000134094399982132 s 2.28
hlo_ffi / HLOOpt / cpu / PreRev 0.000030273 s 0.000014196640004229266 s 2.13
hlo_ffi / HLOOpt / cpu / PostRev 0.000032443 s 0.000013641719942825147 s 2.38
hlo_ffi / HLOOpt / cpu / BothRev 0.000031377 s 0.00001572217998727865 s 2.00
hlo_ffi / PartOpt / cpu / PreRev 0.000030831000000000003 s 0.000014185019972501325 s 2.17
hlo_ffi / PartOpt / cpu / PostRev 0.000036048 s 0.000013708520000363932 s 2.63
hlo_ffi / PartOpt / cpu / BothRev 0.000032112 s 0.000014153360007185256 s 2.27
hlo_ffi / IPartOpt / cpu / PreRev 0.000031482 s 0.000014302419958767133 s 2.20
hlo_ffi / IPartOpt / cpu / PostRev 0.000030427 s 0.000014463979950960492 s 2.10
hlo_ffi / IPartOpt / cpu / BothRev 0.00003028 s 0.00001367498001854983 s 2.21
hlo_ffi / DefOpt / cpu / PreRev 0.000030304 s 0.000014264779993027331 s 2.12
hlo_ffi / DefOpt / cpu / PostRev 0.00003658 s 0.00001391776000673417 s 2.63
hlo_ffi / DefOpt / cpu / BothRev 0.000030637 s 0.00001432760001080169 s 2.14
hlo_ffi / IDefOpt / cpu / PreRev 0.00003155 s 0.000014373059966601432 s 2.20
hlo_ffi / IDefOpt / cpu / PostRev 0.000031179 s 0.000013848860035068356 s 2.25
hlo_ffi / IDefOpt / cpu / BothRev 0.000031804 s 0.000014241379985833192 s 2.23
hlo_ffi / JaXPipe / cpu / Primal 0.000013 s 0.000010819560029631248 s 1.20
hlo_ffi / Jax / cpu / Primal 0.000013 s 0.000009936580036082886 s 1.31
hlo_ffi / HLOOpt / cpu / Primal 0.000013 s 0.000013691060021301382 s 0.95
hlo_ffi / PartOpt / cpu / Primal 0.000013 s 0.000009589139981471815 s 1.36
hlo_ffi / IPartOpt / cpu / Primal 0.000013 s 0.000009444800007258893 s 1.38
hlo_ffi / DefOpt / cpu / Primal 0.000013 s 0.000014216780018614372 s 0.91
hlo_ffi / IDefOpt / cpu / Primal 0.000013 s 0.000009802700005820951 s 1.33
hlo_ffi / JaXPipe / cpu / Forward 0.000019 s 0.000015186179989541416 s 1.25
hlo_ffi / Jax / cpu / Forward 0.000017999999999999997 s 0.000014690119996885189 s 1.23
hlo_ffi / HLOOpt / cpu / Forward 0.000017999999999999997 s 0.00001446758000383852 s 1.24
hlo_ffi / PartOpt / cpu / Forward 0.000017 s 0.000014529620002576847 s 1.17
hlo_ffi / IPartOpt / cpu / Forward 0.000017999999999999997 s 0.000014481360003628652 s 1.24
hlo_ffi / DefOpt / cpu / Forward 0.000017999999999999997 s 0.00001496964000580192 s 1.20
hlo_ffi / IDefOpt / cpu / Forward 0.000017999999999999997 s 0.000014544020041284968 s 1.24
hlo_ffi / JaXPipe / cpu / PreRev 0.000017 s 0.000014137960015432328 s 1.20
hlo_ffi / JaXPipe / cpu / PostRev 0.000017999999999999997 s 0.000014079540051170625 s 1.28
hlo_ffi / JaXPipe / cpu / BothRev 0.000017999999999999997 s 0.000014294059974417906 s 1.26
hlo_ffi / Jax / cpu / BothRev 0.000017 s 0.0000134094399982132 s 1.27
hlo_ffi / HLOOpt / cpu / PreRev 0.000017999999999999997 s 0.000014196640004229266 s 1.27
hlo_ffi / HLOOpt / cpu / PostRev 0.000017999999999999997 s 0.000013641719942825147 s 1.32
hlo_ffi / HLOOpt / cpu / BothRev 0.000017999999999999997 s 0.00001572217998727865 s 1.14
hlo_ffi / PartOpt / cpu / PreRev 0.000017999999999999997 s 0.000014185019972501325 s 1.27
hlo_ffi / PartOpt / cpu / PostRev 0.000017999999999999997 s 0.000013708520000363932 s 1.31
hlo_ffi / PartOpt / cpu / BothRev 0.000017999999999999997 s 0.000014153360007185256 s 1.27
hlo_ffi / IPartOpt / cpu / PreRev 0.000017 s 0.000014302419958767133 s 1.19
hlo_ffi / IPartOpt / cpu / PostRev 0.000017999999999999997 s 0.000014463979950960492 s 1.24
hlo_ffi / IPartOpt / cpu / BothRev 0.000017999999999999997 s 0.00001367498001854983 s 1.32
hlo_ffi / DefOpt / cpu / PreRev 0.000017 s 0.000014264779993027331 s 1.19
hlo_ffi / DefOpt / cpu / PostRev 0.000017999999999999997 s 0.00001391776000673417 s 1.29
hlo_ffi / DefOpt / cpu / BothRev 0.000017999999999999997 s 0.00001432760001080169 s 1.26
hlo_ffi / IDefOpt / cpu / PreRev 0.000017 s 0.000014373059966601432 s 1.18
hlo_ffi / IDefOpt / cpu / PostRev 0.000017999999999999997 s 0.000013848860035068356 s 1.30
hlo_ffi / IDefOpt / cpu / BothRev 0.000017 s 0.000014241379985833192 s 1.19
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Primal 0.0010523991999434 s 0.0011927451999326 s 0.88
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Primal 0.0009259531999305 s 0.0009631297999476 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Primal 0.0009685873998932 s 0.0009927785999025 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Primal 0.0009120263999648 s 0.0009332794001238 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Primal 0.0009178911999697 s 0.0009380363999298 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Primal 0.0009656492000431 s 0.0010069513999951 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Primal 0.0009727714000291 s 0.001006894800048 s 0.97
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Forward 0.0026339946000007 s 0.0027499933999934 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Forward 0.0023685340001065 s 0.0023157078000622 s 1.02
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Forward 0.0023205288000099 s 0.0023180373999821 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Forward 0.0022240189999138 s 0.0022757928001738 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Forward 0.0022822035999524 s 0.0023812692000319 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Forward 0.0023477289998481 s 0.0023198616000627 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Forward 0.0022953267999582 s 0.0022326147999592 s 1.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PreRev 0.0069196460000057 s 0.0069319946000177 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PostRev 0.006045536000056 s 0.0058638810000957 s 1.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / BothRev 0.0055484112001067 s 0.0058791037999071 s 0.94
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / BothRev 0.0058848736000072 s 0.0058926214000166 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PreRev 0.0056255798001075 s 0.0056403882001177 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PostRev 0.0055599140000595 s 0.0056009611998888 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / BothRev 0.0059495055999832 s 0.0056345711999711 s 1.06
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PreRev 0.0041792110000642 s 0.0058212633998664 s 0.72
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PostRev 0.0060412818000259 s 0.0066407462000825 s 0.91
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / BothRev 0.003333176399974 s 0.0056830952000382 s 0.59
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PreRev 0.0056380193999757 s 0.0060815593998995 s 0.93
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PostRev 0.0037222981999548 s 0.0064373504000286 s 0.58
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / BothRev 0.0057932471998356 s 0.0058575226001266 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PreRev 0.0041098855999734 s 0.0059112527999786 s 0.70
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PostRev 0.0055529363999085 s 0.0053786626001055 s 1.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / BothRev 0.0033771235998756 s 0.0056751533999886 s 0.60
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PreRev 0.0058969876000446 s 0.0056428122000397 s 1.05
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PostRev 0.0049572783998883 s 0.0063789011999688 s 0.78
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / BothRev 0.0053860651999457 s 0.0055820960000346 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cuda / Primal 0.000273663 s 0.000279808 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cuda / Primal 0.000272959 s 0.000279647 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cuda / Primal 0.000286495 s 0.00028688 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cuda / Primal 0.000271839 s 0.000280959 s 0.97
