Skip to content

Conversation

@enzymead-bot
Copy link
Contributor

@enzymead-bot enzymead-bot bot commented Jan 4, 2026

@enzymead-bot enzymead-bot bot added the dependencies Pull requests that update a dependency file label Jan 4, 2026
@wsmoses wsmoses merged commit 008dac3 into main Jan 4, 2026
22 checks passed
@wsmoses wsmoses deleted the update-enzymead-enzyme branch January 4, 2026 10:44
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: 6ba10b5 Previous: 008dac3 Ratio
actmtch / JaXPipe / cpu / Primal 0.00000711891996616032 s 0.000006760500036762096 s 1.05
actmtch / Jax / cpu / Primal 0.000007122600018192316 s 0.000006411220001609764 s 1.11
actmtch / HLOOpt / cpu / Primal 0.00000799637999080005 s 0.000008145720021275339 s 0.98
actmtch / PartOpt / cpu / Primal 0.000007919440013210988 s 0.000006577820022357627 s 1.20
actmtch / IPartOpt / cpu / Primal 0.000007384120026472374 s 0.000006845999960205517 s 1.08
actmtch / DefOpt / cpu / Primal 0.000007814239961589919 s 0.000007159520037021139 s 1.09
actmtch / IDefOpt / cpu / Primal 0.000007353140044870088 s 0.000007819799966455321 s 0.94
actmtch / JaXPipe / cpu / Forward 0.00001117671999054437 s 0.00001100208001844294 s 1.02
actmtch / Jax / cpu / Forward 0.000010938559998976416 s 0.00001000555997052288 s 1.09
actmtch / HLOOpt / cpu / Forward 0.000012474639934225708 s 0.000011567560004550616 s 1.08
actmtch / PartOpt / cpu / Forward 0.000012058939973940142 s 0.000010661280030035412 s 1.13
actmtch / IPartOpt / cpu / Forward 0.000012128520020269206 s 0.000011272220053797356 s 1.08
actmtch / DefOpt / cpu / Forward 0.000011871959986820004 s 0.000010422680006740849 s 1.14
actmtch / IDefOpt / cpu / Forward 0.000011829519980892657 s 0.000010585179988993332 s 1.12
actmtch / JaXPipe / cpu / PreRev 0.000012125000002924935 s 0.00001065782000296167 s 1.14
actmtch / JaXPipe / cpu / PostRev 0.000011016960042979916 s 0.000009895780012811885 s 1.11
actmtch / JaXPipe / cpu / BothRev 0.000012555740004245307 s 0.000011262480038567446 s 1.11
actmtch / Jax / cpu / BothRev 0.000011011999995389488 s 0.00001015358000586275 s 1.08
actmtch / HLOOpt / cpu / PreRev 0.00001271603999157378 s 0.000010812100017574266 s 1.18
actmtch / HLOOpt / cpu / PostRev 0.000013917300011598854 s 0.000012807180000891094 s 1.09
actmtch / HLOOpt / cpu / BothRev 0.000012507219971666928 s 0.00001181797997560352 s 1.06
actmtch / PartOpt / cpu / PreRev 0.000012131820012655226 s 0.000010439159987072345 s 1.16
actmtch / PartOpt / cpu / PostRev 0.000010517060027268598 s 0.000010471360028532217 s 1.00
actmtch / PartOpt / cpu / BothRev 0.00001263714000742766 s 0.000011898939937964317 s 1.06
actmtch / IPartOpt / cpu / PreRev 0.000011436459953984012 s 0.000010801919979712692 s 1.06
actmtch / IPartOpt / cpu / PostRev 0.000010947100017801858 s 0.000009592260012141196 s 1.14
actmtch / IPartOpt / cpu / BothRev 0.000012845659975937453 s 0.000011451419995864853 s 1.12
actmtch / DefOpt / cpu / PreRev 0.000011596120002650422 s 0.000010753459991974524 s 1.08
actmtch / DefOpt / cpu / PostRev 0.000012296319982851856 s 0.00001099387999602186 s 1.12
actmtch / DefOpt / cpu / BothRev 0.000012394620016493718 s 0.000011193439931957982 s 1.11
actmtch / IDefOpt / cpu / PreRev 0.0000117150999722071 s 0.000011068260055253632 s 1.06
actmtch / IDefOpt / cpu / PostRev 0.000012418659989634762 s 0.000011407880001570448 s 1.09
actmtch / IDefOpt / cpu / BothRev 0.000012367840035949484 s 0.000011293160023342352 s 1.10
actmtch / JaXPipe / cuda / Primal 0.000002016 s 0.000002015 s 1.00
actmtch / Jax / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / HLOOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / PartOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / IPartOpt / cuda / Primal 0.000002016 s 0.000002015 s 1.00
actmtch / DefOpt / cuda / Primal 0.000002015 s 0.000002016 s 1.00
actmtch / IDefOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
actmtch / JaXPipe / cuda / Forward 0.000010272 s 0.000009664 s 1.06
actmtch / Jax / cuda / Forward 0.000010144 s 0.000009888 s 1.03
actmtch / HLOOpt / cuda / Forward 0.000010273 s 0.000009952 s 1.03
actmtch / PartOpt / cuda / Forward 0.000010432 s 0.000009728 s 1.07
actmtch / IPartOpt / cuda / Forward 0.000009984 s 0.000009888 s 1.01
actmtch / DefOpt / cuda / Forward 0.000010687 s 0.000010016 s 1.07
actmtch / IDefOpt / cuda / Forward 0.000011328 s 0.000010144 s 1.12
actmtch / JaXPipe / cuda / PreRev 0.00001056 s 0.000009569 s 1.10
actmtch / JaXPipe / cuda / PostRev 0.000011424 s 0.00000976 s 1.17
actmtch / JaXPipe / cuda / BothRev 0.00001104 s 0.000009728 s 1.13
actmtch / Jax / cuda / BothRev 0.000010784 s 0.000010271 s 1.05
actmtch / HLOOpt / cuda / PreRev 0.000010304 s 0.000010144 s 1.02
actmtch / HLOOpt / cuda / PostRev 0.000010976 s 0.000009888 s 1.11
actmtch / HLOOpt / cuda / BothRev 0.000010113 s 0.000009985 s 1.01
actmtch / PartOpt / cuda / PreRev 0.000010304 s 0.000010176 s 1.01
actmtch / PartOpt / cuda / PostRev 0.000011648 s 0.000009825 s 1.19
actmtch / PartOpt / cuda / BothRev 0.000011168 s 0.000009408 s 1.19
actmtch / IPartOpt / cuda / PreRev 0.000011424 s 0.000010016 s 1.14
actmtch / IPartOpt / cuda / PostRev 0.000010207 s 0.000009888 s 1.03
actmtch / IPartOpt / cuda / BothRev 0.000010144 s 0.000009696 s 1.05
actmtch / DefOpt / cuda / PreRev 0.000010496 s 0.000010048 s 1.04
actmtch / DefOpt / cuda / PostRev 0.0000104 s 0.000009824 s 1.06
actmtch / DefOpt / cuda / BothRev 0.000010368 s 0.00001024 s 1.01
actmtch / IDefOpt / cuda / PreRev 0.000010431 s 0.000010304 s 1.01
actmtch / IDefOpt / cuda / PostRev 0.000010144 s 0.000009729 s 1.04
actmtch / IDefOpt / cuda / BothRev 0.0000104 s 0.000010111 s 1.03
actmtch / JaXPipe / tpu / Primal 5.63825e-7 s 5.63425e-7 s 1.00
actmtch / Jax / tpu / Primal 5.968500000000001e-7 s 5.966250000000001e-7 s 1.00
actmtch / HLOOpt / tpu / Primal 0.000002096025 s 0.0000021007 s 1.00
actmtch / PartOpt / tpu / Primal 5.965999999999999e-7 s 5.967e-7 s 1.00
actmtch / IPartOpt / tpu / Primal 5.5285e-7 s 5.5285e-7 s 1
actmtch / DefOpt / tpu / Primal 0.000002163875 s 0.00000216535 s 1.00
actmtch / IDefOpt / tpu / Primal 0.0000021069750000000003 s 0.000002095925 s 1.01
actmtch / JaXPipe / tpu / Forward 0.00000383135 s 0.000003833699999999999 s 1.00
actmtch / Jax / tpu / Forward 0.0000012078 s 0.0000012137 s 1.00
actmtch / HLOOpt / tpu / Forward 0.000003951124999999999 s 0.000003934825 s 1.00
actmtch / PartOpt / tpu / Forward 0.000003912975 s 0.000003914125 s 1.00
actmtch / IPartOpt / tpu / Forward 0.000003948775 s 0.0000039562 s 1.00
actmtch / DefOpt / tpu / Forward 0.000003925925 s 0.0000039197 s 1.00
actmtch / IDefOpt / tpu / Forward 0.000003926374999999999 s 0.000003948875 s 0.99
actmtch / JaXPipe / tpu / PreRev 0.000003466625 s 0.000003474775 s 1.00
actmtch / JaXPipe / tpu / PostRev 0.0000016367 s 0.000001638225 s 1.00
actmtch / JaXPipe / tpu / BothRev 0.0000034688 s 0.0000034755000000000004 s 1.00
actmtch / Jax / tpu / BothRev 0.00000163515 s 0.0000016347999999999998 s 1.00
actmtch / HLOOpt / tpu / PreRev 0.00000348815 s 0.0000034835 s 1.00
actmtch / HLOOpt / tpu / PostRev 0.0000034156750000000005 s 0.0000034127000000000004 s 1.00
actmtch / HLOOpt / tpu / BothRev 0.000003482175 s 0.00000346445 s 1.01
actmtch / PartOpt / tpu / PreRev 0.0000034111000000000003 s 0.0000034168 s 1.00
actmtch / PartOpt / tpu / PostRev 0.000001595925 s 0.000001595925 s 1
actmtch / PartOpt / tpu / BothRev 0.00000340925 s 0.00000340855 s 1.00
actmtch / IPartOpt / tpu / PreRev 0.000003485975 s 0.000003468525 s 1.01
actmtch / IPartOpt / tpu / PostRev 0.00000164645 s 0.000001630525 s 1.01
actmtch / IPartOpt / tpu / BothRev 0.0000034793750000000005 s 0.00000348495 s 1.00
actmtch / DefOpt / tpu / PreRev 0.0000034308499999999995 s 0.000003405575 s 1.01
actmtch / DefOpt / tpu / PostRev 0.00000342335 s 0.000003419525 s 1.00
actmtch / DefOpt / tpu / BothRev 0.000003423375 s 0.00000341945 s 1.00
actmtch / IDefOpt / tpu / PreRev 0.0000034704000000000003 s 0.000003468625 s 1.00
actmtch / IDefOpt / tpu / PostRev 0.00000342245 s 0.0000034068500000000003 s 1.00
actmtch / IDefOpt / tpu / BothRev 0.0000034898 s 0.000003474975 s 1.00
actmtch / JaXPipe / cpu / Primal 0.000013342 s 0.000006760500036762096 s 1.97
actmtch / Jax / cpu / Primal 0.000021777 s 0.000006411220001609764 s 3.40
actmtch / HLOOpt / cpu / Primal 0.000014175 s 0.000008145720021275339 s 1.74
actmtch / PartOpt / cpu / Primal 0.000013281 s 0.000006577820022357627 s 2.02
actmtch / IPartOpt / cpu / Primal 0.00001327 s 0.000006845999960205517 s 1.94
actmtch / DefOpt / cpu / Primal 0.000014313 s 0.000007159520037021139 s 2.00
actmtch / IDefOpt / cpu / Primal 0.000013922 s 0.000007819799966455321 s 1.78
actmtch / JaXPipe / cpu / Forward 0.000019161 s 0.00001100208001844294 s 1.74
actmtch / Jax / cpu / Forward 0.000018033 s 0.00001000555997052288 s 1.80
actmtch / HLOOpt / cpu / Forward 0.000019309 s 0.000011567560004550616 s 1.67