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cuda / Primal 0.000273822 s 0.000280799 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cuda / Primal 0.0002867509999999 s 0.000286912 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cuda / Primal 0.000285535 s 0.000286623 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cuda / Forward 0.000558301 s 0.000557374 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cuda / Forward 0.000538397 s 0.00053827 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cuda / Forward 0.000557917 s 0.0005578509999999 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cuda / Forward 0.000557949 s 0.000558527 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cuda / Forward 0.0005588769999999 s 0.000558847 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cuda / Forward 0.0005588769999999 s 0.000557535 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cuda / Forward 0.0005588769999999 s 0.000558015 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cuda / PreRev 0.001016891 s 0.001015901 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cuda / PostRev 0.000983802 s 0.000985726 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cuda / BothRev 0.001007835 s 0.001005853 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cuda / BothRev 0.000981626 s 0.000981853 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cuda / PreRev 0.001007099 s 0.001006429 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cuda / PostRev 0.001030938 s 0.001030781 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cuda / BothRev 0.001006203 s 0.00100547 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cuda / PreRev 0.001009851 s 0.001009086 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cuda / PostRev 0.00097353 s 0.000970686 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cuda / BothRev 0.001011195 s 0.0010082539999999 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cuda / PreRev 0.00101049 s 0.001007837 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cuda / PostRev 0.000971997 s 0.000974046 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cuda / BothRev 0.001009306 s 0.001010366 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cuda / PreRev 0.00102089 s 0.001018013 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cuda / PostRev 0.000955003 s 0.000953342 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cuda / BothRev 0.0010191619999999 s 0.001018141 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cuda / PreRev 0.001020443 s 0.001017918 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cuda / PostRev 0.0010158979999999 s 0.001015069 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cuda / BothRev 0.0010205379999999 s 0.001017086 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / tpu / Primal 0.000123818 s 0.0001291472499999 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / tpu / Primal 0.00012664225 s 0.0001237849999999 s 1.02
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / tpu / Primal 0.000152727 s 0.000158754 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / tpu / Primal 0.00013432625 s 0.000130933 s 1.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / tpu / Primal 0.00013101725 s 0.0001369255 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / tpu / Primal 0.00014793275 s 0.0001449385 s 1.02
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / tpu / Primal 0.000150867 s 0.00015729225 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / tpu / Forward 0.0002117545 s 0.00021352275 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / tpu / Forward 0.000260765 s 0.00026218025 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / tpu / Forward 0.00021165775 s 0.0002201595 s 0.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / tpu / Forward 0.0002187709999999 s 0.000214101 s 1.02
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / tpu / Forward 0.0002116827499999 s 0.0002159875 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / tpu / Forward 0.00021869575 s 0.000217448 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / tpu / Forward 0.00021173125 s 0.00021605975 s 0.98
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / tpu / PreRev 0.0003541014999999 s 0.00035788475 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / tpu / PostRev 0.0002583197499999 s 0.0002563965 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / tpu / BothRev 0.00035485275 s 0.0003581045 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / tpu / BothRev 0.00025905725 s 0.0002573745 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / tpu / PreRev 0.000354503 s 0.0003577665 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / tpu / PostRev 0.000293166 s 0.000291379 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / tpu / BothRev 0.00035453825 s 0.000357762 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / tpu / PreRev 0.00035680125 s 0.00035674925 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / tpu / PostRev 0.00027114775 s 0.00027458775 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / tpu / BothRev 0.00035675325 s 0.0003571505 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / tpu / PreRev 0.0003548 s 0.0003578875 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / tpu / PostRev 0.00027432425 s 0.00027272375 s 1.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / tpu / BothRev 0.0003546955 s 0.0003581575 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / tpu / PreRev 0.00035869325 s 0.00035924325 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / tpu / PostRev 0.000282215 s 0.0002841055 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / tpu / BothRev 0.00035907 s 0.00035885425 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / tpu / PreRev 0.0003573179999999 s 0.00036028175 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / tpu / PostRev 0.0002996379999999 s 0.000299094 s 1.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / tpu / BothRev 0.0003571665 s 0.0003599999999999 s 0.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Primal 0.00348749 s 0.0011927451999326 s 2.92
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Primal 0.0032351109999999 s 0.0009631297999476 s 3.36
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Primal 0.003509946 s 0.0009927785999025 s 3.54
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Primal 0.003346089 s 0.0009332794001238 s 3.59
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Primal 0.003158343 s 0.0009380363999298 s 3.37
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Primal 0.003358634 s 0.0010069513999951 s 3.34
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Primal 0.003035652 s 0.001006894800048 s 3.01
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Forward 0.007587808 s 0.0027499933999934 s 2.76
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Forward 0.007921419 s 0.0023157078000622 s 3.42
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Forward 0.007595566 s 0.0023180373999821 s 3.28
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Forward 0.007698728 s 0.0022757928001738 s 3.38
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Forward 0.0075843379999999 s 0.0023812692000319 s 3.18
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Forward 0.007572568 s 0.0023198616000627 s 3.26
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Forward 0.008326043 s 0.0022326147999592 s 3.73
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PreRev 0.014744518 s 0.0069319946000177 s 2.13
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PostRev 0.011746287 s 0.0058638810000957 s 2.00
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / BothRev 0.010904818 s 0.0058791037999071 s 1.85
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / BothRev 0.011996973 s 0.0058926214000166 s 2.04
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PreRev 0.011428331 s 0.0056403882001177 s 2.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PostRev 0.011752254 s 0.0056009611998888 s 2.10
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / BothRev 0.012220295 s 0.0056345711999711 s 2.17
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PreRev 0.011324029 s 0.0058212633998664 s 1.95
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PostRev 0.012582496 s 0.0066407462000825 s 1.89
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / BothRev 0.01227135 s 0.0056830952000382 s 2.16
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PreRev 0.0118818439999999 s 0.0060815593998995 s 1.95
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PostRev 0.011789991 s 0.0064373504000286 s 1.83
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / BothRev 0.010873374 s 0.0058575226001266 s 1.86
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PreRev 0.011962944 s 0.0059112527999786 s 2.02
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PostRev 0.012018354 s 0.0053786626001055 s 2.23
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / BothRev 0.011640449 s 0.0056751533999886 s 2.05
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PreRev 0.01172352 s 0.0056428122000397 s 2.08
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PostRev 0.011956122 s 0.0063789011999688 s 1.87
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / BothRev 0.012080653 s 0.0055820960000346 s 2.16
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Primal 0.001534 s 0.0011927451999326 s 1.29
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Primal 0.001497 s 0.0009631297999476 s 1.55
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Primal 0.0016439999999999 s 0.0009927785999025 s 1.66
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Primal 0.001453 s 0.0009332794001238 s 1.56
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Primal 0.001581 s 0.0009380363999298 s 1.69
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Primal 0.0015509999999999 s 0.0010069513999951 s 1.54
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Primal 0.001588 s 0.001006894800048 s 1.58
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / Forward 0.006023 s 0.0027499933999934 s 2.19
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / Forward 0.004061 s 0.0023157078000622 s 1.75
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / Forward 0.004404 s 0.0023180373999821 s 1.90
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / Forward 0.0049889999999999 s 0.0022757928001738 s 2.19
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / Forward 0.004248 s 0.0023812692000319 s 1.78
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / Forward 0.004003 s 0.0023198616000627 s 1.73
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Forward 0.004448 s 0.0022326147999592 s 1.99
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PreRev 0.007139 s 0.0069319946000177 s 1.03
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / PostRev 0.00832 s 0.0058638810000957 s 1.42
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / JaXPipe / cpu / BothRev 0.007358 s 0.0058791037999071 s 1.25
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / BothRev 0.010058 s 0.0058926214000166 s 1.71
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PreRev 0.007384 s 0.0056403882001177 s 1.31
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / PostRev 0.006787 s 0.0056009611998888 s 1.21
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / HLOOpt / cpu / BothRev 0.007856 s 0.0056345711999711 s 1.39
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PreRev 0.007163 s 0.0058212633998664 s 1.23
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / PostRev 0.008481 s 0.0066407462000825 s 1.28
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / PartOpt / cpu / BothRev 0.007042 s 0.0056830952000382 s 1.24
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PreRev 0.00823 s 0.0060815593998995 s 1.35
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PostRev 0.009719 s 0.0064373504000286 s 1.51
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / BothRev 0.008028 s 0.0058575226001266 s 1.37
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PreRev 0.007437 s 0.0059112527999786 s 1.26
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PostRev 0.007464 s 0.0053786626001055 s 1.39
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / BothRev 0.007453 s 0.0056751533999886 s 1.31
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PreRev 0.007264 s 0.0056428122000397 s 1.29
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PostRev 0.007386 s 0.0063789011999688 s 1.16
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / BothRev 0.007918 s 0.0055820960000346 s 1.42
scatter_sum / JaXPipe / cpu / Primal 0.000009369700019306037 s 0.000010604419985611456 s 0.88
scatter_sum / Jax / cpu / Primal 0.000008509920007782057 s 0.000009438319957553175 s 0.90
scatter_sum / HLOOpt / cpu / Primal 0.000012776759986081744 s 0.000013068099988231552 s 0.98
scatter_sum / PartOpt / cpu / Primal 0.000008707800034244429 s 0.000008753420006542001 s 0.99
scatter_sum / IPartOpt / cpu / Primal 0.000009744979988681737 s 0.00000876433997291315 s 1.11
scatter_sum / DefOpt / cpu / Primal 0.000008852820019455977 s 0.000007953300009830855 s 1.11
scatter_sum / IDefOpt / cpu / Primal 0.00000922470000659814 s 0.000008649779974803096 s 1.07
scatter_sum / JaXPipe / cpu / Forward 0.000013571119998232462 s 0.00001413226002114243 s 0.96
scatter_sum / Jax / cpu / Forward 0.000012878179950348569 s 0.000013407739998001488 s 0.96
scatter_sum / HLOOpt / cpu / Forward 0.000018949259992950828 s 0.000014226940056687452 s 1.33
scatter_sum / PartOpt / cpu / Forward 0.000019242960015617427 s 0.000018281640004715884 s 1.05
scatter_sum / IPartOpt / cpu / Forward 0.000013149200021871366 s 0.000013232420023996384 s 0.99
scatter_sum / DefOpt / cpu / Forward 0.000019200119968445503 s 0.00001845781998781604 s 1.04
scatter_sum / IDefOpt / cpu / Forward 0.00001336465996246261 s 0.000012912159972984228 s 1.04
scatter_sum / JaXPipe / cpu / PreRev 0.000014629959996455 s 0.000012809519985239603 s 1.14
scatter_sum / JaXPipe / cpu / PostRev 0.000012589459956870996 s 0.000012419260001479416 s 1.01
scatter_sum / JaXPipe / cpu / BothRev 0.000017784900001061034 s 0.000017465760029153897 s 1.02
scatter_sum / Jax / cpu / BothRev 0.000012874879976152444 s 0.000012575700047818827 s 1.02
scatter_sum / HLOOpt / cpu / PreRev 0.000012359579977783142 s 0.000012642240026252694 s 0.98
scatter_sum / HLOOpt / cpu / PostRev 0.000017716339980324845 s 0.00001703805998658936 s 1.04
scatter_sum / HLOOpt / cpu / BothRev 0.000014836919981462414 s 0.00001894454000648693 s 0.78
scatter_sum / PartOpt / cpu / PreRev 0.00001251217998287757 s 0.00001337118003903015 s 0.94
scatter_sum / PartOpt / cpu / PostRev 0.000012783620022673858 s 0.000012543000011646654 s 1.02
scatter_sum / PartOpt / cpu / BothRev 0.000012367240033199778 s 0.000012205999992147554 s 1.01
scatter_sum / IPartOpt / cpu / PreRev 0.00001838603993746801 s 0.000017159359977085842 s 1.07
scatter_sum / IPartOpt / cpu / PostRev 0.000013259900060802466 s 0.000011862860010296572 s 1.12
scatter_sum / IPartOpt / cpu / BothRev 0.000012661200007642038 s 0.000013094580026518088 s 0.97
scatter_sum / DefOpt / cpu / PreRev 0.00001327687998127658 s 0.000012335439996604692 s 1.08
scatter_sum / DefOpt / cpu / PostRev 0.000012783939964720049 s 0.00001290288002564921 s 0.99
scatter_sum / DefOpt / cpu / BothRev 0.00001304368001910916 s 0.000012867500008724163 s 1.01
scatter_sum / IDefOpt / cpu / PreRev 0.000012527079998108092 s 0.000012888519986518076 s 0.97
scatter_sum / IDefOpt / cpu / PostRev 0.000012619719982467358 s 0.000012327399981586496 s 1.02
scatter_sum / IDefOpt / cpu / BothRev 0.000013305399970704456 s 0.000012559520027934923 s 1.06
scatter_sum / JaXPipe / cuda / Primal 0.000010304 s 0.00001008 s 1.02
scatter_sum / Jax / cuda / Primal 0.000010465 s 0.000010496 s 1.00
scatter_sum / HLOOpt / cuda / Primal 0.000010112 s 0.0000104 s 0.97
scatter_sum / PartOpt / cuda / Primal 0.000010272 s 0.00001008 s 1.02
scatter_sum / IPartOpt / cuda / Primal 0.000010175 s 0.000010144 s 1.00
scatter_sum / DefOpt / cuda / Primal 0.000010048 s 0.000010048 s 1
scatter_sum / IDefOpt / cuda / Primal 0.0000104 s 0.000010239 s 1.02
scatter_sum / JaXPipe / cuda / Forward 0.000017279 s 0.000017344 s 1.00
scatter_sum / Jax / cuda / Forward 0.000017184 s 0.000017503999999999997 s 0.98
scatter_sum / HLOOpt / cuda / Forward 0.0000184 s 0.00001664 s 1.11
scatter_sum / PartOpt / cuda / Forward 0.000022303 s 0.000017536 s 1.27
scatter_sum / IPartOpt / cuda / Forward 0.000017216 s 0.000017344 s 0.99
scatter_sum / DefOpt / cuda / Forward 0.000017728 s 0.000016768000000000003 s 1.06
scatter_sum / IDefOpt / cuda / Forward 0.000017183 s 0.000017408 s 0.99
scatter_sum / JaXPipe / cuda / PreRev 0.000018784 s 0.000017152 s 1.10
scatter_sum / JaXPipe / cuda / PostRev 0.000018944 s 0.00001936 s 0.98
scatter_sum / JaXPipe / cuda / BothRev 0.0000192 s 0.000019264 s 1.00
scatter_sum / Jax / cuda / BothRev 0.000016768000000000003 s 0.000017184 s 0.98
scatter_sum / HLOOpt / cuda / PreRev 0.000018751 s 0.00001712 s 1.10
scatter_sum / HLOOpt / cuda / PostRev 0.00001712 s 0.000017088 s 1.00
scatter_sum / HLOOpt / cuda / BothRev 0.000017344 s 0.000017024 s 1.02
scatter_sum / PartOpt / cuda / PreRev 0.000018976 s 0.000019488 s 0.97
scatter_sum / PartOpt / cuda / PostRev 0.000022304 s 0.00001936 s 1.15
scatter_sum / PartOpt / cuda / BothRev 0.00001712 s 0.000019232 s 0.89
scatter_sum / IPartOpt / cuda / PreRev 0.000017728 s 0.000019743 s 0.90
scatter_sum / IPartOpt / cuda / PostRev 0.000017503999999999997 s 0.000017217 s 1.02
scatter_sum / IPartOpt / cuda / BothRev 0.000017184 s 0.000017088 s 1.01
scatter_sum / DefOpt / cuda / PreRev 0.000016736 s 0.000017472 s 0.96
scatter_sum / DefOpt / cuda / PostRev 0.000016704 s 0.000017375999999999998 s 0.96
scatter_sum / DefOpt / cuda / BothRev 0.000017312 s 0.000017344 s 1.00
scatter_sum / IDefOpt / cuda / PreRev 0.000016383999999999998 s 0.000017632 s 0.93
scatter_sum / IDefOpt / cuda / PostRev 0.000016992 s 0.000017024 s 1.00
scatter_sum / IDefOpt / cuda / BothRev 0.000016992 s 0.000017152 s 0.99
scatter_sum / JaXPipe / tpu / Primal 0.0000013442 s 0.0000013511500000000002 s 0.99
scatter_sum / Jax / tpu / Primal 0.0000014144000000000002 s 0.000001352275 s 1.05
scatter_sum / HLOOpt / tpu / Primal 0.000001353 s 0.0000013598750000000005 s 0.99
scatter_sum / PartOpt / tpu / Primal 0.0000014139250000000002 s 0.000001353125 s 1.04
scatter_sum / IPartOpt / tpu / Primal 0.0000013530249999999998 s 0.000001360825 s 0.99
scatter_sum / DefOpt / tpu / Primal 0.00000141445 s 0.000001353725 s 1.04