actmtch / PartOpt / cpu / Forward 0.000018950000000000003 s 0.000010661280030035412 s 1.78
actmtch / IPartOpt / cpu / Forward 0.000019582 s 0.000011272220053797356 s 1.74
actmtch / DefOpt / cpu / Forward 0.000018987 s 0.000010422680006740849 s 1.82
actmtch / IDefOpt / cpu / Forward 0.000019105 s 0.000010585179988993332 s 1.80
actmtch / JaXPipe / cpu / PreRev 0.000019604 s 0.00001065782000296167 s 1.84
actmtch / JaXPipe / cpu / PostRev 0.000017500000000000002 s 0.000009895780012811885 s 1.77
actmtch / JaXPipe / cpu / BothRev 0.00001944 s 0.000011262480038567446 s 1.73
actmtch / Jax / cpu / BothRev 0.000018013999999999997 s 0.00001015358000586275 s 1.77
actmtch / HLOOpt / cpu / PreRev 0.000019606 s 0.000010812100017574266 s 1.81
actmtch / HLOOpt / cpu / PostRev 0.000019596 s 0.000012807180000891094 s 1.53
actmtch / HLOOpt / cpu / BothRev 0.000019635 s 0.00001181797997560352 s 1.66
actmtch / PartOpt / cpu / PreRev 0.000019655 s 0.000010439159987072345 s 1.88
actmtch / PartOpt / cpu / PostRev 0.000017952 s 0.000010471360028532217 s 1.71
actmtch / PartOpt / cpu / BothRev 0.000019632 s 0.000011898939937964317 s 1.65
actmtch / IPartOpt / cpu / PreRev 0.000019309 s 0.000010801919979712692 s 1.79
actmtch / IPartOpt / cpu / PostRev 0.000017257 s 0.000009592260012141196 s 1.80
actmtch / IPartOpt / cpu / BothRev 0.000019651 s 0.000011451419995864853 s 1.72
actmtch / DefOpt / cpu / PreRev 0.000019562 s 0.000010753459991974524 s 1.82
actmtch / DefOpt / cpu / PostRev 0.000019599 s 0.00001099387999602186 s 1.78
actmtch / DefOpt / cpu / BothRev 0.000019875 s 0.000011193439931957982 s 1.78
actmtch / IDefOpt / cpu / PreRev 0.000019064 s 0.000011068260055253632 s 1.72
actmtch / IDefOpt / cpu / PostRev 0.000019604 s 0.000011407880001570448 s 1.72
actmtch / IDefOpt / cpu / BothRev 0.000019671 s 0.000011293160023342352 s 1.74
actmtch / JaXPipe / cpu / Primal 0.00001 s 0.000006760500036762096 s 1.48
actmtch / Jax / cpu / Primal 0.000008999999999999999 s 0.000006411220001609764 s 1.40
actmtch / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.000008145720021275339 s 1.10
actmtch / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000006577820022357627 s 1.37
actmtch / IPartOpt / cpu / Primal 0.000008999999999999999 s 0.000006845999960205517 s 1.31
actmtch / DefOpt / cpu / Primal 0.00001 s 0.000007159520037021139 s 1.40
actmtch / IDefOpt / cpu / Primal 0.00001 s 0.000007819799966455321 s 1.28
actmtch / JaXPipe / cpu / Forward 0.000014 s 0.00001100208001844294 s 1.27
actmtch / Jax / cpu / Forward 0.000012 s 0.00001000555997052288 s 1.20
actmtch / HLOOpt / cpu / Forward 0.000014 s 0.000011567560004550616 s 1.21
actmtch / PartOpt / cpu / Forward 0.000015 s 0.000010661280030035412 s 1.41
actmtch / IPartOpt / cpu / Forward 0.000014 s 0.000011272220053797356 s 1.24
actmtch / DefOpt / cpu / Forward 0.000013 s 0.000010422680006740849 s 1.25
actmtch / IDefOpt / cpu / Forward 0.000014 s 0.000010585179988993332 s 1.32
actmtch / JaXPipe / cpu / PreRev 0.000013 s 0.00001065782000296167 s 1.22
actmtch / JaXPipe / cpu / PostRev 0.000011 s 0.000009895780012811885 s 1.11
actmtch / JaXPipe / cpu / BothRev 0.000014 s 0.000011262480038567446 s 1.24
actmtch / Jax / cpu / BothRev 0.000012 s 0.00001015358000586275 s 1.18
actmtch / HLOOpt / cpu / PreRev 0.000014 s 0.000010812100017574266 s 1.29
actmtch / HLOOpt / cpu / PostRev 0.000013 s 0.000012807180000891094 s 1.02
actmtch / HLOOpt / cpu / BothRev 0.000014 s 0.00001181797997560352 s 1.18
actmtch / PartOpt / cpu / PreRev 0.000014 s 0.000010439159987072345 s 1.34
actmtch / PartOpt / cpu / PostRev 0.000013 s 0.000010471360028532217 s 1.24
actmtch / PartOpt / cpu / BothRev 0.000014 s 0.000011898939937964317 s 1.18
actmtch / IPartOpt / cpu / PreRev 0.000014 s 0.000010801919979712692 s 1.30
actmtch / IPartOpt / cpu / PostRev 0.000013 s 0.000009592260012141196 s 1.36
actmtch / IPartOpt / cpu / BothRev 0.000014 s 0.000011451419995864853 s 1.22
actmtch / DefOpt / cpu / PreRev 0.000014 s 0.000010753459991974524 s 1.30
actmtch / DefOpt / cpu / PostRev 0.000014 s 0.00001099387999602186 s 1.27
actmtch / DefOpt / cpu / BothRev 0.000014 s 0.000011193439931957982 s 1.25
actmtch / IDefOpt / cpu / PreRev 0.000013 s 0.000011068260055253632 s 1.17
actmtch / IDefOpt / cpu / PostRev 0.000014 s 0.000011407880001570448 s 1.23
actmtch / IDefOpt / cpu / BothRev 0.000013 s 0.000011293160023342352 s 1.15
add_one / JaXPipe / cpu / Primal 0.00000696844003869046 s 0.000007061899959808216 s 0.99
add_one / Jax / cpu / Primal 0.000007604100010212278 s 0.000007725300010861246 s 0.98
add_one / HLOOpt / cpu / Primal 0.000007626020033058012 s 0.000007127159988158383 s 1.07
add_one / PartOpt / cpu / Primal 0.0000075094999556313265 s 0.0000064727400240371934 s 1.16
add_one / IPartOpt / cpu / Primal 0.000007266620004884317 s 0.000006855559977338999 s 1.06
add_one / DefOpt / cpu / Primal 0.000008245340013672831 s 0.000006799359998694854 s 1.21
add_one / IDefOpt / cpu / Primal 0.00000758053997742536 s 0.000006532919996971032 s 1.16
add_one / JaXPipe / cpu / Forward 0.000011090380012319655 s 0.000010692179985198891 s 1.04
add_one / Jax / cpu / Forward 0.000011256580019107786 s 0.000009852040057012344 s 1.14
add_one / HLOOpt / cpu / Forward 0.000011349379992680042 s 0.00001061569998455525 s 1.07
add_one / PartOpt / cpu / Forward 0.000010835320008482084 s 0.000010313280026821304 s 1.05
add_one / IPartOpt / cpu / Forward 0.000011383780038158876 s 0.000010236940006507212 s 1.11
add_one / DefOpt / cpu / Forward 0.000011074819976784056 s 0.00001021022000713856 s 1.08
add_one / IDefOpt / cpu / Forward 0.000010992300030920888 s 0.00000989591997495154 s 1.11
add_one / JaXPipe / cpu / PreRev 0.000013310240010468988 s 0.000011687020005410889 s 1.14
add_one / JaXPipe / cpu / PostRev 0.00001303181997172942 s 0.000011235759975534164 s 1.16
add_one / JaXPipe / cpu / BothRev 0.000013435880000542963 s 0.000012070920010955888 s 1.11
add_one / Jax / cpu / BothRev 0.000013003439989915931 s 0.00001110520000111137 s 1.17
add_one / HLOOpt / cpu / PreRev 0.000012714859958578017 s 0.0000117855600456096 s 1.08
add_one / HLOOpt / cpu / PostRev 0.00001498494006227702 s 0.0000174333799895976 s 0.86
add_one / HLOOpt / cpu / BothRev 0.00001277210000807827 s 0.000011470099998405203 s 1.11
add_one / PartOpt / cpu / PreRev 0.000012748560011459632 s 0.000011000640006386676 s 1.16
add_one / PartOpt / cpu / PostRev 0.00001256424001439882 s 0.000011238540000704234 s 1.12
add_one / PartOpt / cpu / BothRev 0.000013391959973887425 s 0.000011834960005217 s 1.13
add_one / IPartOpt / cpu / PreRev 0.00001262571999177453 s 0.000011671399961414864 s 1.08
add_one / IPartOpt / cpu / PostRev 0.000013093240004309336 s 0.000012116419975427562 s 1.08
add_one / IPartOpt / cpu / BothRev 0.000013001139977859566 s 0.000011671220026983064 s 1.11
add_one / DefOpt / cpu / PreRev 0.000012567380008476902 s 0.000011632380019364063 s 1.08
add_one / DefOpt / cpu / PostRev 0.00001325919995906588 s 0.000011632059986368405 s 1.14
add_one / DefOpt / cpu / BothRev 0.000012897880005766637 s 0.00001143328000580368 s 1.13
add_one / IDefOpt / cpu / PreRev 0.000013151539988029982 s 0.00001101984002161771 s 1.19
add_one / IDefOpt / cpu / PostRev 0.000012991299936402356 s 0.000011647559977063794 s 1.12
add_one / IDefOpt / cpu / BothRev 0.000012694779979938177 s 0.00001107572001274093 s 1.15
add_one / JaXPipe / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_one / Jax / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
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 0.000010209 s 0.000010144 s 1.01
add_one / Jax / cuda / Forward 0.000009984 s 0.000010112 s 0.99
add_one / HLOOpt / cuda / Forward 0.00000992 s 0.000011391 s 0.87
add_one / PartOpt / cuda / Forward 0.000009665 s 0.000009824 s 0.98
add_one / IPartOpt / cuda / Forward 0.00001008 s 0.000011392 s 0.88
add_one / DefOpt / cuda / Forward 0.000010304 s 0.000011328 s 0.91
add_one / IDefOpt / cuda / Forward 0.000010175 s 0.000009953 s 1.02
add_one / JaXPipe / cuda / PreRev 0.00002528 s 0.000029536 s 0.86
add_one / JaXPipe / cuda / PostRev 0.000025536 s 0.000029408 s 0.87
add_one / JaXPipe / cuda / BothRev 0.000025441 s 0.000024704 s 1.03
add_one / Jax / cuda / BothRev 0.000025664 s 0.000023936 s 1.07
add_one / HLOOpt / cuda / PreRev 0.000025344 s 0.000029056 s 0.87
add_one / HLOOpt / cuda / PostRev 0.000025473 s 0.000024895 s 1.02
add_one / HLOOpt / cuda / BothRev 0.000025376 s 0.000028992 s 0.88
add_one / PartOpt / cuda / PreRev 0.000025472000000000003 s 0.000024801 s 1.03
add_one / PartOpt / cuda / PostRev 0.000025824 s 0.00002464 s 1.05
add_one / PartOpt / cuda / BothRev 0.000026048 s 0.000024736 s 1.05
add_one / IPartOpt / cuda / PreRev 0.000025889 s 0.00002496 s 1.04
add_one / IPartOpt / cuda / PostRev 0.00002544 s 0.000024607 s 1.03
add_one / IPartOpt / cuda / BothRev 0.000025344 s 0.000024544 s 1.03
add_one / DefOpt / cuda / PreRev 0.00002528 s 0.000024896 s 1.02