scatter_sum / IDefOpt / tpu / Primal 0.0000013531 s 0.0000013608000000000002 s 0.99
scatter_sum / JaXPipe / tpu / Forward 0.0000027131 s 0.0000026897 s 1.01
scatter_sum / Jax / tpu / Forward 0.0000027318 s 0.000002744725 s 1.00
scatter_sum / HLOOpt / tpu / Forward 0.0000027196 s 0.000002690275 s 1.01
scatter_sum / PartOpt / tpu / Forward 0.000002699025 s 0.0000027090500000000003 s 1.00
scatter_sum / IPartOpt / tpu / Forward 0.000002712575 s 0.000002687325 s 1.01
scatter_sum / DefOpt / tpu / Forward 0.0000027015250000000004 s 0.000002706450000000001 s 1.00
scatter_sum / IDefOpt / tpu / Forward 0.0000027195 s 0.0000026882 s 1.01
scatter_sum / JaXPipe / tpu / PreRev 0.0000026907 s 0.0000027001500000000003 s 1.00
scatter_sum / JaXPipe / tpu / PostRev 0.0000027012000000000005 s 0.0000026967000000000004 s 1.00
scatter_sum / JaXPipe / tpu / BothRev 0.000002720875 s 0.000002717675 s 1.00
scatter_sum / Jax / tpu / BothRev 0.0000027482250000000003 s 0.00000274035 s 1.00
scatter_sum / HLOOpt / tpu / PreRev 0.0000027101750000000003 s 0.000002710475 s 1.00
scatter_sum / HLOOpt / tpu / PostRev 0.0000027536 s 0.0000027447749999999995 s 1.00
scatter_sum / HLOOpt / tpu / BothRev 0.0000027069 s 0.000002715025 s 1.00
scatter_sum / PartOpt / tpu / PreRev 0.0000027529 s 0.0000027438249999999995 s 1.00
scatter_sum / PartOpt / tpu / PostRev 0.0000027038250000000003 s 0.00000271565 s 1.00
scatter_sum / PartOpt / tpu / BothRev 0.000002757175 s 0.0000027421 s 1.01
scatter_sum / IPartOpt / tpu / PreRev 0.0000027066 s 0.000002718275 s 1.00
scatter_sum / IPartOpt / tpu / PostRev 0.000002750625 s 0.000002748325 s 1.00
scatter_sum / IPartOpt / tpu / BothRev 0.000002710275 s 0.000002719525 s 1.00
scatter_sum / DefOpt / tpu / PreRev 0.00000274915 s 0.00000274415 s 1.00
scatter_sum / DefOpt / tpu / PostRev 0.000002706175 s 0.000002721425 s 0.99
scatter_sum / DefOpt / tpu / BothRev 0.0000027518 s 0.000002747275 s 1.00
scatter_sum / IDefOpt / tpu / PreRev 0.000002705875 s 0.0000027163749999999995 s 1.00
scatter_sum / IDefOpt / tpu / PostRev 0.00000274995 s 0.00000274505 s 1.00
scatter_sum / IDefOpt / tpu / BothRev 0.000002710075 s 0.0000027152500000000004 s 1.00
scatter_sum / JaXPipe / cpu / Primal 0.00001996 s 0.000010604419985611456 s 1.88
scatter_sum / Jax / cpu / Primal 0.0000193 s 0.000009438319957553175 s 2.04
scatter_sum / HLOOpt / cpu / Primal 0.000019824 s 0.000013068099988231552 s 1.52
scatter_sum / PartOpt / cpu / Primal 0.000019425 s 0.000008753420006542001 s 2.22
scatter_sum / IPartOpt / cpu / Primal 0.000019823 s 0.00000876433997291315 s 2.26
scatter_sum / DefOpt / cpu / Primal 0.000019774 s 0.000007953300009830855 s 2.49
scatter_sum / IDefOpt / cpu / Primal 0.000020515 s 0.000008649779974803096 s 2.37
scatter_sum / JaXPipe / cpu / Forward 0.000029333 s 0.00001413226002114243 s 2.08
scatter_sum / Jax / cpu / Forward 0.00002836 s 0.000013407739998001488 s 2.12
scatter_sum / HLOOpt / cpu / Forward 0.000028308 s 0.000014226940056687452 s 1.99
scatter_sum / PartOpt / cpu / Forward 0.000027503 s 0.000018281640004715884 s 1.50
scatter_sum / IPartOpt / cpu / Forward 0.000028469 s 0.000013232420023996384 s 2.15
scatter_sum / DefOpt / cpu / Forward 0.000028258 s 0.00001845781998781604 s 1.53
scatter_sum / IDefOpt / cpu / Forward 0.000027938000000000003 s 0.000012912159972984228 s 2.16
scatter_sum / JaXPipe / cpu / PreRev 0.00002824 s 0.000012809519985239603 s 2.20
scatter_sum / JaXPipe / cpu / PostRev 0.000028505 s 0.000012419260001479416 s 2.30
scatter_sum / JaXPipe / cpu / BothRev 0.000028579 s 0.000017465760029153897 s 1.64
scatter_sum / Jax / cpu / BothRev 0.000028944 s 0.000012575700047818827 s 2.30
scatter_sum / HLOOpt / cpu / PreRev 0.000027757 s 0.000012642240026252694 s 2.20
scatter_sum / HLOOpt / cpu / PostRev 0.00002928 s 0.00001703805998658936 s 1.72
scatter_sum / HLOOpt / cpu / BothRev 0.000028927000000000003 s 0.00001894454000648693 s 1.53
scatter_sum / PartOpt / cpu / PreRev 0.000028475 s 0.00001337118003903015 s 2.13
scatter_sum / PartOpt / cpu / PostRev 0.000029249 s 0.000012543000011646654 s 2.33
scatter_sum / PartOpt / cpu / BothRev 0.000029237 s 0.000012205999992147554 s 2.40
scatter_sum / IPartOpt / cpu / PreRev 0.000028607 s 0.000017159359977085842 s 1.67
scatter_sum / IPartOpt / cpu / PostRev 0.00002796 s 0.000011862860010296572 s 2.36
scatter_sum / IPartOpt / cpu / BothRev 0.000028455 s 0.000013094580026518088 s 2.17
scatter_sum / DefOpt / cpu / PreRev 0.000027885 s 0.000012335439996604692 s 2.26
scatter_sum / DefOpt / cpu / PostRev 0.000028617 s 0.00001290288002564921 s 2.22
scatter_sum / DefOpt / cpu / BothRev 0.000028071 s 0.000012867500008724163 s 2.18
scatter_sum / IDefOpt / cpu / PreRev 0.000028485 s 0.000012888519986518076 s 2.21
scatter_sum / IDefOpt / cpu / PostRev 0.000029669 s 0.000012327399981586496 s 2.41
scatter_sum / IDefOpt / cpu / BothRev 0.000028287 s 0.000012559520027934923 s 2.25
scatter_sum / JaXPipe / cpu / Primal 0.000011 s 0.000010604419985611456 s 1.04
scatter_sum / Jax / cpu / Primal 0.000011 s 0.000009438319957553175 s 1.17
scatter_sum / HLOOpt / cpu / Primal 0.00001 s 0.000013068099988231552 s 0.77
scatter_sum / PartOpt / cpu / Primal 0.00001 s 0.000008753420006542001 s 1.14
scatter_sum / IPartOpt / cpu / Primal 0.00001 s 0.00000876433997291315 s 1.14
scatter_sum / DefOpt / cpu / Primal 0.00001 s 0.000007953300009830855 s 1.26
scatter_sum / IDefOpt / cpu / Primal 0.00001 s 0.000008649779974803096 s 1.16
scatter_sum / JaXPipe / cpu / Forward 0.000016 s 0.00001413226002114243 s 1.13
scatter_sum / Jax / cpu / Forward 0.000016 s 0.000013407739998001488 s 1.19
scatter_sum / HLOOpt / cpu / Forward 0.000016 s 0.000014226940056687452 s 1.12
scatter_sum / PartOpt / cpu / Forward 0.000015 s 0.000018281640004715884 s 0.82
scatter_sum / IPartOpt / cpu / Forward 0.000015 s 0.000013232420023996384 s 1.13
scatter_sum / DefOpt / cpu / Forward 0.000016 s 0.00001845781998781604 s 0.87
scatter_sum / IDefOpt / cpu / Forward 0.000016 s 0.000012912159972984228 s 1.24
scatter_sum / JaXPipe / cpu / PreRev 0.000016 s 0.000012809519985239603 s 1.25
scatter_sum / JaXPipe / cpu / PostRev 0.000016 s 0.000012419260001479416 s 1.29
scatter_sum / JaXPipe / cpu / BothRev 0.000016 s 0.000017465760029153897 s 0.92
scatter_sum / Jax / cpu / BothRev 0.000016 s 0.000012575700047818827 s 1.27
scatter_sum / HLOOpt / cpu / PreRev 0.000016 s 0.000012642240026252694 s 1.27
scatter_sum / HLOOpt / cpu / PostRev 0.000016 s 0.00001703805998658936 s 0.94
scatter_sum / HLOOpt / cpu / BothRev 0.000017 s 0.00001894454000648693 s 0.90
scatter_sum / PartOpt / cpu / PreRev 0.000016 s 0.00001337118003903015 s 1.20
scatter_sum / PartOpt / cpu / PostRev 0.000016 s 0.000012543000011646654 s 1.28
scatter_sum / PartOpt / cpu / BothRev 0.000016 s 0.000012205999992147554 s 1.31
scatter_sum / IPartOpt / cpu / PreRev 0.000016 s 0.000017159359977085842 s 0.93
scatter_sum / IPartOpt / cpu / PostRev 0.000016 s 0.000011862860010296572 s 1.35
scatter_sum / IPartOpt / cpu / BothRev 0.000016 s 0.000013094580026518088 s 1.22
scatter_sum / DefOpt / cpu / PreRev 0.000016 s 0.000012335439996604692 s 1.30
scatter_sum / DefOpt / cpu / PostRev 0.000016 s 0.00001290288002564921 s 1.24
scatter_sum / DefOpt / cpu / BothRev 0.000015 s 0.000012867500008724163 s 1.17
scatter_sum / IDefOpt / cpu / PreRev 0.000015 s 0.000012888519986518076 s 1.16
scatter_sum / IDefOpt / cpu / PostRev 0.000016 s 0.000012327399981586496 s 1.30
scatter_sum / IDefOpt / cpu / BothRev 0.000015 s 0.000012559520027934923 s 1.19
slicing / JaXPipe / cpu / Primal 0.000007709380006417633 s 0.000007591579978907248 s 1.02
slicing / Jax / cpu / Primal 0.000007044280027912464 s 0.000006696759974147426 s 1.05
slicing / HLOOpt / cpu / Primal 0.000010941520022242912 s 0.000010924839925792184 s 1.00
slicing / PartOpt / cpu / Primal 0.000007226000025184476 s 0.000006778399983886629 s 1.07
slicing / IPartOpt / cpu / Primal 0.000006458899961216958 s 0.0000066117999904236055 s 0.98
slicing / DefOpt / cpu / Primal 0.000010305679998054984 s 0.00001195267998809868 s 0.86
slicing / IDefOpt / cpu / Primal 0.000006528700032504275 s 0.000006800619985369849 s 0.96
slicing / JaXPipe / cpu / Forward 0.00001075037998816697 s 0.000010853660014618072 s 0.99
slicing / Jax / cpu / Forward 0.000011494580003272858 s 0.00001074058001904632 s 1.07
slicing / HLOOpt / cpu / Forward 0.00001556375999825832 s 0.000014565760020559535 s 1.07
slicing / PartOpt / cpu / Forward 0.000015067179947436671 s 0.000014514440063067013 s 1.04
slicing / IPartOpt / cpu / Forward 0.000011096719972556456 s 0.000009968380063583026 s 1.11
slicing / DefOpt / cpu / Forward 0.000015668920013922616 s 0.000014301899982456234 s 1.10
slicing / IDefOpt / cpu / Forward 0.000010880819954763866 s 0.00001044116000230133 s 1.04
slicing / JaXPipe / cpu / PreRev 0.000011117639987787695 s 0.000012225100008436127 s 0.91
slicing / JaXPipe / cpu / PostRev 0.00001206911996632698 s 0.000011756279991459451 s 1.03
slicing / JaXPipe / cpu / BothRev 0.00001554814000883198 s 0.000011134380010844324 s 1.40
slicing / Jax / cpu / BothRev 0.000010870640007851762 s 0.000011381400008758648 s 0.96
slicing / HLOOpt / cpu / PreRev 0.000010555519993431518 s 0.000011067560026276623 s 0.95
slicing / HLOOpt / cpu / PostRev 0.000011335219996908564 s 0.000011994699962087909 s 0.95
slicing / HLOOpt / cpu / BothRev 0.000012319360002948088 s 0.000012893240018456707 s 0.96
slicing / PartOpt / cpu / PreRev 0.000011506379978527548 s 0.00001100773999496596 s 1.05
slicing / PartOpt / cpu / PostRev 0.000011483719999887398 s 0.00001192775996059936 s 0.96
slicing / PartOpt / cpu / BothRev 0.000011138479976580127 s 0.00001121483997849282 s 0.99
slicing / IPartOpt / cpu / PreRev 0.00001589297999998962 s 0.00001651082003263582 s 0.96
slicing / IPartOpt / cpu / PostRev 0.000011990760012849931 s 0.000012523619998319191 s 0.96
slicing / IPartOpt / cpu / BothRev 0.000010985339986291364 s 0.000010976600015055735 s 1.00
slicing / DefOpt / cpu / PreRev 0.000011482039999464178 s 0.000011150139980600216 s 1.03
slicing / DefOpt / cpu / PostRev 0.000011204140037079924 s 0.000012347219990260782 s 0.91
slicing / DefOpt / cpu / BothRev 0.000011081779975938845 s 0.000011590060003072722 s 0.96
slicing / IDefOpt / cpu / PreRev 0.000010581019978417316 s 0.000011069459987993467 s 0.96
slicing / IDefOpt / cpu / PostRev 0.00001178187997538771 s 0.00001156921998699545 s 1.02
slicing / IDefOpt / cpu / BothRev 0.000010775259997899411 s 0.00001105164001273806 s 0.97
slicing / JaXPipe / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / Jax / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / HLOOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / PartOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / IPartOpt / cuda / Primal 0.000001887 s 0.0000019200000000000003 s 0.98
slicing / DefOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / IDefOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / JaXPipe / cuda / Forward 0.000010175 s 0.000009953 s 1.02
slicing / Jax / cuda / Forward 0.000009856 s 0.000010048 s 0.98
slicing / HLOOpt / cuda / Forward 0.000009537 s 0.000010304 s 0.93
slicing / PartOpt / cuda / Forward 0.000010304 s 0.000010143 s 1.02
slicing / IPartOpt / cuda / Forward 0.000010143 s 0.000009344 s 1.09
slicing / DefOpt / cuda / Forward 0.000010047 s 0.000010143 s 0.99
slicing / IDefOpt / cuda / Forward 0.000009536 s 0.000009983 s 0.96
slicing / JaXPipe / cuda / PreRev 0.000010368 s 0.00001024 s 1.01
slicing / JaXPipe / cuda / PostRev 0.000010112 s 0.000009696 s 1.04
slicing / JaXPipe / cuda / BothRev 0.00001024 s 0.000010144 s 1.01
slicing / Jax / cuda / BothRev 0.00000976 s 0.000009792 s 1.00
slicing / HLOOpt / cuda / PreRev 0.000009984 s 0.000009888 s 1.01
slicing / HLOOpt / cuda / PostRev 0.00001008 s 0.00001024 s 0.98
slicing / HLOOpt / cuda / BothRev 0.000010304 s 0.000010016 s 1.03
slicing / PartOpt / cuda / PreRev 0.000010303 s 0.000010176 s 1.01
slicing / PartOpt / cuda / PostRev 0.000010208 s 0.000010144 s 1.01
slicing / PartOpt / cuda / BothRev 0.000010592 s 0.000010272 s 1.03
slicing / IPartOpt / cuda / PreRev 0.000010271 s 0.000010112 s 1.02
slicing / IPartOpt / cuda / PostRev 0.000010144 s 0.00000944 s 1.07
slicing / IPartOpt / cuda / BothRev 0.000010368 s 0.000009824 s 1.06
slicing / DefOpt / cuda / PreRev 0.000012576 s 0.000012769 s 0.98
slicing / DefOpt / cuda / PostRev 0.000010016 s 0.000009632 s 1.04
slicing / DefOpt / cuda / BothRev 0.000009824 s 0.000012224 s 0.80
slicing / IDefOpt / cuda / PreRev 0.00001008 s 0.000011167 s 0.90
slicing / IDefOpt / cuda / PostRev 0.000010016 s 0.000009664 s 1.04
slicing / IDefOpt / cuda / BothRev 0.000010304 s 0.000009985 s 1.03
slicing / JaXPipe / tpu / Primal 9.80725e-7 s 9.985e-7 s 0.98
slicing / Jax / tpu / Primal 9.716e-7 s 9.60375e-7 s 1.01
slicing / HLOOpt / tpu / Primal 9.6695e-7 s 0.0000010006750000000002 s 0.97
slicing / PartOpt / tpu / Primal 9.636e-7 s 9.628499999999998e-7 s 1.00
slicing / IPartOpt / tpu / Primal 9.6355e-7 s 0.000001004575 s 0.96
slicing / DefOpt / tpu / Primal 9.70625e-7 s 9.6015e-7 s 1.01
slicing / IDefOpt / tpu / Primal 9.75475e-7 s 9.96375e-7 s 0.98
slicing / JaXPipe / tpu / Forward 0.0000014026250000000002 s 0.000001407325 s 1.00
slicing / Jax / tpu / Forward 0.0000014192 s 0.000001449325 s 0.98
slicing / HLOOpt / tpu / Forward 0.000001517975 s 0.0000015219750000000002 s 1.00
slicing / PartOpt / tpu / Forward 0.0000014394 s 0.0000014721500000000002 s 0.98
slicing / IPartOpt / tpu / Forward 0.0000015131000000000002 s 0.000001523 s 0.99
slicing / DefOpt / tpu / Forward 0.0000014425 s 0.000001473475 s 0.98
slicing / IDefOpt / tpu / Forward 0.0000015126249999999998 s 0.000001516075 s 1.00
slicing / JaXPipe / tpu / PreRev 0.0000023871 s 0.000002475775 s 0.96
slicing / JaXPipe / tpu / PostRev 0.000002523075 s 0.000002532425 s 1.00
slicing / JaXPipe / tpu / BothRev 0.000002405325 s 0.000002512625 s 0.96
slicing / Jax / tpu / BothRev 0.0000025396 s 0.000002542725 s 1.00
slicing / HLOOpt / tpu / PreRev 0.00000239435 s 0.000002502275 s 0.96
slicing / HLOOpt / tpu / PostRev 0.000002533725 s 0.000002536025 s 1.00
slicing / HLOOpt / tpu / BothRev 0.00000239425 s 0.000002515725 s 0.95
slicing / PartOpt / tpu / PreRev 0.0000025467 s 0.00000252235 s 1.01
slicing / PartOpt / tpu / PostRev 0.0000023905 s 0.000002490925 s 0.96
slicing / PartOpt / tpu / BothRev 0.0000025455250000000005 s 0.0000025242750000000005 s 1.01
slicing / IPartOpt / tpu / PreRev 0.0000024063 s 0.00000251575 s 0.96
slicing / IPartOpt / tpu / PostRev 0.000002540025 s 0.0000025392749999999995 s 1.00
slicing / IPartOpt / tpu / BothRev 0.00000240325 s 0.00000249935 s 0.96
slicing / DefOpt / tpu / PreRev 0.00000253205 s 0.0000025303500000000003 s 1.00
slicing / DefOpt / tpu / PostRev 0.000002407975 s 0.000002497675 s 0.96
slicing / DefOpt / tpu / BothRev 0.0000025409749999999994 s 0.00000253695 s 1.00
slicing / IDefOpt / tpu / PreRev 0.0000024002250000000004 s 0.0000024994 s 0.96
slicing / IDefOpt / tpu / PostRev 0.00000253555 s 0.00000253355 s 1.00
slicing / IDefOpt / tpu / BothRev 0.0000023984 s 0.000002510625 s 0.96
slicing / JaXPipe / cpu / Primal 0.000015802 s 0.000007591579978907248 s 2.08
slicing / Jax / cpu / Primal 0.000015619 s 0.000006696759974147426 s 2.33
slicing / HLOOpt / cpu / Primal 0.000015847 s 0.000010924839925792184 s 1.45
slicing / PartOpt / cpu / Primal 0.00001594 s 0.000006778399983886629 s 2.35
slicing / IPartOpt / cpu / Primal 0.000015845 s 0.0000066117999904236055 s 2.40
slicing / DefOpt / cpu / Primal 0.000015803 s 0.00001195267998809868 s 1.32
slicing / IDefOpt / cpu / Primal 0.000015762999999999998 s 0.000006800619985369849 s 2.32
slicing / JaXPipe / cpu / Forward 0.000021477 s 0.000010853660014618072 s 1.98
slicing / Jax / cpu / Forward 0.000021055 s 0.00001074058001904632 s 1.96
slicing / HLOOpt / cpu / Forward 0.000020731 s 0.000014565760020559535 s 1.42
slicing / PartOpt / cpu / Forward 0.000021219 s 0.000014514440063067013 s 1.46
slicing / IPartOpt / cpu / Forward 0.000021013 s 0.000009968380063583026 s 2.11
slicing / DefOpt / cpu / Forward 0.000021441 s 0.000014301899982456234 s 1.50
slicing / IDefOpt / cpu / Forward 0.00002151 s 0.00001044116000230133 s 2.06
slicing / JaXPipe / cpu / PreRev 0.00002223 s 0.000012225100008436127 s 1.82
slicing / JaXPipe / cpu / PostRev 0.00002163 s 0.000011756279991459451 s 1.84
slicing / JaXPipe / cpu / BothRev 0.000022101 s 0.000011134380010844324 s 1.98
slicing / Jax / cpu / BothRev 0.000022195 s 0.000011381400008758648 s 1.95
slicing / HLOOpt / cpu / PreRev 0.000021913 s 0.000011067560026276623 s 1.98
slicing / HLOOpt / cpu / PostRev 0.000021891 s 0.000011994699962087909 s 1.83
slicing / HLOOpt / cpu / BothRev 0.00002251 s 0.000012893240018456707 s 1.75
slicing / PartOpt / cpu / PreRev 0.000022287 s 0.00001100773999496596 s 2.02
slicing / PartOpt / cpu / PostRev 0.000022211 s 0.00001192775996059936 s 1.86
slicing / PartOpt / cpu / BothRev 0.000022276 s 0.00001121483997849282 s 1.99
slicing / IPartOpt / cpu / PreRev 0.000022062 s 0.00001651082003263582 s 1.34
slicing / IPartOpt / cpu / PostRev 0.000022190000000000003 s 0.000012523619998319191 s 1.77
slicing / IPartOpt / cpu / BothRev 0.000022819 s 0.000010976600015055735 s 2.08
slicing / DefOpt / cpu / PreRev 0.000021716 s 0.000011150139980600216 s 1.95
slicing / DefOpt / cpu / PostRev 0.00002158 s 0.000012347219990260782 s 1.75
slicing / DefOpt / cpu / BothRev 0.000022827 s 0.000011590060003072722 s 1.97
slicing / IDefOpt / cpu / PreRev 0.000021643 s 0.000011069459987993467 s 1.96
slicing / IDefOpt / cpu / PostRev 0.000022544000000000003 s 0.00001156921998699545 s 1.95
slicing / IDefOpt / cpu / BothRev 0.000022473 s 0.00001105164001273806 s 2.03
slicing / JaXPipe / cpu / Primal 0.000008 s 0.000007591579978907248 s 1.05