add_one / DefOpt / cuda / PostRev 0.000025792 s 0.00002464 s 1.05
add_one / DefOpt / cuda / BothRev 0.000025377 s 0.00002528 s 1.00
add_one / IDefOpt / cuda / PreRev 0.000026336 s 0.000024736 s 1.06
add_one / IDefOpt / cuda / PostRev 0.000025472000000000003 s 0.000024896 s 1.02
add_one / IDefOpt / cuda / BothRev 0.000025537 s 0.000024544 s 1.04
add_one / JaXPipe / tpu / Primal 0.000001433725 s 0.000001425475 s 1.01
add_one / Jax / tpu / Primal 0.0000013976250000000002 s 0.000001400975 s 1.00
add_one / HLOOpt / tpu / Primal 0.000001438375 s 0.00000143205 s 1.00
add_one / PartOpt / tpu / Primal 0.00000140335 s 0.0000014018749999999998 s 1.00
add_one / IPartOpt / tpu / Primal 0.0000014325750000000002 s 0.00000142615 s 1.00
add_one / DefOpt / tpu / Primal 0.00000140265 s 0.0000014055 s 1.00
add_one / IDefOpt / tpu / Primal 0.0000014316000000000002 s 0.000001431325 s 1.00
add_one / JaXPipe / tpu / Forward 0.000001851575 s 0.0000018029 s 1.03
add_one / Jax / tpu / Forward 0.0000018438 s 0.00000184345 s 1.00
add_one / HLOOpt / tpu / Forward 0.000001855375 s 0.000001803225 s 1.03
add_one / PartOpt / tpu / Forward 0.0000018495 s 0.000001844125 s 1.00
add_one / IPartOpt / tpu / Forward 0.000001846575 s 0.000001798375 s 1.03
add_one / DefOpt / tpu / Forward 0.000001836425 s 0.00000183825 s 1.00
add_one / IDefOpt / tpu / Forward 0.0000018495 s 0.0000018087 s 1.02
add_one / JaXPipe / tpu / PreRev 0.000002242525 s 0.0000022364 s 1.00
add_one / JaXPipe / tpu / PostRev 0.0000022342 s 0.000002194825 s 1.02
add_one / JaXPipe / tpu / BothRev 0.00000223295 s 0.000002238375 s 1.00
add_one / Jax / tpu / BothRev 0.000002243025 s 0.00000217835 s 1.03
add_one / HLOOpt / tpu / PreRev 0.00000223825 s 0.000002233875 s 1.00
add_one / HLOOpt / tpu / PostRev 0.00000223965 s 0.0000021854000000000003 s 1.02
add_one / HLOOpt / tpu / BothRev 0.000002232425 s 0.0000022366 s 1.00
add_one / PartOpt / tpu / PreRev 0.000002237675 s 0.0000021858250000000004 s 1.02
add_one / PartOpt / tpu / PostRev 0.0000022399 s 0.00000223615 s 1.00
add_one / PartOpt / tpu / BothRev 0.000002233975 s 0.0000021889500000000004 s 1.02
add_one / IPartOpt / tpu / PreRev 0.00000223925 s 0.000002251875 s 0.99
add_one / IPartOpt / tpu / PostRev 0.000002246925 s 0.0000021842 s 1.03
add_one / IPartOpt / tpu / BothRev 0.000002237425 s 0.00000223985 s 1.00
add_one / DefOpt / tpu / PreRev 0.00000224875 s 0.000002184575 s 1.03
add_one / DefOpt / tpu / PostRev 0.000002242625 s 0.000002233075 s 1.00
add_one / DefOpt / tpu / BothRev 0.00000223925 s 0.0000021877 s 1.02
add_one / IDefOpt / tpu / PreRev 0.000002234025 s 0.000002241625 s 1.00
add_one / IDefOpt / tpu / PostRev 0.00000223905 s 0.0000021856000000000003 s 1.02
add_one / IDefOpt / tpu / BothRev 0.0000022398750000000003 s 0.000002233375 s 1.00
add_one / JaXPipe / cpu / Primal 0.000013064 s 0.000007061899959808216 s 1.85
add_one / Jax / cpu / Primal 0.000012721 s 0.000007725300010861246 s 1.65
add_one / HLOOpt / cpu / Primal 0.000013033 s 0.000007127159988158383 s 1.83
add_one / PartOpt / cpu / Primal 0.000012617 s 0.0000064727400240371934 s 1.95
add_one / IPartOpt / cpu / Primal 0.000012756 s 0.000006855559977338999 s 1.86
add_one / DefOpt / cpu / Primal 0.000013079 s 0.000006799359998694854 s 1.92
add_one / IDefOpt / cpu / Primal 0.000013021 s 0.000006532919996971032 s 1.99
add_one / JaXPipe / cpu / Forward 0.000018097 s 0.000010692179985198891 s 1.69
add_one / Jax / cpu / Forward 0.000017835 s 0.000009852040057012344 s 1.81
add_one / HLOOpt / cpu / Forward 0.000018051 s 0.00001061569998455525 s 1.70
add_one / PartOpt / cpu / Forward 0.000017622 s 0.000010313280026821304 s 1.71
add_one / IPartOpt / cpu / Forward 0.000017848 s 0.000010236940006507212 s 1.74
add_one / DefOpt / cpu / Forward 0.000017856 s 0.00001021022000713856 s 1.75
add_one / IDefOpt / cpu / Forward 0.000018105 s 0.00000989591997495154 s 1.83
add_one / JaXPipe / cpu / PreRev 0.000020057 s 0.000011687020005410889 s 1.72
add_one / JaXPipe / cpu / PostRev 0.000020068 s 0.000011235759975534164 s 1.79
add_one / JaXPipe / cpu / BothRev 0.00001954 s 0.000012070920010955888 s 1.62
add_one / Jax / cpu / BothRev 0.000019419 s 0.00001110520000111137 s 1.75
add_one / HLOOpt / cpu / PreRev 0.000019479 s 0.0000117855600456096 s 1.65
add_one / HLOOpt / cpu / PostRev 0.000019814 s 0.0000174333799895976 s 1.14
add_one / HLOOpt / cpu / BothRev 0.000019784 s 0.000011470099998405203 s 1.72
add_one / PartOpt / cpu / PreRev 0.000020019 s 0.000011000640006386676 s 1.82
add_one / PartOpt / cpu / PostRev 0.000020128 s 0.000011238540000704234 s 1.79
add_one / PartOpt / cpu / BothRev 0.000020206 s 0.000011834960005217 s 1.71
add_one / IPartOpt / cpu / PreRev 0.000020295 s 0.000011671399961414864 s 1.74
add_one / IPartOpt / cpu / PostRev 0.000019967 s 0.000012116419975427562 s 1.65
add_one / IPartOpt / cpu / BothRev 0.000027645 s 0.000011671220026983064 s 2.37
add_one / DefOpt / cpu / PreRev 0.000020536 s 0.000011632380019364063 s 1.77
add_one / DefOpt / cpu / PostRev 0.000019978 s 0.000011632059986368405 s 1.72
add_one / DefOpt / cpu / BothRev 0.000019946 s 0.00001143328000580368 s 1.74
add_one / IDefOpt / cpu / PreRev 0.00001976 s 0.00001101984002161771 s 1.79
add_one / IDefOpt / cpu / PostRev 0.000019696000000000003 s 0.000011647559977063794 s 1.69
add_one / IDefOpt / cpu / BothRev 0.000020116 s 0.00001107572001274093 s 1.82
add_one / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000007061899959808216 s 1.27
add_one / Jax / cpu / Primal 0.000008999999999999999 s 0.000007725300010861246 s 1.17
add_one / HLOOpt / cpu / Primal 0.000008 s 0.000007127159988158383 s 1.12
add_one / PartOpt / cpu / Primal 0.000008999999999999999 s 0.0000064727400240371934 s 1.39
add_one / IPartOpt / cpu / Primal 0.000008 s 0.000006855559977338999 s 1.17
add_one / DefOpt / cpu / Primal 0.000008 s 0.000006799359998694854 s 1.18
add_one / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000006532919996971032 s 1.38
add_one / JaXPipe / cpu / Forward 0.000012 s 0.000010692179985198891 s 1.12
add_one / Jax / cpu / Forward 0.000012 s 0.000009852040057012344 s 1.22
add_one / HLOOpt / cpu / Forward 0.000012 s 0.00001061569998455525 s 1.13
add_one / PartOpt / cpu / Forward 0.000012 s 0.000010313280026821304 s 1.16
add_one / IPartOpt / cpu / Forward 0.000012 s 0.000010236940006507212 s 1.17
add_one / DefOpt / cpu / Forward 0.000012 s 0.00001021022000713856 s 1.18
add_one / IDefOpt / cpu / Forward 0.000012 s 0.00000989591997495154 s 1.21
add_one / JaXPipe / cpu / PreRev 0.000014 s 0.000011687020005410889 s 1.20
add_one / JaXPipe / cpu / PostRev 0.000014 s 0.000011235759975534164 s 1.25
add_one / JaXPipe / cpu / BothRev 0.000014 s 0.000012070920010955888 s 1.16
add_one / Jax / cpu / BothRev 0.000014 s 0.00001110520000111137 s 1.26
add_one / HLOOpt / cpu / PreRev 0.000014 s 0.0000117855600456096 s 1.19
add_one / HLOOpt / cpu / PostRev 0.000014 s 0.0000174333799895976 s 0.80
add_one / HLOOpt / cpu / BothRev 0.000014 s 0.000011470099998405203 s 1.22
add_one / PartOpt / cpu / PreRev 0.000015 s 0.000011000640006386676 s 1.36
add_one / PartOpt / cpu / PostRev 0.000014 s 0.000011238540000704234 s 1.25
add_one / PartOpt / cpu / BothRev 0.000014 s 0.000011834960005217 s 1.18
add_one / IPartOpt / cpu / PreRev 0.000014 s 0.000011671399961414864 s 1.20
add_one / IPartOpt / cpu / PostRev 0.000014 s 0.000012116419975427562 s 1.16
add_one / IPartOpt / cpu / BothRev 0.000014 s 0.000011671220026983064 s 1.20
add_one / DefOpt / cpu / PreRev 0.000014 s 0.000011632380019364063 s 1.20
add_one / DefOpt / cpu / PostRev 0.000014 s 0.000011632059986368405 s 1.20
add_one / DefOpt / cpu / BothRev 0.000014 s 0.00001143328000580368 s 1.22
add_one / IDefOpt / cpu / PreRev 0.000014 s 0.00001101984002161771 s 1.27
add_one / IDefOpt / cpu / PostRev 0.000014 s 0.000011647559977063794 s 1.20
add_one / IDefOpt / cpu / BothRev 0.000014 s 0.00001107572001274093 s 1.26
add_two / JaXPipe / cpu / Primal 0.0000072838799951568944 s 0.000006867799966130406 s 1.06
add_two / Jax / cpu / Primal 0.000007277559998328797 s 0.000006908039995323634 s 1.05
add_two / HLOOpt / cpu / Primal 0.00000759058001676749 s 0.000007006639998508036 s 1.08
add_two / PartOpt / cpu / Primal 0.000007864159988457686 s 0.000007027579995337874 s 1.12
add_two / IPartOpt / cpu / Primal 0.000008142500028043288 s 0.000007588979979118449 s 1.07
add_two / DefOpt / cpu / Primal 0.000007316539977182402 s 0.0000068686600297951375 s 1.07
add_two / IDefOpt / cpu / Primal 0.000007225020008263527 s 0.000006989980029175058 s 1.03
add_two / JaXPipe / cpu / Forward 0.000011805000003732855 s 0.000010030479979832308 s 1.18
add_two / Jax / cpu / Forward 0.000011475920018710896 s 0.000009928719991876278 s 1.16
add_two / HLOOpt / cpu / Forward 0.000011753259987017372 s 0.00001017656001749856 s 1.15
add_two / PartOpt / cpu / Forward 0.000011706080003932584 s 0.000010274779997416772 s 1.14