slicing / Jax / cpu / Primal 0.000008 s 0.000006696759974147426 s 1.19
slicing / HLOOpt / cpu / Primal 0.000008 s 0.000010924839925792184 s 0.73
slicing / PartOpt / cpu / Primal 0.000008 s 0.000006778399983886629 s 1.18
slicing / IPartOpt / cpu / Primal 0.000008 s 0.0000066117999904236055 s 1.21
slicing / DefOpt / cpu / Primal 0.000008 s 0.00001195267998809868 s 0.67
slicing / IDefOpt / cpu / Primal 0.000008 s 0.000006800619985369849 s 1.18
slicing / JaXPipe / cpu / Forward 0.000011 s 0.000010853660014618072 s 1.01
slicing / Jax / cpu / Forward 0.000011 s 0.00001074058001904632 s 1.02
slicing / HLOOpt / cpu / Forward 0.000011 s 0.000014565760020559535 s 0.76
slicing / PartOpt / cpu / Forward 0.000011 s 0.000014514440063067013 s 0.76
slicing / IPartOpt / cpu / Forward 0.000012 s 0.000009968380063583026 s 1.20
slicing / DefOpt / cpu / Forward 0.000011 s 0.000014301899982456234 s 0.77
slicing / IDefOpt / cpu / Forward 0.000011 s 0.00001044116000230133 s 1.05
slicing / JaXPipe / cpu / PreRev 0.000011 s 0.000012225100008436127 s 0.90
slicing / JaXPipe / cpu / PostRev 0.000011 s 0.000011756279991459451 s 0.94
slicing / JaXPipe / cpu / BothRev 0.000012 s 0.000011134380010844324 s 1.08
slicing / Jax / cpu / BothRev 0.000011 s 0.000011381400008758648 s 0.97
slicing / HLOOpt / cpu / PreRev 0.000011 s 0.000011067560026276623 s 0.99
slicing / HLOOpt / cpu / PostRev 0.000011 s 0.000011994699962087909 s 0.92
slicing / HLOOpt / cpu / BothRev 0.000011 s 0.000012893240018456707 s 0.85
slicing / PartOpt / cpu / PreRev 0.000011 s 0.00001100773999496596 s 1.00
slicing / PartOpt / cpu / PostRev 0.000012 s 0.00001192775996059936 s 1.01
slicing / PartOpt / cpu / BothRev 0.000012 s 0.00001121483997849282 s 1.07
slicing / IPartOpt / cpu / PreRev 0.000011 s 0.00001651082003263582 s 0.67
slicing / IPartOpt / cpu / PostRev 0.000012 s 0.000012523619998319191 s 0.96
slicing / IPartOpt / cpu / BothRev 0.000012 s 0.000010976600015055735 s 1.09
slicing / DefOpt / cpu / PreRev 0.000011 s 0.000011150139980600216 s 0.99
slicing / DefOpt / cpu / PostRev 0.000011 s 0.000012347219990260782 s 0.89
slicing / DefOpt / cpu / BothRev 0.000011 s 0.000011590060003072722 s 0.95
slicing / IDefOpt / cpu / PreRev 0.000012 s 0.000011069459987993467 s 1.08
slicing / IDefOpt / cpu / PostRev 0.000012 s 0.00001156921998699545 s 1.04
slicing / IDefOpt / cpu / BothRev 0.000012 s 0.00001105164001273806 s 1.09
sum / JaXPipe / cpu / Primal 0.00000897073999112763 s 0.000008944359979068395 s 1.00
sum / Jax / cpu / Primal 0.00000796704001913895 s 0.000008394739988943912 s 0.95
sum / HLOOpt / cpu / Primal 0.000012563559994305252 s 0.000012469859966586229 s 1.01
sum / PartOpt / cpu / Primal 0.000008638700010124011 s 0.000008866120006132404 s 0.97
sum / IPartOpt / cpu / Primal 0.000008357160004379693 s 0.000009332300041933197 s 0.90
sum / DefOpt / cpu / Primal 0.000013028680004936176 s 0.000012841500001741224 s 1.01
sum / IDefOpt / cpu / Primal 0.000008801860012681572 s 0.000008591099995101103 s 1.02
sum / JaXPipe / cpu / Forward 0.000012453900008040363 s 0.000012326300020504277 s 1.01
sum / Jax / cpu / Forward 0.000012981099989701762 s 0.000013074359976599226 s 0.99
sum / HLOOpt / cpu / Forward 0.000017418239931430435 s 0.000017311880010311143 s 1.01
sum / PartOpt / cpu / Forward 0.000016878399983397684 s 0.0000170316599724174 s 0.99
sum / IPartOpt / cpu / Forward 0.000012651920005737338 s 0.000012417759990057675 s 1.02
sum / DefOpt / cpu / Forward 0.00001764980003827077 s 0.000017328099993392243 s 1.02
sum / IDefOpt / cpu / Forward 0.000012428859999999986 s 0.000012513379988376983 s 0.99
sum / JaXPipe / cpu / PreRev 0.000012619739964065955 s 0.000012807400034944294 s 0.99
sum / JaXPipe / cpu / PostRev 0.000013014619999012213 s 0.00001236651999533933 s 1.05
sum / JaXPipe / cpu / BothRev 0.000012923299973408577 s 0.000016243300015048588 s 0.80
sum / Jax / cpu / BothRev 0.00001244928003870882 s 0.000012590739979714271 s 0.99
sum / HLOOpt / cpu / PreRev 0.000011871919987243018 s 0.000012084299987691338 s 0.98
sum / HLOOpt / cpu / PostRev 0.000012574719985423144 s 0.000015077259995450733 s 0.83
sum / HLOOpt / cpu / BothRev 0.000018325920009374383 s 0.000013451559989334782 s 1.36
sum / PartOpt / cpu / PreRev 0.00001172238002254744 s 0.00001157794003120216 s 1.01
sum / PartOpt / cpu / PostRev 0.000012561680014187005 s 0.000012062860005244149 s 1.04
sum / PartOpt / cpu / BothRev 0.000012099200021111757 s 0.00001210502000503766 s 1.00
sum / IPartOpt / cpu / PreRev 0.000013820399963151432 s 0.000012810119978894362 s 1.08
sum / IPartOpt / cpu / PostRev 0.000012502099971243296 s 0.00001243552001142234 s 1.01
sum / IPartOpt / cpu / BothRev 0.000011970880013905116 s 0.000011493500041979132 s 1.04
sum / DefOpt / cpu / PreRev 0.000011762160011130616 s 0.000011542479978743358 s 1.02
sum / DefOpt / cpu / PostRev 0.00001287915998545941 s 0.00001140446000135853 s 1.13
sum / DefOpt / cpu / BothRev 0.000012576319986692396 s 0.000011926980014322908 s 1.05
sum / IDefOpt / cpu / PreRev 0.000012363359983282864 s 0.000011830739940705826 s 1.05
sum / IDefOpt / cpu / PostRev 0.000012434499976734517 s 0.00001181827999971574 s 1.05
sum / IDefOpt / cpu / BothRev 0.00001182321998385305 s 0.000011645999948086682 s 1.02
sum / JaXPipe / cuda / Primal 0.000002047 s 0.00000208 s 0.98
sum / Jax / cuda / Primal 0.000002047 s 0.00000208 s 0.98
sum / HLOOpt / cuda / Primal 0.000002048 s 0.00000208 s 0.98
sum / PartOpt / cuda / Primal 0.000002048 s 0.000002079 s 0.99
sum / IPartOpt / cuda / Primal 0.000002047 s 0.000002079 s 0.98
sum / DefOpt / cuda / Primal 0.000002048 s 0.000002079 s 0.99
sum / IDefOpt / cuda / Primal 0.000002047 s 0.000002079 s 0.98
sum / JaXPipe / cuda / Forward 0.000010304 s 0.00001168 s 0.88
sum / Jax / cuda / Forward 0.000010336 s 0.00001072 s 0.96
sum / HLOOpt / cuda / Forward 0.00001024 s 0.00001088 s 0.94
sum / PartOpt / cuda / Forward 0.00001056 s 0.000011072 s 0.95
sum / IPartOpt / cuda / Forward 0.000010431 s 0.000009984 s 1.04
sum / DefOpt / cuda / Forward 0.0000104 s 0.000010335 s 1.01
sum / IDefOpt / cuda / Forward 0.000010304 s 0.000009664 s 1.07
sum / JaXPipe / cuda / PreRev 0.00000992 s 0.000009536 s 1.04
sum / JaXPipe / cuda / PostRev 0.000009984 s 0.000009632 s 1.04
sum / JaXPipe / cuda / BothRev 0.000011936 s 0.000009791 s 1.22
sum / Jax / cuda / BothRev 0.00001008 s 0.000010047 s 1.00
sum / HLOOpt / cuda / PreRev 0.000010144 s 0.000009984 s 1.02
sum / HLOOpt / cuda / PostRev 0.000010368 s 0.0000096 s 1.08
sum / HLOOpt / cuda / BothRev 0.000009952 s 0.000009792 s 1.02
sum / PartOpt / cuda / PreRev 0.000010976 s 0.00001008 s 1.09
sum / PartOpt / cuda / PostRev 0.000009888 s 0.000009984 s 0.99
sum / PartOpt / cuda / BothRev 0.00000992 s 0.000010016 s 0.99
sum / IPartOpt / cuda / PreRev 0.000010112 s 0.000009536 s 1.06
sum / IPartOpt / cuda / PostRev 0.000009889 s 0.000009792 s 1.01
sum / IPartOpt / cuda / BothRev 0.000010143 s 0.000012224 s 0.83
sum / DefOpt / cuda / PreRev 0.000010016 s 0.00001136 s 0.88
sum / DefOpt / cuda / PostRev 0.000012031 s 0.000011584 s 1.04
sum / DefOpt / cuda / BothRev 0.000010208 s 0.000009953 s 1.03
sum / IDefOpt / cuda / PreRev 0.00001008 s 0.000010208 s 0.99
sum / IDefOpt / cuda / PostRev 0.00001008 s 0.000010304 s 0.98
sum / IDefOpt / cuda / BothRev 0.000009984 s 0.000010048 s 0.99
sum / JaXPipe / tpu / Primal 5.10625e-7 s 5.03125e-7 s 1.01
sum / Jax / tpu / Primal 5.612e-7 s 5.5775e-7 s 1.01
sum / HLOOpt / tpu / Primal 5.204250000000001e-7 s 5.135750000000001e-7 s 1.01
sum / PartOpt / tpu / Primal 5.5695e-7 s 5.576e-7 s 1.00
sum / IPartOpt / tpu / Primal 5.2005e-7 s 5.135750000000001e-7 s 1.01
sum / DefOpt / tpu / Primal 5.57075e-7 s 5.5785e-7 s 1.00
sum / IDefOpt / tpu / Primal 5.2025e-7 s 5.134e-7 s 1.01
sum / JaXPipe / tpu / Forward 0.000001550925 s 0.0000015539 s 1.00
sum / Jax / tpu / Forward 0.00000150455 s 0.0000014981 s 1.00
sum / HLOOpt / tpu / Forward 0.00000153445 s 0.0000015324499999999998 s 1.00
sum / PartOpt / tpu / Forward 0.00000149365 s 0.00000149055 s 1.00
sum / IPartOpt / tpu / Forward 0.0000015279999999999998 s 0.00000153895 s 0.99
sum / DefOpt / tpu / Forward 0.000001493275 s 0.0000014910750000000002 s 1.00
sum / IDefOpt / tpu / Forward 0.0000015354499999999998 s 0.000001529975 s 1.00
sum / JaXPipe / tpu / PreRev 0.000001001125 s 0.00000101955 s 0.98
sum / JaXPipe / tpu / PostRev 0.0000010366499999999998 s 0.0000010621 s 0.98
sum / JaXPipe / tpu / BothRev 0.00000101525 s 0.000001022175 s 0.99
sum / Jax / tpu / BothRev 0.0000010427 s 0.000001060925 s 0.98
sum / HLOOpt / tpu / PreRev 0.000001000825 s 0.000001021125 s 0.98
sum / HLOOpt / tpu / PostRev 0.0000010356 s 0.0000010625750000000002 s 0.97
sum / HLOOpt / tpu / BothRev 9.99575e-7 s 0.000001014425 s 0.99
sum / PartOpt / tpu / PreRev 0.00000104805 s 0.0000010603750000000002 s 0.99
sum / PartOpt / tpu / PostRev 0.0000010041500000000002 s 0.000001019175 s 0.99
sum / PartOpt / tpu / BothRev 0.000001046625 s 0.000001062275 s 0.99
sum / IPartOpt / tpu / PreRev 0.0000010024 s 0.000001017325 s 0.99
sum / IPartOpt / tpu / PostRev 0.00000103595 s 0.0000010679 s 0.97
sum / IPartOpt / tpu / BothRev 0.000001000875 s 0.00000101495 s 0.99
sum / DefOpt / tpu / PreRev 0.000001038975 s 0.000001064175 s 0.98
sum / DefOpt / tpu / PostRev 0.000001010325 s 0.000001021825 s 0.99
sum / DefOpt / tpu / BothRev 0.00000104785 s 0.0000010673499999999998 s 0.98
sum / IDefOpt / tpu / PreRev 0.0000010046 s 0.000001019725 s 0.99
sum / IDefOpt / tpu / PostRev 0.000001037675 s 0.0000010642 s 0.98
sum / IDefOpt / tpu / BothRev 0.000001004825 s 0.000001020925 s 0.98
sum / JaXPipe / cpu / Primal 0.000018824 s 0.000008944359979068395 s 2.10
sum / Jax / cpu / Primal 0.000018348 s 0.000008394739988943912 s 2.19
sum / HLOOpt / cpu / Primal 0.00001838 s 0.000012469859966586229 s 1.47
sum / PartOpt / cpu / Primal 0.000018087 s 0.000008866120006132404 s 2.04
sum / IPartOpt / cpu / Primal 0.00001825 s 0.000009332300041933197 s 1.96
sum / DefOpt / cpu / Primal 0.000017683 s 0.000012841500001741224 s 1.38
sum / IDefOpt / cpu / Primal 0.000018143 s 0.000008591099995101103 s 2.11
sum / JaXPipe / cpu / Forward 0.00002514 s 0.000012326300020504277 s 2.04
sum / Jax / cpu / Forward 0.000024955 s 0.000013074359976599226 s 1.91
sum / HLOOpt / cpu / Forward 0.000024985 s 0.000017311880010311143 s 1.44
sum / PartOpt / cpu / Forward 0.000024763 s 0.0000170316599724174 s 1.45
sum / IPartOpt / cpu / Forward 0.000025117 s 0.000012417759990057675 s 2.02
sum / DefOpt / cpu / Forward 0.000031877 s 0.000017328099993392243 s 1.84
sum / IDefOpt / cpu / Forward 0.000025129 s 0.000012513379988376983 s 2.01
sum / JaXPipe / cpu / PreRev 0.000023849 s 0.000012807400034944294 s 1.86
sum / JaXPipe / cpu / PostRev 0.000023556 s 0.00001236651999533933 s 1.90
sum / JaXPipe / cpu / BothRev 0.000023809 s 0.000016243300015048588 s 1.47
sum / Jax / cpu / BothRev 0.000023176 s 0.000012590739979714271 s 1.84
sum / HLOOpt / cpu / PreRev 0.000023606 s 0.000012084299987691338 s 1.95
sum / HLOOpt / cpu / PostRev 0.00002392 s 0.000015077259995450733 s 1.59
sum / HLOOpt / cpu / BothRev 0.000023256 s 0.000013451559989334782 s 1.73
sum / PartOpt / cpu / PreRev 0.000024122 s 0.00001157794003120216 s 2.08
sum / PartOpt / cpu / PostRev 0.000023634 s 0.000012062860005244149 s 1.96
sum / PartOpt / cpu / BothRev 0.000024436 s 0.00001210502000503766 s 2.02
sum / IPartOpt / cpu / PreRev 0.000023549 s 0.000012810119978894362 s 1.84
sum / IPartOpt / cpu / PostRev 0.000024012 s 0.00001243552001142234 s 1.93
sum / IPartOpt / cpu / BothRev 0.000023664 s 0.000011493500041979132 s 2.06
sum / DefOpt / cpu / PreRev 0.000023464 s 0.000011542479978743358 s 2.03
sum / DefOpt / cpu / PostRev 0.000024454 s 0.00001140446000135853 s 2.14
sum / DefOpt / cpu / BothRev 0.000024824 s 0.000011926980014322908 s 2.08
sum / IDefOpt / cpu / PreRev 0.000023555 s 0.000011830739940705826 s 1.99
sum / IDefOpt / cpu / PostRev 0.000023704 s 0.00001181827999971574 s 2.01
sum / IDefOpt / cpu / BothRev 0.000023899 s 0.000011645999948086682 s 2.05
sum / JaXPipe / cpu / Primal 0.00001 s 0.000008944359979068395 s 1.12
sum / Jax / cpu / Primal 0.00001 s 0.000008394739988943912 s 1.19
sum / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.000012469859966586229 s 0.72
sum / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000008866120006132404 s 1.02
sum / IPartOpt / cpu / Primal 0.00001 s 0.000009332300041933197 s 1.07
sum / DefOpt / cpu / Primal 0.00001 s 0.000012841500001741224 s 0.78
sum / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000008591099995101103 s 1.05
sum / JaXPipe / cpu / Forward 0.000013 s 0.000012326300020504277 s 1.05
sum / Jax / cpu / Forward 0.000014 s 0.000013074359976599226 s 1.07
sum / HLOOpt / cpu / Forward 0.000013 s 0.000017311880010311143 s 0.75
sum / PartOpt / cpu / Forward 0.000014 s 0.0000170316599724174 s 0.82
sum / IPartOpt / cpu / Forward 0.000013 s 0.000012417759990057675 s 1.05
sum / DefOpt / cpu / Forward 0.000013 s 0.000017328099993392243 s 0.75
sum / IDefOpt / cpu / Forward 0.000014 s 0.000012513379988376983 s 1.12
sum / JaXPipe / cpu / PreRev 0.000012 s 0.000012807400034944294 s 0.94
sum / JaXPipe / cpu / PostRev 0.000013 s 0.00001236651999533933 s 1.05
sum / JaXPipe / cpu / BothRev 0.000013 s 0.000016243300015048588 s 0.80
sum / Jax / cpu / BothRev 0.000013 s 0.000012590739979714271 s 1.03
sum / HLOOpt / cpu / PreRev 0.000013 s 0.000012084299987691338 s 1.08
sum / HLOOpt / cpu / PostRev 0.000013 s 0.000015077259995450733 s 0.86
sum / HLOOpt / cpu / BothRev 0.000013 s 0.000013451559989334782 s 0.97
sum / PartOpt / cpu / PreRev 0.000013 s 0.00001157794003120216 s 1.12
sum / PartOpt / cpu / PostRev 0.000013 s 0.000012062860005244149 s 1.08
sum / PartOpt / cpu / BothRev 0.000012 s 0.00001210502000503766 s 0.99
sum / IPartOpt / cpu / PreRev 0.000013 s 0.000012810119978894362 s 1.01
sum / IPartOpt / cpu / PostRev 0.000012 s 0.00001243552001142234 s 0.96
sum / IPartOpt / cpu / BothRev 0.000013 s 0.000011493500041979132 s 1.13
sum / DefOpt / cpu / PreRev 0.000013 s 0.000011542479978743358 s 1.13
sum / DefOpt / cpu / PostRev 0.000013 s 0.00001140446000135853 s 1.14
sum / DefOpt / cpu / BothRev 0.000013 s 0.000011926980014322908 s 1.09
sum / IDefOpt / cpu / PreRev 0.000012 s 0.000011830739940705826 s 1.01
sum / IDefOpt / cpu / PostRev 0.000013 s 0.00001181827999971574 s 1.10
sum / IDefOpt / cpu / BothRev 0.000013 s 0.000011645999948086682 s 1.12
value_and_grad / JaXPipe / cpu / Primal 0.000016157740001290223 s 0.000015609660003974568 s 1.04
value_and_grad / Jax / cpu / Primal 0.000015475239961233457 s 0.00001694587999736541 s 0.91
value_and_grad / HLOOpt / cpu / Primal 0.000015901099977781996 s 0.000015964899976097513 s 1.00
value_and_grad / PartOpt / cpu / Primal 0.000015634899982615023 s 0.000015452320003532804 s 1.01
value_and_grad / IPartOpt / cpu / Primal 0.00001576279999426333 s 0.000015345120036727167 s 1.03
value_and_grad / DefOpt / cpu / Primal 0.000015485739995710902 s 0.00001504838006439968 s 1.03
value_and_grad / IDefOpt / cpu / Primal 0.000015489620000153082 s 0.000015279319959518035 s 1.01
value_and_grad / JaXPipe / cuda / Primal 0.000032800000000000004 s 0.000034336 s 0.96
value_and_grad / Jax / cuda / Primal 0.000032992 s 0.00003408 s 0.97
value_and_grad / HLOOpt / cuda / Primal 0.000032383000000000005 s 0.00003408 s 0.95
value_and_grad / PartOpt / cuda / Primal 0.000032959 s 0.000034689 s 0.95
value_and_grad / IPartOpt / cuda / Primal 0.000033471 s 0.000034208 s 0.98
value_and_grad / DefOpt / cuda / Primal 0.000033119999999999995 s 0.000034336 s 0.96
value_and_grad / IDefOpt / cuda / Primal 0.00003264 s 0.000034592 s 0.94
value_and_grad / JaXPipe / tpu / Primal 0 s 0 s 1
value_and_grad / Jax / tpu / Primal 0 s 0 s 1
value_and_grad / HLOOpt / tpu / Primal 0 s 0 s 1
value_and_grad / PartOpt / tpu / Primal 0 s 0 s 1
value_and_grad / IPartOpt / tpu / Primal 0 s 0 s 1
value_and_grad / DefOpt / tpu / Primal 0 s 0 s 1
value_and_grad / IDefOpt / tpu / Primal 0 s 0 s 1
value_and_grad / JaXPipe / cpu / Primal 0.000028677 s 0.000015609660003974568 s 1.84
value_and_grad / Jax / cpu / Primal 0.000028026 s 0.00001694587999736541 s 1.65
value_and_grad / HLOOpt / cpu / Primal 0.000028418 s 0.000015964899976097513 s 1.78
value_and_grad / PartOpt / cpu / Primal 0.000034069 s 0.000015452320003532804 s 2.20
value_and_grad / IPartOpt / cpu / Primal 0.00002843 s 0.000015345120036727167 s 1.85
value_and_grad / DefOpt / cpu / Primal 0.000028668 s 0.00001504838006439968 s 1.91
value_and_grad / IDefOpt / cpu / Primal 0.000028631 s 0.000015279319959518035 s 1.87
value_and_grad / JaXPipe / cpu / Primal 0.000014 s 0.000015609660003974568 s 0.90
value_and_grad / Jax / cpu / Primal 0.000015 s 0.00001694587999736541 s 0.89
value_and_grad / HLOOpt / cpu / Primal 0.000015 s 0.000015964899976097513 s 0.94
value_and_grad / PartOpt / cpu / Primal 0.000015 s 0.000015452320003532804 s 0.97
value_and_grad / IPartOpt / cpu / Primal 0.000015 s 0.000015345120036727167 s 0.98
value_and_grad / DefOpt / cpu / Primal 0.000015 s 0.00001504838006439968 s 1.00
value_and_grad / IDefOpt / cpu / Primal 0.000016 s 0.000015279319959518035 s 1.05
jaxmd20 / JaXPipe / cuda / Primal 0.001533431 s 0.00155254 s 0.99
jaxmd20 / Jax / cuda / Primal 0.001511672 s 0.0014699169999999 s 1.03
jaxmd20 / HLOOpt / cuda / Primal 0.00113369 s 0.001089854 s 1.04
jaxmd20 / PartOpt / cuda / Primal 0.001385176 s 0.001328637 s 1.04
jaxmd20 / IPartOpt / cuda / Primal 0.001341655 s 0.001323804 s 1.01
jaxmd20 / DefOpt / cuda / Primal 0.000530333 s 0.000537983 s 0.99
jaxmd20 / IDefOpt / cuda / Primal 0.000511229 s 0.000483038 s 1.06
jaxmd20 / JaXPipe / cuda / Forward 0.000815516 s 0.000816734 s 1.00
jaxmd20 / Jax / cuda / Forward 0.001804021 s 0.001810108 s 1.00