add_two / IPartOpt / cpu / Forward 0.00001152200002252357 s 0.000010454699986439663 s 1.10
add_two / DefOpt / cpu / Forward 0.000011475180053821532 s 0.000010264359962093295 s 1.12
add_two / IDefOpt / cpu / Forward 0.000011278700003458652 s 0.00001037967996126099 s 1.09
add_two / JaXPipe / cpu / PreRev 0.000015527780042248197 s 0.000013865560013073265 s 1.12
add_two / JaXPipe / cpu / PostRev 0.000015411120002681856 s 0.000014588599997296116 s 1.06
add_two / JaXPipe / cpu / BothRev 0.000014870159975544085 s 0.000014942600009817398 s 1.00
add_two / Jax / cpu / BothRev 0.000015494660019612638 s 0.000013541219987018849 s 1.14
add_two / HLOOpt / cpu / PreRev 0.000015030979984658188 s 0.000014437700037888137 s 1.04
add_two / HLOOpt / cpu / PostRev 0.000018457420001141145 s 0.00001636072001019784 s 1.13
add_two / HLOOpt / cpu / BothRev 0.00001555643999381573 s 0.000014554019971910747 s 1.07
add_two / PartOpt / cpu / PreRev 0.00001528863998828456 s 0.0000141802000325697 s 1.08
add_two / PartOpt / cpu / PostRev 0.000015791539972269675 s 0.0000140553999972326 s 1.12
add_two / PartOpt / cpu / BothRev 0.0000157322399354598 s 0.000014440060003835245 s 1.09
add_two / IPartOpt / cpu / PreRev 0.000014992700025686645 s 0.000014146200001050602 s 1.06
add_two / IPartOpt / cpu / PostRev 0.000015091780014699906 s 0.00001453995997508173 s 1.04
add_two / IPartOpt / cpu / BothRev 0.000015593139969496406 s 0.000014311239992821356 s 1.09
add_two / DefOpt / cpu / PreRev 0.000015304039989132436 s 0.000013563660022555269 s 1.13
add_two / DefOpt / cpu / PostRev 0.00001577400001224305 s 0.00001501925997217768 s 1.05
add_two / DefOpt / cpu / BothRev 0.000015295739976863843 s 0.00001491654000346898 s 1.03
add_two / IDefOpt / cpu / PreRev 0.000015469020008822555 s 0.00001355600003080326 s 1.14
add_two / IDefOpt / cpu / PostRev 0.000016254820029644178 s 0.000014451440019911388 s 1.12
add_two / IDefOpt / cpu / BothRev 0.00001532030003545515 s 0.000014129720002529213 s 1.08
add_two / JaXPipe / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / Jax / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / HLOOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / PartOpt / cuda / Primal 0.0000019200000000000003 s 0.000001919 s 1.00
add_two / IPartOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / DefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / IDefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
add_two / JaXPipe / cuda / Forward 0.000009664 s 0.000009664 s 1
add_two / Jax / cuda / Forward 0.000009952 s 0.000009376 s 1.06
add_two / HLOOpt / cuda / Forward 0.00000992 s 0.000009856 s 1.01
add_two / PartOpt / cuda / Forward 0.000010081 s 0.000009888 s 1.02
add_two / IPartOpt / cuda / Forward 0.000010176 s 0.000009696 s 1.05
add_two / DefOpt / cuda / Forward 0.000009952 s 0.000009792 s 1.02
add_two / IDefOpt / cuda / Forward 0.000010048 s 0.00000976 s 1.03
add_two / JaXPipe / cuda / PreRev 0.00003264 s 0.000032608 s 1.00
add_two / JaXPipe / cuda / PostRev 0.000032416 s 0.000032352 s 1.00
add_two / JaXPipe / cuda / BothRev 0.000032929 s 0.00003264 s 1.01
add_two / Jax / cuda / BothRev 0.000032928 s 0.000032513 s 1.01
add_two / HLOOpt / cuda / PreRev 0.00003232 s 0.00003264 s 0.99
add_two / HLOOpt / cuda / PostRev 0.000032417 s 0.000031616 s 1.03
add_two / HLOOpt / cuda / BothRev 0.0000336 s 0.000032896000000000005 s 1.02
add_two / PartOpt / cuda / PreRev 0.000032864 s 0.000032767999999999995 s 1.00
add_two / PartOpt / cuda / PostRev 0.000033056 s 0.000032064 s 1.03
add_two / PartOpt / cuda / BothRev 0.000032576 s 0.000032576 s 1
add_two / IPartOpt / cuda / PreRev 0.000032736 s 0.000032928 s 0.99
add_two / IPartOpt / cuda / PostRev 0.000032608 s 0.00003168 s 1.03
add_two / IPartOpt / cuda / BothRev 0.000033184 s 0.000032576 s 1.02
add_two / DefOpt / cuda / PreRev 0.000032928 s 0.000032671 s 1.01
add_two / DefOpt / cuda / PostRev 0.000033536999999999994 s 0.00003168 s 1.06
add_two / DefOpt / cuda / BothRev 0.000033793 s 0.000031647000000000004 s 1.07
add_two / IDefOpt / cuda / PreRev 0.000037216 s 0.000032064 s 1.16
add_two / IDefOpt / cuda / PostRev 0.000034432 s 0.000032096 s 1.07
add_two / IDefOpt / cuda / BothRev 0.000033344 s 0.000031937 s 1.04
add_two / JaXPipe / tpu / Primal 0.0000014277500000000002 s 0.0000014319 s 1.00
add_two / Jax / tpu / Primal 0.000001471475 s 0.000001423975 s 1.03
add_two / HLOOpt / tpu / Primal 0.000001427275 s 0.0000014321 s 1.00
add_two / PartOpt / tpu / Primal 0.000001468975 s 0.000001420075 s 1.03
add_two / IPartOpt / tpu / Primal 0.0000014271500000000002 s 0.00000142885 s 1.00
add_two / DefOpt / tpu / Primal 0.0000014790000000000002 s 0.000001434525 s 1.03
add_two / IDefOpt / tpu / Primal 0.000001440425 s 0.000001431175 s 1.01
add_two / JaXPipe / tpu / Forward 0.00000182745 s 0.0000018271 s 1.00
add_two / Jax / tpu / Forward 0.000001834175 s 0.000001827825 s 1.00
add_two / HLOOpt / tpu / Forward 0.0000018287 s 0.00000183625 s 1.00
add_two / PartOpt / tpu / Forward 0.0000018411 s 0.00000182075 s 1.01
add_two / IPartOpt / tpu / Forward 0.0000018284 s 0.000001831175 s 1.00
add_two / DefOpt / tpu / Forward 0.000001822275 s 0.000001834675 s 0.99
add_two / IDefOpt / tpu / Forward 0.000001825975 s 0.000001831225 s 1.00
add_two / JaXPipe / tpu / PreRev 0.00000284175 s 0.000002835075 s 1.00
add_two / JaXPipe / tpu / PostRev 0.000002745 s 0.000002755875 s 1.00
add_two / JaXPipe / tpu / BothRev 0.0000028318500000000003 s 0.0000028421000000000003 s 1.00
add_two / Jax / tpu / BothRev 0.000002749075 s 0.000002751975 s 1.00
add_two / HLOOpt / tpu / PreRev 0.0000028364 s 0.0000028349 s 1.00
add_two / HLOOpt / tpu / PostRev 0.0000027504 s 0.0000027513250000000003 s 1.00
add_two / HLOOpt / tpu / BothRev 0.000002840675 s 0.000002831625 s 1.00
add_two / PartOpt / tpu / PreRev 0.0000027542250000000003 s 0.000002742475 s 1.00
add_two / PartOpt / tpu / PostRev 0.0000028429 s 0.00000283765 s 1.00
add_two / PartOpt / tpu / BothRev 0.0000027578000000000005 s 0.000002765825 s 1.00
add_two / IPartOpt / tpu / PreRev 0.00000283835 s 0.000002844575 s 1.00
add_two / IPartOpt / tpu / PostRev 0.00000275475 s 0.000002751 s 1.00
add_two / IPartOpt / tpu / BothRev 0.000002837825 s 0.00000283395 s 1.00
add_two / DefOpt / tpu / PreRev 0.000002749525 s 0.000002757975 s 1.00
add_two / DefOpt / tpu / PostRev 0.0000028314 s 0.0000028295749999999995 s 1.00
add_two / DefOpt / tpu / BothRev 0.0000027486 s 0.00000274965 s 1.00
add_two / IDefOpt / tpu / PreRev 0.00000283715 s 0.00000283355 s 1.00
add_two / IDefOpt / tpu / PostRev 0.000002753325 s 0.0000027563749999999995 s 1.00
add_two / IDefOpt / tpu / BothRev 0.00000283775 s 0.000002845875 s 1.00
add_two / JaXPipe / cpu / Primal 0.000013192 s 0.000006867799966130406 s 1.92
add_two / Jax / cpu / Primal 0.00001314 s 0.000006908039995323634 s 1.90
add_two / HLOOpt / cpu / Primal 0.000013448 s 0.000007006639998508036 s 1.92
add_two / PartOpt / cpu / Primal 0.000013142 s 0.000007027579995337874 s 1.87
add_two / IPartOpt / cpu / Primal 0.000013041 s 0.000007588979979118449 s 1.72
add_two / DefOpt / cpu / Primal 0.000013153 s 0.0000068686600297951375 s 1.91
add_two / IDefOpt / cpu / Primal 0.000013167 s 0.000006989980029175058 s 1.88
add_two / JaXPipe / cpu / Forward 0.000018313 s 0.000010030479979832308 s 1.83
add_two / Jax / cpu / Forward 0.000018104 s 0.000009928719991876278 s 1.82
add_two / HLOOpt / cpu / Forward 0.000017848 s 0.00001017656001749856 s 1.75
add_two / PartOpt / cpu / Forward 0.000018313 s 0.000010274779997416772 s 1.78
add_two / IPartOpt / cpu / Forward 0.000017822 s 0.000010454699986439663 s 1.70
add_two / DefOpt / cpu / Forward 0.000018012 s 0.000010264359962093295 s 1.75
add_two / IDefOpt / cpu / Forward 0.000017963 s 0.00001037967996126099 s 1.73
add_two / JaXPipe / cpu / PreRev 0.000024148 s 0.000013865560013073265 s 1.74
add_two / JaXPipe / cpu / PostRev 0.000023124 s 0.000014588599997296116 s 1.59
add_two / JaXPipe / cpu / BothRev 0.000024023 s 0.000014942600009817398 s 1.61
add_two / Jax / cpu / BothRev 0.00002309 s 0.000013541219987018849 s 1.71
add_two / HLOOpt / cpu / PreRev 0.000023796 s 0.000014437700037888137 s 1.65
add_two / HLOOpt / cpu / PostRev 0.000023576 s 0.00001636072001019784 s 1.44
add_two / HLOOpt / cpu / BothRev 0.000023373 s 0.000014554019971910747 s 1.61
add_two / PartOpt / cpu / PreRev 0.000024191 s 0.0000141802000325697 s 1.71
add_two / PartOpt / cpu / PostRev 0.00002295 s 0.0000140553999972326 s 1.63
add_two / PartOpt / cpu / BothRev 0.000023719 s 0.000014440060003835245 s 1.64
add_two / IPartOpt / cpu / PreRev 0.000024456 s 0.000014146200001050602 s 1.73
add_two / IPartOpt / cpu / PostRev 0.000023923 s 0.00001453995997508173 s 1.65
add_two / IPartOpt / cpu / BothRev 0.00002436 s 0.000014311239992821356 s 1.70
add_two / DefOpt / cpu / PreRev 0.000024082 s 0.000013563660022555269 s 1.78
add_two / DefOpt / cpu / PostRev 0.000023831000000000003 s 0.00001501925997217768 s 1.59
add_two / DefOpt / cpu / BothRev 0.000023588 s 0.00001491654000346898 s 1.58
add_two / IDefOpt / cpu / PreRev 0.000024589 s 0.00001355600003080326 s 1.81
add_two / IDefOpt / cpu / PostRev 0.000023677 s 0.000014451440019911388 s 1.64
add_two / IDefOpt / cpu / BothRev 0.000024209 s 0.000014129720002529213 s 1.71
add_two / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000006867799966130406 s 1.31
add_two / Jax / cpu / Primal 0.000008999999999999999 s 0.000006908039995323634 s 1.30
add_two / HLOOpt / cpu / Primal 0.000008 s 0.000007006639998508036 s 1.14
add_two / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000007027579995337874 s 1.28
add_two / IPartOpt / cpu / Primal 0.000008 s 0.000007588979979118449 s 1.05
add_two / DefOpt / cpu / Primal 0.000008999999999999999 s 0.0000068686600297951375 s 1.31
add_two / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000006989980029175058 s 1.29
add_two / JaXPipe / cpu / Forward 0.000013 s 0.000010030479979832308 s 1.30
add_two / Jax / cpu / Forward 0.000012 s 0.000009928719991876278 s 1.21
add_two / HLOOpt / cpu / Forward 0.000012 s 0.00001017656001749856 s 1.18
add_two / PartOpt / cpu / Forward 0.000012 s 0.000010274779997416772 s 1.17
add_two / IPartOpt / cpu / Forward 0.000012 s 0.000010454699986439663 s 1.15
add_two / DefOpt / cpu / Forward 0.000012 s 0.000010264359962093295 s 1.17
add_two / IDefOpt / cpu / Forward 0.000012 s 0.00001037967996126099 s 1.16
add_two / JaXPipe / cpu / PreRev 0.000016 s 0.000013865560013073265 s 1.15
add_two / JaXPipe / cpu / PostRev 0.000016 s 0.000014588599997296116 s 1.10
add_two / JaXPipe / cpu / BothRev 0.000016 s 0.000014942600009817398 s 1.07
add_two / Jax / cpu / BothRev 0.000016 s 0.000013541219987018849 s 1.18
add_two / HLOOpt / cpu / PreRev 0.000016 s 0.000014437700037888137 s 1.11
add_two / HLOOpt / cpu / PostRev 0.000017 s 0.00001636072001019784 s 1.04
add_two / HLOOpt / cpu / BothRev 0.000016 s 0.000014554019971910747 s 1.10
add_two / PartOpt / cpu / PreRev 0.000016 s 0.0000141802000325697 s 1.13
add_two / PartOpt / cpu / PostRev 0.000017 s 0.0000140553999972326 s 1.21
add_two / PartOpt / cpu / BothRev 0.000017 s 0.000014440060003835245 s 1.18
add_two / IPartOpt / cpu / PreRev 0.000017 s 0.000014146200001050602 s 1.20
add_two / IPartOpt / cpu / PostRev 0.000017 s 0.00001453995997508173 s 1.17
add_two / IPartOpt / cpu / BothRev 0.000017 s 0.000014311239992821356 s 1.19
add_two / DefOpt / cpu / PreRev 0.000017 s 0.000013563660022555269 s 1.25
add_two / DefOpt / cpu / PostRev 0.000051 s 0.00001501925997217768 s 3.40
add_two / DefOpt / cpu / BothRev 0.000017 s 0.00001491654000346898 s 1.14
add_two / IDefOpt / cpu / PreRev 0.000017 s 0.00001355600003080326 s 1.25
add_two / IDefOpt / cpu / PostRev 0.000017 s 0.000014451440019911388 s 1.18
add_two / IDefOpt / cpu / BothRev 0.000017 s 0.000014129720002529213 s 1.20
cache / JaXPipe / cpu / Primal 0.000006835900003352435 s 0.000006572259999302332 s 1.04
cache / Jax / cpu / Primal 0.000007040180007606978 s 0.000006501319994640653 s 1.08
cache / HLOOpt / cpu / Primal 0.000006756659968232271 s 0.00000615048002146068 s 1.10
cache / PartOpt / cpu / Primal 0.000006869119997645612 s 0.000006232679988897871 s 1.10
cache / IPartOpt / cpu / Primal 0.00000657152003441297 s 0.000006268780007303576 s 1.05
cache / DefOpt / cpu / Primal 0.00000701981999554846 s 0.000006588100022781873 s 1.07
cache / IDefOpt / cpu / Primal 0.000007289320019481238 s 0.000006464800007961458 s 1.13
cache / JaXPipe / cpu / Forward 0.00001537807999739016 s 0.000014711600006194204 s 1.05
cache / Jax / cpu / Forward 0.000014446060022237362 s 0.00001440755996554799 s 1.00
cache / HLOOpt / cpu / Forward 0.00001559604000249237 s 0.00001561347999086138 s 1.00
cache / PartOpt / cpu / Forward 0.000014526159984598051 s 0.000014749180008948317 s 0.98
cache / IPartOpt / cpu / Forward 0.000014918300012141115 s 0.000016127100006997353 s 0.93
cache / DefOpt / cpu / Forward 0.000015244140031427378 s 0.000015200499992715775 s 1.00
cache / IDefOpt / cpu / Forward 0.000015149060000112512 s 0.000015227600015350615 s 0.99
cache / JaXPipe / cpu / PreRev 0.000016188099953069467 s 0.000016161640023710787 s 1.00
cache / JaXPipe / cpu / PostRev 0.00002090654002131487 s 0.000021423939997475828 s 0.98
cache / JaXPipe / cpu / BothRev 0.00001634539998121909 s 0.00001640085999497387 s 1.00
cache / Jax / cpu / BothRev 0.00002091987997118849 s 0.000021007039958931272 s 1.00
cache / HLOOpt / cpu / PreRev 0.00001634156001273368 s 0.000016289080003843993 s 1.00
cache / HLOOpt / cpu / PostRev 0.000018052619989248345 s 0.000019273419984529027 s 0.94
cache / HLOOpt / cpu / BothRev 0.000016606460021648673 s 0.000017125759968621422 s 0.97
cache / PartOpt / cpu / PreRev 0.00001606105995051621 s 0.000016171379975276068 s 0.99
cache / PartOpt / cpu / PostRev 0.000020910839948555805 s 0.000019693940039360317 s 1.06
cache / PartOpt / cpu / BothRev 0.000016004479957700822 s 0.000016042779989220435 s 1.00
cache / IPartOpt / cpu / PreRev 0.000016057220009315642 s 0.000015654539993192884 s 1.03
cache / IPartOpt / cpu / PostRev 0.000020891899994239792 s 0.00001996903996769106 s 1.05
cache / IPartOpt / cpu / BothRev 0.00001739608005664195 s 0.000015126959997360244 s 1.15
cache / DefOpt / cpu / PreRev 0.000016460439992442845 s 0.000015228679976644345 s 1.08
cache / DefOpt / cpu / PostRev 0.00001647531992603035 s 0.000015360740017058562 s 1.07
cache / DefOpt / cpu / BothRev 0.00001636726002288924 s 0.000015114279985937172 s 1.08
cache / IDefOpt / cpu / PreRev 0.00001765007999892987 s 0.000015915160092845325 s 1.11
cache / IDefOpt / cpu / PostRev 0.000016989580044537432 s 0.000016638339993733097 s 1.02
cache / IDefOpt / cpu / BothRev 0.000017095980010708444 s 0.000016164739990927045 s 1.06
cache / JaXPipe / cuda / Primal 0.000002336 s 0.000002303 s 1.01
cache / Jax / cuda / Primal 0.0000023670000000000004 s 0.000002304 s 1.03
cache / HLOOpt / cuda / Primal 0.000002272 s 0.00000224 s 1.01
cache / PartOpt / cuda / Primal 0.000002304 s 0.00000224 s 1.03
cache / IPartOpt / cuda / Primal 0.000002336 s 0.000002303 s 1.01
cache / DefOpt / cuda / Primal 0.000002304 s 0.00000224 s 1.03
cache / IDefOpt / cuda / Primal 0.000002272 s 0.000002208 s 1.03
cache / JaXPipe / cuda / Forward 0.000002368 s 0.000002335 s 1.01
cache / Jax / cuda / Forward 0.000002368 s 0.000002304 s 1.03
cache / HLOOpt / cuda / Forward 0.0000023670000000000004 s 0.000002335 s 1.01
cache / PartOpt / cuda / Forward 0.000002368 s 0.000002335 s 1.01
cache / IPartOpt / cuda / Forward 0.000002368 s 0.000002335 s 1.01
cache / DefOpt / cuda / Forward 0.000002303 s 0.000002272 s 1.01
cache / IDefOpt / cuda / Forward 0.0000023670000000000004 s 0.000002335 s 1.01
cache / JaXPipe / cuda / PreRev 0.000010849 s 0.000009984 s 1.09
cache / JaXPipe / cuda / PostRev 0.000010784 s 0.000010528 s 1.02
cache / JaXPipe / cuda / BothRev 0.000011008 s 0.0000104 s 1.06
cache / Jax / cuda / BothRev 0.000011008 s 0.000010656 s 1.03
cache / HLOOpt / cuda / PreRev 0.000013536 s 0.000013504 s 1.00
cache / HLOOpt / cuda / PostRev 0.000013536 s 0.000013536 s 1
cache / HLOOpt / cuda / BothRev 0.000013535 s 0.000013536 s 1.00
cache / PartOpt / cuda / PreRev 0.000011648 s 0.000011745 s 0.99
cache / PartOpt / cuda / PostRev 0.000011104 s 0.000011520000000000002 s 0.96
cache / PartOpt / cuda / BothRev 0.000011424 s 0.000010463 s 1.09
cache / IPartOpt / cuda / PreRev 0.000011392 s 0.000010624 s 1.07
cache / IPartOpt / cuda / PostRev 0.000010944 s 0.000010432 s 1.05
cache / IPartOpt / cuda / BothRev 0.000010977 s 0.0000104 s 1.06
cache / DefOpt / cuda / PreRev 0.000011297 s 0.00001072 s 1.05
cache / DefOpt / cuda / PostRev 0.000011104 s 0.000010336 s 1.07
cache / DefOpt / cuda / BothRev 0.000011008 s 0.000010752 s 1.02
cache / IDefOpt / cuda / PreRev 0.000010816 s 0.000010591 s 1.02
cache / IDefOpt / cuda / PostRev 0.000011008 s 0.000010304 s 1.07
cache / IDefOpt / cuda / BothRev 0.000010529 s 0.000010464 s 1.01
cache / JaXPipe / tpu / Primal 0.000002464 s 0.0000024717 s 1.00
cache / Jax / tpu / Primal 0.00000246485 s 0.000002463125 s 1.00
cache / HLOOpt / tpu / Primal 0.0000024720250000000003 s 0.000002479075 s 1.00
cache / PartOpt / tpu / Primal 0.000002448025 s 0.00000247065 s 0.99
cache / IPartOpt / tpu / Primal 0.000002458 s 0.0000024608 s 1.00
cache / DefOpt / tpu / Primal 0.000002457525 s 0.0000024597 s 1.00
cache / IDefOpt / tpu / Primal 0.000002463025 s 0.000002462275 s 1.00
cache / JaXPipe / tpu / Forward 0.000003549125 s 0.000003541175 s 1.00
cache / Jax / tpu / Forward 0.0000035601 s 0.000003548125 s 1.00
cache / HLOOpt / tpu / Forward 0.000003578825 s 0.000003560675 s 1.01
cache / PartOpt / tpu / Forward 0.00000353345 s 0.000003532225 s 1.00
cache / IPartOpt / tpu / Forward 0.0000035606500000000004 s 0.00000355105 s 1.00
cache / DefOpt / tpu / Forward 0.000003558325 s 0.0000035362 s 1.01
cache / IDefOpt / tpu / Forward 0.0000035610000000000003 s 0.000003544875 s 1.00
cache / JaXPipe / tpu / PreRev 0.000004959475 s 0.00000495055 s 1.00
cache / JaXPipe / tpu / PostRev 0.000004942550000000001 s 0.00000496035 s 1.00
cache / JaXPipe / tpu / BothRev 0.00000499495 s 0.000004976425 s 1.00
cache / Jax / tpu / BothRev 0.0000049708 s 0.000004994475 s 1.00
cache / HLOOpt / tpu / PreRev 0.000003951925 s 0.0000039297 s 1.01
cache / HLOOpt / tpu / PostRev 0.000004114825 s 0.000004137525000000001 s 0.99
cache / HLOOpt / tpu / BothRev 0.000003943675 s 0.0000039311 s 1.00
cache / PartOpt / tpu / PreRev 0.000004993275 s 0.0000049712 s 1.00
cache / PartOpt / tpu / PostRev 0.000004969700000000001 s 0.000004976825 s 1.00
cache / PartOpt / tpu / BothRev 0.0000049761 s 0.00000496935 s 1.00
cache / IPartOpt / tpu / PreRev 0.0000049814 s 0.000004975975 s 1.00
cache / IPartOpt / tpu / PostRev 0.000004969875 s 0.0000049639 s 1.00
cache / IPartOpt / tpu / BothRev 0.0000049627 s 0.0000049710750000000005 s 1.00
cache / DefOpt / tpu / PreRev 0.000004966 s 0.000004988775 s 1.00
cache / DefOpt / tpu / PostRev 0.00000496735 s 0.00000495615 s 1.00
cache / DefOpt / tpu / BothRev 0.00000495815 s 0.00000498125 s 1.00
cache / IDefOpt / tpu / PreRev 0.0000049830500000000005 s 0.000004979 s 1.00
cache / IDefOpt / tpu / PostRev 0.00000497565 s 0.000004971375 s 1.00
cache / IDefOpt / tpu / BothRev 0.00000498555 s 0.000004941475 s 1.01
cache / JaXPipe / cpu / Primal 0.000012777 s 0.000006572259999302332 s 1.94
cache / Jax / cpu / Primal 0.000012644 s 0.000006501319994640653 s 1.94
cache / HLOOpt / cpu / Primal 0.000012705 s 0.00000615048002146068 s 2.07
cache / PartOpt / cpu / Primal 0.000012297 s 0.000006232679988897871 s 1.97
cache / IPartOpt / cpu / Primal 0.00001231 s 0.000006268780007303576 s 1.96
cache / DefOpt / cpu / Primal 0.000012496 s 0.000006588100022781873 s 1.90
cache / IDefOpt / cpu / Primal 0.000012515 s 0.000006464800007961458 s 1.94
cache / JaXPipe / cpu / Forward 0.000024018000000000003 s 0.000014711600006194204 s 1.63
cache / Jax / cpu / Forward 0.000024995 s 0.00001440755996554799 s 1.73
cache / HLOOpt / cpu / Forward 0.000017057 s 0.00001561347999086138 s 1.09
cache / PartOpt / cpu / Forward 0.000016593 s 0.000014749180008948317 s 1.13
cache / IPartOpt / cpu / Forward 0.000016854 s 0.000016127100006997353 s 1.05
cache / DefOpt / cpu / Forward 0.000017284 s 0.000015200499992715775 s 1.14
cache / IDefOpt / cpu / Forward 0.000025163000000000003 s 0.000015227600015350615 s 1.65
cache / JaXPipe / cpu / PreRev 0.00002357 s 0.000016161640023710787 s 1.46
cache / JaXPipe / cpu / PostRev 0.000030297 s 0.000021423939997475828 s 1.41
cache / JaXPipe / cpu / BothRev 0.000026197 s 0.00001640085999497387 s 1.60
cache / Jax / cpu / BothRev 0.000033052 s 0.000021007039958931272 s 1.57
cache / HLOOpt / cpu / PreRev 0.000030746 s 0.000016289080003843993 s 1.89
cache / HLOOpt / cpu / PostRev 0.000027282 s 0.000019273419984529027 s 1.42
cache / HLOOpt / cpu / BothRev 0.000031953000000000004 s 0.000017125759968621422 s 1.87
cache / PartOpt / cpu / PreRev 0.000025775 s 0.000016171379975276068 s 1.59
cache / PartOpt / cpu / PostRev 0.000035183 s 0.000019693940039360317 s 1.79
cache / PartOpt / cpu / BothRev 0.000025125 s 0.000016042779989220435 s 1.57
cache / IPartOpt / cpu / PreRev 0.000026972 s 0.000015654539993192884 s 1.72
cache / IPartOpt / cpu / PostRev 0.000036011000000000005 s 0.00001996903996769106 s 1.80
cache / IPartOpt / cpu / BothRev 0.000017860999999999997 s 0.000015126959997360244 s 1.18
cache / DefOpt / cpu / PreRev 0.000028718 s 0.000015228679976644345 s 1.89
cache / DefOpt / cpu / PostRev 0.000028055 s 0.000015360740017058562 s 1.83
cache / DefOpt / cpu / BothRev 0.000022313 s 0.000015114279985937172 s 1.48
cache / IDefOpt / cpu / PreRev 0.000024726 s 0.000015915160092845325 s 1.55
cache / IDefOpt / cpu / PostRev 0.000024749 s 0.000016638339993733097 s 1.49
cache / IDefOpt / cpu / BothRev 0.000018362 s 0.000016164739990927045 s 1.14
cache / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000006572259999302332 s 1.37
cache / Jax / cpu / Primal 0.000008999999999999999 s 0.000006501319994640653 s 1.38
cache / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.00000615048002146068 s 1.46
cache / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000006232679988897871 s 1.44
cache / IPartOpt / cpu / Primal 0.000008999999999999999 s 0.000006268780007303576 s 1.44
cache / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000006588100022781873 s 1.37
cache / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000006464800007961458 s 1.39
cache / JaXPipe / cpu / Forward 0.000011 s 0.000014711600006194204 s 0.75
cache / Jax / cpu / Forward 0.000011 s 0.00001440755996554799 s 0.76
cache / HLOOpt / cpu / Forward 0.00001 s 0.00001561347999086138 s 0.64
cache / PartOpt / cpu / Forward 0.000011 s 0.000014749180008948317 s 0.75
cache / IPartOpt / cpu / Forward 0.000023 s 0.000016127100006997353 s 1.43
cache / DefOpt / cpu / Forward 0.000011 s 0.000015200499992715775 s 0.72
cache / IDefOpt / cpu / Forward 0.000011 s 0.000015227600015350615 s 0.72
cache / JaXPipe / cpu / PreRev 0.000011 s 0.000016161640023710787 s 0.68
cache / JaXPipe / cpu / PostRev 0.000011 s 0.000021423939997475828 s 0.51
cache / JaXPipe / cpu / BothRev 0.000011 s 0.00001640085999497387 s 0.67
cache / Jax / cpu / BothRev 0.000011 s 0.000021007039958931272 s 0.52
cache / HLOOpt / cpu / PreRev 0.000011 s 0.000016289080003843993 s 0.68
cache / HLOOpt / cpu / PostRev 0.000037 s 0.000019273419984529027 s 1.92
cache / HLOOpt / cpu / BothRev 0.000012 s 0.000017125759968621422 s 0.70
cache / PartOpt / cpu / PreRev 0.000011 s 0.000016171379975276068 s 0.68
cache / PartOpt / cpu / PostRev 0.000035999999999999994 s 0.000019693940039360317 s 1.83
cache / PartOpt / cpu / BothRev 0.000011 s 0.000016042779989220435 s 0.69
cache / IPartOpt / cpu / PreRev 0.000011 s 0.000015654539993192884 s 0.70
cache / IPartOpt / cpu / PostRev 0.000039 s 0.00001996903996769106 s 1.95
cache / IPartOpt / cpu / BothRev 0.000011 s 0.000015126959997360244 s 0.73
cache / DefOpt / cpu / PreRev 0.000011 s 0.000015228679976644345 s 0.72
cache / DefOpt / cpu / PostRev 0.000011 s 0.000015360740017058562 s 0.72
cache / DefOpt / cpu / BothRev 0.000011 s 0.000015114279985937172 s 0.73
cache / IDefOpt / cpu / PreRev 0.000011 s 0.000015915160092845325 s 0.69
cache / IDefOpt / cpu / PostRev 0.000015 s 0.000016638339993733097 s 0.90
cache / IDefOpt / cpu / BothRev 0.000011 s 0.000016164739990927045 s 0.68
Concat / JaXPipe / cpu / Primal 0.0000072969599477801235 s 0.000007459219978045439 s 0.98
Concat / Jax / cpu / Primal 0.000008260900003733695 s 0.0000072401800116495 s 1.14
Concat / HLOOpt / cpu / Primal 0.000007134419956855709 s 0.0000072882999938883585 s 0.98
Concat / PartOpt / cpu / Primal 0.000007362760006799363 s 0.000007322039991777274 s 1.01
Concat / IPartOpt / cpu / Primal 0.000007141939968278166 s 0.000006696399968859623 s 1.07
Concat / DefOpt / cpu / Primal 0.000007510720024583861 s 0.000006468439987656894 s 1.16
Concat / IDefOpt / cpu / Primal 0.000007224419969134032 s 0.000006598119989575935 s 1.09
Concat / JaXPipe / cpu / Forward 0.00001141933998951572 s 0.000009809859993765712 s 1.16
Concat / Jax / cpu / Forward 0.000010927760013146324 s 0.000010438159988552795 s 1.05
Concat / HLOOpt / cpu / Forward 0.000011138319996462089 s 0.000010427059996800382 s 1.07
Concat / PartOpt / cpu / Forward 0.000011108959979537758 s 0.00001027719998091925 s 1.08
Concat / IPartOpt / cpu / Forward 0.000011096420012108864 s 0.000010685560009733309 s 1.04
Concat / DefOpt / cpu / Forward 0.000011038080010621345 s 0.000010202059975199518 s 1.08
Concat / IDefOpt / cpu / Forward 0.000010976559988193912 s 0.00000999803997729032 s 1.10
Concat / JaXPipe / cpu / PreRev 0.000012555639987112954 s 0.000012138919973949667 s 1.03
Concat / JaXPipe / cpu / PostRev 0.0000127675999647181 s 0.00001107852001041465 s 1.15
Concat / JaXPipe / cpu / BothRev 0.000012911479989270448 s 0.000010961260022668284 s 1.18
Concat / Jax / cpu / BothRev 0.00001226989996212069 s 0.00001153500001237262 s 1.06
Concat / HLOOpt / cpu / PreRev 0.000012453680010366952 s 0.000012123960059398087 s 1.03
Concat / HLOOpt / cpu / PostRev 0.000014550600044458406 s 0.00001364110003123642 s 1.07
Concat / HLOOpt / cpu / BothRev 0.00001251344004231214 s 0.000010911959998338716 s 1.15
Concat / PartOpt / cpu / PreRev 0.000012512799985415768 s 0.00001119947999541182 s 1.12
Concat / PartOpt / cpu / PostRev 0.000012610860003405832 s 0.000011387060012566508 s 1.11
Concat / PartOpt / cpu / BothRev 0.000012970619991392596 s 0.000012363559999357677 s 1.05
Concat / IPartOpt / cpu / PreRev 0.000012577560019053635 s 0.000012330119961916352 s 1.02
Concat / IPartOpt / cpu / PostRev 0.000012699599983534426 s 0.000011503100004119916 s 1.10
Concat / IPartOpt / cpu / BothRev 0.00001243582000824972 s 0.000011435539990998222 s 1.09
Concat / DefOpt / cpu / PreRev 0.000012569259997690096 s 0.000011610259998633407 s 1.08
Concat / DefOpt / cpu / PostRev 0.000013453339988700463 s 0.000011580839955058763 s 1.16
Concat / DefOpt / cpu / BothRev 0.000012422979962138925 s 0.000011326979984005448 s 1.10
Concat / IDefOpt / cpu / PreRev 0.000012357260002318072 s 0.000011609920002229044 s 1.06
Concat / IDefOpt / cpu / PostRev 0.000012732999985018975 s 0.000011728899999070565 s 1.09
Concat / IDefOpt / cpu / BothRev 0.000012817399938285234 s 0.0000115970599836146 s 1.11
Concat / JaXPipe / cuda / Primal 0.0000019200000000000003 s 0.000001888 s 1.02
Concat / Jax / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / HLOOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / PartOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / IPartOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / DefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / IDefOpt / cuda / Primal 0.0000019200000000000003 s 0.0000019200000000000003 s 1
Concat / JaXPipe / cuda / Forward 0.000010176 s 0.000009887 s 1.03
Concat / Jax / cuda / Forward 0.000010209 s 0.000010176 s 1.00
Concat / HLOOpt / cuda / Forward 0.000010207 s 0.000009824 s 1.04
Concat / PartOpt / cuda / Forward 0.000010112 s 0.000009631 s 1.05
Concat / IPartOpt / cuda / Forward 0.000010176 s 0.000010304 s 0.99
Concat / DefOpt / cuda / Forward 0.000010241 s 0.000010048 s 1.02
Concat / IDefOpt / cuda / Forward 0.000010176 s 0.00000944 s 1.08
Concat / JaXPipe / cuda / PreRev 0.000016737 s 0.000016545 s 1.01
Concat / JaXPipe / cuda / PostRev 0.000016672 s 0.00001584 s 1.05
Concat / JaXPipe / cuda / BothRev 0.000017216 s 0.00001632 s 1.05
Concat / Jax / cuda / BothRev 0.000016929 s 0.000016032 s 1.06
Concat / HLOOpt / cuda / PreRev 0.00001712 s 0.000015968 s 1.07
Concat / HLOOpt / cuda / PostRev 0.000016288 s 0.00001584 s 1.03
Concat / HLOOpt / cuda / BothRev 0.000017088 s 0.000016576000000000002 s 1.03
Concat / PartOpt / cuda / PreRev 0.000016991 s 0.000016 s 1.06
Concat / PartOpt / cuda / PostRev 0.0000168 s 0.000016063999999999997 s 1.05
Concat / PartOpt / cuda / BothRev 0.000016864 s 0.000016288 s 1.04
Concat / IPartOpt / cuda / PreRev 0.00001712 s 0.000016416 s 1.04
Concat / IPartOpt / cuda / PostRev 0.000016927999999999998 s 0.00001584 s 1.07
Concat / IPartOpt / cuda / BothRev 0.000016864 s 0.000016 s 1.05
Concat / DefOpt / cuda / PreRev 0.000017151 s 0.000016192 s 1.06
Concat / DefOpt / cuda / PostRev 0.000017087 s 0.000016511 s 1.03
Concat / DefOpt / cuda / BothRev 0.000016864 s 0.00001584 s 1.06
Concat / IDefOpt / cuda / PreRev 0.000017056 s 0.00001616 s 1.06
Concat / IDefOpt / cuda / PostRev 0.00001712 s 0.000015712 s 1.09
Concat / IDefOpt / cuda / BothRev 0.000016993 s 0.000016608 s 1.02
Concat / JaXPipe / tpu / Primal 0.000001537025 s 0.0000014826 s 1.04
Concat / Jax / tpu / Primal 0.000001520225 s 0.0000014854999999999998 s 1.02
Concat / HLOOpt / tpu / Primal 0.0000015407499999999998 s 0.0000014761 s 1.04
Concat / PartOpt / tpu / Primal 0.0000015239249999999998 s 0.0000014868 s 1.02
Concat / IPartOpt / tpu / Primal 0.000001534975 s 0.000001476675 s 1.04
Concat / DefOpt / tpu / Primal 0.000001527025 s 0.0000014921 s 1.02
Concat / IDefOpt / tpu / Primal 0.000001533525 s 0.0000014804 s 1.04
Concat / JaXPipe / tpu / Forward 0.00000157035 s 0.0000015552 s 1.01
Concat / Jax / tpu / Forward 0.0000015491 s 0.000001504375 s 1.03
Concat / HLOOpt / tpu / Forward 0.000001573175 s 0.00000154035 s 1.02
Concat / PartOpt / tpu / Forward 0.000001556725 s 0.00000151385 s 1.03
Concat / IPartOpt / tpu / Forward 0.00000156855 s 0.000001547175 s 1.01
Concat / DefOpt / tpu / Forward 0.0000015471999999999998 s 0.000001514675 s 1.02
Concat / IDefOpt / tpu / Forward 0.000001572 s 0.000001551225 s 1.01
Concat / JaXPipe / tpu / PreRev 0.0000020152500000000003 s 0.0000019532250000000003 s 1.03
Concat / JaXPipe / tpu / PostRev 0.000002086075 s 0.000002041625 s 1.02
Concat / JaXPipe / tpu / BothRev 0.000002009925 s 0.0000019443750000000003 s 1.03
Concat / Jax / tpu / BothRev 0.0000020720500000000003 s 0.000002032475 s 1.02
Concat / HLOOpt / tpu / PreRev 0.0000020186 s 0.0000019443750000000003 s 1.04
Concat / HLOOpt / tpu / PostRev 0.00000206835 s 0.000002035425 s 1.02
Concat / HLOOpt / tpu / BothRev 0.00000200685 s 0.00000195195 s 1.03
Concat / PartOpt / tpu / PreRev 0.000002077575 s 0.000002037175 s 1.02
Concat / PartOpt / tpu / PostRev 0.000002003525 s 0.000001951725 s 1.03
Concat / PartOpt / tpu / BothRev 0.000002065475 s 0.0000020347 s 1.02
Concat / IPartOpt / tpu / PreRev 0.00000200435 s 0.00000195195 s 1.03
Concat / IPartOpt / tpu / PostRev 0.000002080375 s 0.0000020362 s 1.02
Concat / IPartOpt / tpu / BothRev 0.000002016525 s 0.00000194775 s 1.04
Concat / DefOpt / tpu / PreRev 0.000002071775 s 0.0000020317 s 1.02
Concat / DefOpt / tpu / PostRev 0.000002004475 s 0.000001952 s 1.03
Concat / DefOpt / tpu / BothRev 0.000002076475 s 0.000002034425 s 1.02
Concat / IDefOpt / tpu / PreRev 0.0000020074 s 0.00000194375 s 1.03
Concat / IDefOpt / tpu / PostRev 0.00000207185 s 0.00000203705 s 1.02
Concat / IDefOpt / tpu / BothRev 0.0000020083500000000004 s 0.000001943275 s 1.03
Concat / JaXPipe / cpu / Primal 0.00001243 s 0.000007459219978045439 s 1.67
Concat / Jax / cpu / Primal 0.000012667 s 0.0000072401800116495 s 1.75
Concat / HLOOpt / cpu / Primal 0.000012746 s 0.0000072882999938883585 s 1.75
Concat / PartOpt / cpu / Primal 0.000012437 s 0.000007322039991777274 s 1.70
Concat / IPartOpt / cpu / Primal 0.000012789 s 0.000006696399968859623 s 1.91
Concat / DefOpt / cpu / Primal 0.000012689 s 0.000006468439987656894 s 1.96
Concat / IDefOpt / cpu / Primal 0.000012749 s 0.000006598119989575935 s 1.93
Concat / JaXPipe / cpu / Forward 0.000018327 s 0.000009809859993765712 s 1.87
Concat / Jax / cpu / Forward 0.000017466 s 0.000010438159988552795 s 1.67
Concat / HLOOpt / cpu / Forward 0.00001734 s 0.000010427059996800382 s 1.66
Concat / PartOpt / cpu / Forward 0.000017754 s 0.00001027719998091925 s 1.73
Concat / IPartOpt / cpu / Forward 0.000017912 s 0.000010685560009733309 s 1.68
Concat / DefOpt / cpu / Forward 0.000017364000000000002 s 0.000010202059975199518 s 1.70
Concat / IDefOpt / cpu / Forward 0.000017714 s 0.00000999803997729032 s 1.77
Concat / JaXPipe / cpu / PreRev 0.000020572 s 0.000012138919973949667 s 1.69
Concat / JaXPipe / cpu / PostRev 0.000019322 s 0.00001107852001041465 s 1.74
Concat / JaXPipe / cpu / BothRev 0.000019689 s 0.000010961260022668284 s 1.80
Concat / Jax / cpu / BothRev 0.000019904 s 0.00001153500001237262 s 1.73
Concat / HLOOpt / cpu / PreRev 0.000020015 s 0.000012123960059398087 s 1.65
Concat / HLOOpt / cpu / PostRev 0.00002029 s 0.00001364110003123642 s 1.49
Concat / HLOOpt / cpu / BothRev 0.000020131 s 0.000010911959998338716 s 1.84
Concat / PartOpt / cpu / PreRev 0.00002025 s 0.00001119947999541182 s 1.81
Concat / PartOpt / cpu / PostRev 0.000019779 s 0.000011387060012566508 s 1.74
Concat / PartOpt / cpu / BothRev 0.000020224 s 0.000012363559999357677 s 1.64
Concat / IPartOpt / cpu / PreRev 0.000019883000000000003 s 0.000012330119961916352 s 1.61
Concat / IPartOpt / cpu / PostRev 0.000020278 s 0.000011503100004119916 s 1.76
Concat / IPartOpt / cpu / BothRev 0.000019814 s 0.000011435539990998222 s 1.73
Concat / DefOpt / cpu / PreRev 0.000020402 s 0.000011610259998633407 s 1.76
Concat / DefOpt / cpu / PostRev 0.000019361 s 0.000011580839955058763 s 1.67
Concat / DefOpt / cpu / BothRev 0.000019534 s 0.000011326979984005448 s 1.72
Concat / IDefOpt / cpu / PreRev 0.000020429 s 0.000011609920002229044 s 1.76
Concat / IDefOpt / cpu / PostRev 0.000019933 s 0.000011728899999070565 s 1.70
Concat / IDefOpt / cpu / BothRev 0.000020074 s 0.0000115970599836146 s 1.73
Concat / JaXPipe / cpu / Primal 0.000008999999999999999 s 0.000007459219978045439 s 1.21
Concat / Jax / cpu / Primal 0.000008 s 0.0000072401800116495 s 1.10
Concat / HLOOpt / cpu / Primal 0.000008 s 0.0000072882999938883585 s 1.10
Concat / PartOpt / cpu / Primal 0.000008 s 0.000007322039991777274 s 1.09
Concat / IPartOpt / cpu / Primal 0.000008999999999999999 s 0.000006696399968859623 s 1.34
Concat / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000006468439987656894 s 1.39
Concat / IDefOpt / cpu / Primal 0.000008 s 0.000006598119989575935 s 1.21
Concat / JaXPipe / cpu / Forward 0.000012 s 0.000009809859993765712 s 1.22
Concat / Jax / cpu / Forward 0.000012 s 0.000010438159988552795 s 1.15
Concat / HLOOpt / cpu / Forward 0.000012 s 0.000010427059996800382 s 1.15
Concat / PartOpt / cpu / Forward 0.000012 s 0.00001027719998091925 s 1.17
Concat / IPartOpt / cpu / Forward 0.000012 s 0.000010685560009733309 s 1.12
Concat / DefOpt / cpu / Forward 0.000012 s 0.000010202059975199518 s 1.18
Concat / IDefOpt / cpu / Forward 0.000012 s 0.00000999803997729032 s 1.20
Concat / JaXPipe / cpu / PreRev 0.000013 s 0.000012138919973949667 s 1.07
Concat / JaXPipe / cpu / PostRev 0.000013 s 0.00001107852001041465 s 1.17
Concat / JaXPipe / cpu / BothRev 0.000014 s 0.000010961260022668284 s 1.28
Concat / Jax / cpu / BothRev 0.000013 s 0.00001153500001237262 s 1.13
Concat / HLOOpt / cpu / PreRev 0.000013 s 0.000012123960059398087 s 1.07
Concat / HLOOpt / cpu / PostRev 0.000013 s 0.00001364110003123642 s 0.95
Concat / HLOOpt / cpu / BothRev 0.000013 s 0.000010911959998338716 s 1.19
Concat / PartOpt / cpu / PreRev 0.000014 s 0.00001119947999541182 s 1.25
Concat / PartOpt / cpu / PostRev 0.000014 s 0.000011387060012566508 s 1.23
Concat / PartOpt / cpu / BothRev 0.000015 s 0.000012363559999357677 s 1.21
Concat / IPartOpt / cpu / PreRev 0.000014 s 0.000012330119961916352 s 1.14
Concat / IPartOpt / cpu / PostRev 0.000014 s 0.000011503100004119916 s 1.22
Concat / IPartOpt / cpu / BothRev 0.000014 s 0.000011435539990998222 s 1.22
Concat / DefOpt / cpu / PreRev 0.000014 s 0.000011610259998633407 s 1.21
Concat / DefOpt / cpu / PostRev 0.000014 s 0.000011580839955058763 s 1.21
Concat / DefOpt / cpu / BothRev 0.000014 s 0.000011326979984005448 s 1.24
Concat / IDefOpt / cpu / PreRev 0.000014 s 0.000011609920002229044 s 1.21
Concat / IDefOpt / cpu / PostRev 0.000014 s 0.000011728899999070565 s 1.19
Concat / IDefOpt / cpu / BothRev 0.000014 s 0.0000115970599836146 s 1.21
const_scatter / JaXPipe / cpu / Primal 0.000006609980036955676 s 0.000006414199997379911 s 1.03
const_scatter / Jax / cpu / Primal 0.0000070419999428850136 s 0.000006381759994837921 s 1.10
const_scatter / HLOOpt / cpu / Primal 0.000007514640019508079 s 0.000007092419973560027 s 1.06
const_scatter / PartOpt / cpu / Primal 0.000006942040008652839 s 0.000006388200008586864 s 1.09
const_scatter / IPartOpt / cpu / Primal 0.000006966119963180973 s 0.000007287040007213363 s 0.96
const_scatter / DefOpt / cpu / Primal 0.000007272399971043342 s 0.000006763159999536583 s 1.08
const_scatter / IDefOpt / cpu / Primal 0.000007624060008311062 s 0.000007052119999571005 s 1.08
const_scatter / JaXPipe / cpu / Forward 0.00001145965997238818 s 0.000010790360001919908 s 1.06
const_scatter / Jax / cpu / Forward 0.000010578720011835683 s 0.000009555260021443246 s 1.11
const_scatter / HLOOpt / cpu / Forward 0.00001196986000650213 s 0.000011073819969169565 s 1.08
const_scatter / PartOpt / cpu / Forward 0.000011872019986185478 s 0.000010634000009304144 s 1.12
const_scatter / IPartOpt / cpu / Forward 0.000011645619952105336 s 0.000010531140032981057 s 1.11
const_scatter / DefOpt / cpu / Forward 0.000011462500024208568 s 0.000010720419995777774 s 1.07
const_scatter / IDefOpt / cpu / Forward 0.00001136234001023695 s 0.000010679459992388729 s 1.06
const_scatter / JaXPipe / cpu / PreRev 0.000288531199958 s 0.0002888223200261 s 1.00
const_scatter / JaXPipe / cpu / PostRev 0.0002812392600117 s 0.0002839717799724 s 0.99
const_scatter / JaXPipe / cpu / BothRev 0.000281834239995 s 0.0002854160999595 s 0.99
const_scatter / Jax / cpu / BothRev 0.0002810300999954 s 0.0002842025400059 s 0.99
const_scatter / HLOOpt / cpu / PreRev 0.0002870756799893 s 0.0002855643800376 s 1.01
const_scatter / HLOOpt / cpu / PostRev 0.0002851579999969 s 0.0002889120800318 s 0.99
const_scatter / HLOOpt / cpu / BothRev 0.0002815610800007 s 0.0002857153400054 s 0.99
const_scatter / PartOpt / cpu / PreRev 0.0002823165000245 s 0.000283256499988 s 1.00
const_scatter / PartOpt / cpu / PostRev 0.0002836064599523 s 0.0002831990800314 s 1.00
const_scatter / PartOpt / cpu / BothRev 0.0002832542199485 s 0.0002844155800357 s 1.00
const_scatter / IPartOpt / cpu / PreRev 0.0002811633599685 s 0.0002835344200138 s 0.99
const_scatter / IPartOpt / cpu / PostRev 0.0002800351599944 s 0.0002824978799799 s 0.99
const_scatter / IPartOpt / cpu / BothRev 0.0002835342199887 s 0.0002845528399939 s 1.00
const_scatter / DefOpt / cpu / PreRev 0.0002838122399589 s 0.0002848829400136 s 1.00
const_scatter / DefOpt / cpu / PostRev 0.0002827460200023 s 0.0002848792600161 s 0.99
const_scatter / DefOpt / cpu / BothRev 0.0002820648600118 s 0.0002847787200244 s 0.99
const_scatter / IDefOpt / cpu / PreRev 0.0002833821599779 s 0.0002828045599926 s 1.00
const_scatter / IDefOpt / cpu / PostRev 0.000283653760016 s 0.0002841483999873 s 1.00
const_scatter / IDefOpt / cpu / BothRev 0.0002832252799998 s 0.0002834731599705 s 1.00
const_scatter / JaXPipe / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / Jax / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / HLOOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / PartOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
const_scatter / IPartOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / DefOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / IDefOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
const_scatter / JaXPipe / cuda / Forward 0.000010144 s 0.000009632 s 1.05
const_scatter / Jax / cuda / Forward 0.000010016 s 0.00001008 s 0.99
const_scatter / HLOOpt / cuda / Forward 0.000009568 s 0.000010016 s 0.96
const_scatter / PartOpt / cuda / Forward 0.000009824 s 0.000009568 s 1.03
const_scatter / IPartOpt / cuda / Forward 0.000010304 s 0.000010081 s 1.02
const_scatter / DefOpt / cuda / Forward 0.000010016 s 0.000010048 s 1.00
const_scatter / IDefOpt / cuda / Forward 0.000009856 s 0.000010144 s 0.97
const_scatter / JaXPipe / cuda / PreRev 0.000016576000000000002 s 0.000016576000000000002 s 1
const_scatter / JaXPipe / cuda / PostRev 0.000016704 s 0.000016096 s 1.04
const_scatter / JaXPipe / cuda / BothRev 0.000016832 s 0.000015935999999999998 s 1.06
const_scatter / Jax / cuda / BothRev 0.000017087 s 0.000016576000000000002 s 1.03
const_scatter / HLOOpt / cuda / PreRev 0.000016768000000000003 s 0.000016416 s 1.02
const_scatter / HLOOpt / cuda / PostRev 0.00001648 s 0.000016383999999999998 s 1.01
const_scatter / HLOOpt / cuda / BothRev 0.000016255999999999998 s 0.00001616 s 1.01
const_scatter / PartOpt / cuda / PreRev 0.000016736 s 0.000015744 s 1.06
const_scatter / PartOpt / cuda / PostRev 0.000017632 s 0.00001616 s 1.09
const_scatter / PartOpt / cuda / BothRev 0.000016416 s 0.000016031 s 1.02
const_scatter / IPartOpt / cuda / PreRev 0.000017056 s 0.000016063999999999997 s 1.06
const_scatter / IPartOpt / cuda / PostRev 0.00001696 s 0.00001664 s 1.02
const_scatter / IPartOpt / cuda / BothRev 0.000016927999999999998 s 0.000016255999999999998 s 1.04
const_scatter / DefOpt / cuda / PreRev 0.000016832 s 0.000016288 s 1.03
const_scatter / DefOpt / cuda / PostRev 0.000016704 s 0.000015935999999999998 s 1.05
const_scatter / DefOpt / cuda / BothRev 0.000016927999999999998 s 0.000015904000000000002 s 1.06
const_scatter / IDefOpt / cuda / PreRev 0.0000168 s 0.000015808 s 1.06
const_scatter / IDefOpt / cuda / PostRev 0.000016895 s 0.00001712 s 0.99
const_scatter / IDefOpt / cuda / BothRev 0.000016544 s 0.000015935999999999998 s 1.04
const_scatter / JaXPipe / tpu / Primal 0.00000378415 s 0.000003813725 s 0.99
const_scatter / Jax / tpu / Primal 0.0000038044 s 0.000003810925 s 1.00
const_scatter / HLOOpt / tpu / Primal 0.00000379835 s 0.000003826475 s 0.99
const_scatter / PartOpt / tpu / Primal 0.000003802575 s 0.000003823775 s 0.99
const_scatter / IPartOpt / tpu / Primal 0.000003804250000000001 s 0.000003800625 s 1.00
const_scatter / DefOpt / tpu / Primal 0.000003821550000000001 s 0.0000038115 s 1.00
const_scatter / IDefOpt / tpu / Primal 0.000003796825 s 0.000003787725 s 1.00
const_scatter / JaXPipe / tpu / Forward 0.000006471525000000001 s 0.000006466275 s 1.00
const_scatter / Jax / tpu / Forward 0.000006498925 s 0.0000065071 s 1.00
const_scatter / HLOOpt / tpu / Forward 0.00000645435 s 0.000006460749999999999 s 1.00
const_scatter / PartOpt / tpu / Forward 0.000006492124999999999 s 0.000006514025 s 1.00
const_scatter / IPartOpt / tpu / Forward 0.0000064791750000000005 s 0.00000643815 s 1.01
const_scatter / DefOpt / tpu / Forward 0.0000064808 s 0.00000649575 s 1.00
const_scatter / IDefOpt / tpu / Forward 0.000006453075 s 0.000006461475 s 1.00
const_scatter / JaXPipe / tpu / PreRev 0.00000661565 s 0.00000667555 s 0.99
const_scatter / JaXPipe / tpu / PostRev 0.000006630225 s 0.000006664774999999999 s 0.99
const_scatter / JaXPipe / tpu / BothRev 0.000006618125 s 0.0000066576250000000005 s 0.99
const_scatter / Jax / tpu / BothRev 0.00000662305 s 0.0000066598 s 0.99
const_scatter / HLOOpt / tpu / PreRev 0.000006591475 s 0.0000066682 s 0.99
const_scatter / HLOOpt / tpu / PostRev 0.000006613425000000001 s 0.000006652350000000001 s 0.99
const_scatter / HLOOpt / tpu / BothRev 0.00000659255 s 0.000006684425000000001 s 0.99
const_scatter / PartOpt / tpu / PreRev 0.0000066237 s 0.000006659425 s 0.99
const_scatter / PartOpt / tpu / PostRev 0.00000658535 s 0.0000066629 s 0.99
const_scatter / PartOpt / tpu / BothRev 0.000006611375 s 0.000006664474999999999 s 0.99
const_scatter / IPartOpt