jaxmd20 / HLOOpt / cuda / Forward 0.000832796 s 0.000834014 s 1.00
jaxmd20 / PartOpt / cuda / Forward 0.000825852 s 0.000822974 s 1.00
jaxmd20 / IPartOpt / cuda / Forward 0.000823771 s 0.000823486 s 1.00
jaxmd20 / DefOpt / cuda / Forward 0.000824859 s 0.000831743 s 0.99
jaxmd20 / IDefOpt / cuda / Forward 0.000825596 s 0.000831134 s 0.99
jaxmd20 / JaXPipe / cuda / PreRev 0.001694262 s 0.0016501719999999 s 1.03
jaxmd20 / JaXPipe / cuda / PostRev 0.005322819 s 0.0052785149999999 s 1.01
jaxmd20 / JaXPipe / cuda / BothRev 0.0016485669999999 s 0.001692028 s 0.97
jaxmd20 / Jax / cuda / BothRev 0.005759393 s 0.005282833 s 1.09
jaxmd20 / HLOOpt / cuda / PreRev 0.001770358 s 0.001790971 s 0.99
jaxmd20 / HLOOpt / cuda / PostRev 0.0051849 s 0.005920145 s 0.88
jaxmd20 / HLOOpt / cuda / BothRev 0.001651414 s 0.001740156 s 0.95
jaxmd20 / PartOpt / cuda / PreRev 0.001763958 s 0.0016984889999999 s 1.04
jaxmd20 / PartOpt / cuda / PostRev 0.005398917 s 0.0053996339999999 s 1.00
jaxmd20 / PartOpt / cuda / BothRev 0.001737558 s 0.001708827 s 1.02
jaxmd20 / IPartOpt / cuda / PreRev 0.0017294629999999 s 0.001710396 s 1.01
jaxmd20 / IPartOpt / cuda / PostRev 0.005407841 s 0.00529346 s 1.02
jaxmd20 / IPartOpt / cuda / BothRev 0.001673111 s 0.001627835 s 1.03
jaxmd20 / DefOpt / cuda / PreRev 0.001729335 s 0.001701085 s 1.02
jaxmd20 / DefOpt / cuda / PostRev 0.002727962 s 0.0027234159999999 s 1.00
jaxmd20 / DefOpt / cuda / BothRev 0.001689462 s 0.001657948 s 1.02
jaxmd20 / IDefOpt / cuda / PreRev 0.001761237 s 0.001719611 s 1.02
jaxmd20 / IDefOpt / cuda / PostRev 0.002005462 s 0.001983322 s 1.01
jaxmd20 / IDefOpt / cuda / BothRev 0.00166687 s 0.001653083 s 1.01
jaxmd20 / JaXPipe / tpu / Primal 0.009264360625 s 0.00926560375 s 1.00
jaxmd20 / Jax / tpu / Primal 0.009277481875 s 0.00927575 s 1.00
jaxmd20 / HLOOpt / tpu / Primal 0.00915289875 s 0.0091462187499999 s 1.00
jaxmd20 / PartOpt / tpu / Primal 0.0092049125 s 0.009196774375 s 1.00
jaxmd20 / IPartOpt / tpu / Primal 0.009205248125 s 0.009196335 s 1.00
jaxmd20 / DefOpt / tpu / Primal 0.008756299375 s 0.008754564375 s 1.00
jaxmd20 / IDefOpt / tpu / Primal 0.008629999375 s 0.0086285299999999 s 1.00
jaxmd20 / JaXPipe / tpu / Forward 0.01725498375 s 0.017255860625 s 1.00
jaxmd20 / Jax / tpu / Forward 0.018724606875 s 0.018740535 s 1.00
jaxmd20 / HLOOpt / tpu / Forward 0.01723884125 s 0.01723875375 s 1.00
jaxmd20 / PartOpt / tpu / Forward 0.01726343125 s 0.017255685 s 1.00
jaxmd20 / IPartOpt / tpu / Forward 0.017255110625 s 0.01725528375 s 1.00
jaxmd20 / DefOpt / tpu / Forward 0.017266596875 s 0.0172592481249999 s 1.00
jaxmd20 / IDefOpt / tpu / Forward 0.017255185625 s 0.01726168625 s 1.00
jaxmd20 / JaXPipe / tpu / PreRev 0.025364616875 s 0.02535351625 s 1.00
jaxmd20 / JaXPipe / tpu / PostRev 0.021849159375 s 0.021859570625 s 1.00
jaxmd20 / JaXPipe / tpu / BothRev 0.0253645131249999 s 0.025357175 s 1.00
jaxmd20 / Jax / tpu / BothRev 0.021850155 s 0.0218619075 s 1.00
jaxmd20 / HLOOpt / tpu / PreRev 0.025356334375 s 0.0253555 s 1.00
jaxmd20 / HLOOpt / tpu / PostRev 0.0207011 s 0.02096289 s 0.99
jaxmd20 / HLOOpt / tpu / BothRev 0.025272425625 s 0.025256925 s 1.00
jaxmd20 / PartOpt / tpu / PreRev 0.025341625625 s 0.02535721875 s 1.00
jaxmd20 / PartOpt / tpu / PostRev 0.02151078625 s 0.0215000425 s 1.00
jaxmd20 / PartOpt / tpu / BothRev 0.025251536875 s 0.02527976 s 1.00
jaxmd20 / IPartOpt / tpu / PreRev 0.02535884625 s 0.0253414299999999 s 1.00
jaxmd20 / IPartOpt / tpu / PostRev 0.0215084275 s 0.021496316875 s 1.00
jaxmd20 / IPartOpt / tpu / BothRev 0.025273363125 s 0.0252540024999999 s 1.00
jaxmd20 / DefOpt / tpu / PreRev 0.02534130125 s 0.0253602562499999 s 1.00
jaxmd20 / DefOpt / tpu / PostRev 0.018894123125 s 0.018940609375 s 1.00
jaxmd20 / DefOpt / tpu / BothRev 0.02525188 s 0.02527341375 s 1.00
jaxmd20 / IDefOpt / tpu / PreRev 0.0253570662499999 s 0.025349679375 s 1.00
jaxmd20 / IDefOpt / tpu / PostRev 0.01836286375 s 0.018402348125 s 1.00
jaxmd20 / IDefOpt / tpu / BothRev 0.0252670074999999 s 0.02527000125 s 1.00
jaxmd40 / JaXPipe / cpu / Primal 0.0921604499999999 s 0.086540163 s 1.06
jaxmd40 / Jax / cpu / Primal 0.065779993 s 0.072547063 s 0.91
jaxmd40 / HLOOpt / cpu / Primal 0.1119468669999999 s 0.12096931 s 0.93
jaxmd40 / PartOpt / cpu / Primal 0.087218423 s 0.085755572 s 1.02
jaxmd40 / IPartOpt / cpu / Primal 0.091370942 s 0.0849084639999999 s 1.08
jaxmd40 / DefOpt / cpu / Primal 0.11946146 s 0.108344307 s 1.10
jaxmd40 / IDefOpt / cpu / Primal 0.113589541 s 0.103137147 s 1.10
jaxmd40 / JaXPipe / cpu / Forward 0.218499315 s 0.20085107 s 1.09
jaxmd40 / Jax / cpu / Forward 0.116966213 s 0.1088797919999999 s 1.07
jaxmd40 / HLOOpt / cpu / Forward 0.215814815 s 0.200093359 s 1.08
jaxmd40 / PartOpt / cpu / Forward 0.210972408 s 0.1919018649999999 s 1.10
jaxmd40 / IPartOpt / cpu / Forward 0.218689276 s 0.19644185 s 1.11
jaxmd40 / DefOpt / cpu / Forward 0.215531977 s 0.199895264 s 1.08
jaxmd40 / IDefOpt / cpu / Forward 0.209596699 s 0.19931931 s 1.05
jaxmd40 / JaXPipe / cpu / PreRev 0.2917424409999999 s 0.258208964 s 1.13
jaxmd40 / JaXPipe / cpu / PostRev 0.177779821 s 0.1776087609999999 s 1.00
jaxmd40 / JaXPipe / cpu / BothRev 0.273956639 s 0.256219958 s 1.07
jaxmd40 / Jax / cpu / BothRev 0.1836596979999999 s 0.154869852 s 1.19
jaxmd40 / HLOOpt / cpu / PreRev 0.291914947 s 0.256441499 s 1.14
jaxmd40 / HLOOpt / cpu / PostRev 0.2454067919999999 s 0.211236406 s 1.16
jaxmd40 / HLOOpt / cpu / BothRev 0.321410827 s 0.281489736 s 1.14
jaxmd40 / PartOpt / cpu / PreRev 0.2783214939999999 s 0.2645066279999999 s 1.05
jaxmd40 / PartOpt / cpu / PostRev 0.167544312 s 0.159215249 s 1.05
jaxmd40 / PartOpt / cpu / BothRev 0.312836129 s 0.307561513 s 1.02
jaxmd40 / IPartOpt / cpu / PreRev 0.279950214 s 0.257804417 s 1.09
jaxmd40 / IPartOpt / cpu / PostRev 0.175525194 s 0.149934956 s 1.17
jaxmd40 / IPartOpt / cpu / BothRev 0.324499032 s 0.273620266 s 1.19
jaxmd40 / DefOpt / cpu / PreRev 0.275662784 s 0.254484876 s 1.08
jaxmd40 / DefOpt / cpu / PostRev 0.2199199889999999 s 0.207815575 s 1.06
jaxmd40 / DefOpt / cpu / BothRev 0.319323106 s 0.277477551 s 1.15
jaxmd40 / IDefOpt / cpu / PreRev 0.290771709 s 0.266567457 s 1.09
jaxmd40 / IDefOpt / cpu / PostRev 0.247079627 s 0.201902107 s 1.22
jaxmd40 / IDefOpt / cpu / BothRev 0.294098969 s 0.2905241209999999 s 1.01
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / JaXPipe / cuda / Primal 1.704559346 s 1.701648624 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / Jax / cuda / Primal 1.707272507 s 1.704462625 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / HLOOpt / cuda / Primal 1.718296577 s 1.7169312440000002 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / PartOpt / cuda / Primal 1.698482416 s 1.696569026 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IPartOpt / cuda / Primal 1.696644364 s 1.694477234 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / DefOpt / cuda / Primal 1.710956954 s 1.7079774 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IDefOpt / cuda / Primal 1.960933902 s 1.957368557 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / JaXPipe / tpu / Primal 4.0122493275 s 3.994860898125 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / Jax / tpu / Primal 3.03918676 s 3.038576945625 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / HLOOpt / tpu / Primal 3.12164779625 s 3.121049703125 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / PartOpt / tpu / Primal 3.059246383125 s 3.0587828525 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IPartOpt / tpu / Primal 3.059368750625 s 3.058850341875 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / DefOpt / tpu / Primal 2.2637617875 s 2.263365245 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IDefOpt / tpu / Primal 4.74358085375 s 4.7427646175 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / JaXPipe / cpu / Primal 7.421353696 s 6.980644399 s 1.06
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / Jax / cpu / Primal 7.387777943 s 6.826147064 s 1.08
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / HLOOpt / cpu / Primal 7.134158391 s 6.931855829 s 1.03
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / PartOpt / cpu / Primal 7.60048433 s 7.02564593 s 1.08
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / IPartOpt / cpu / Primal 7.418664468 s 6.977051482 s 1.06
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / DefOpt / cpu / Primal 3.097766698 s 2.730640324 s 1.13
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / IDefOpt / cpu / Primal 8.068911395 s 7.594417568 s 1.06

This comment was automatically generated by workflow using github-action-benchmark.

@vimarsh6739 vimarsh6739 merged commit 2685ba6 into main Feb 5, 2026
22 of 25 checks passed
@vimarsh6739 vimarsh6739 deleted the hiprt-conversion branch February 5, 2026 19:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants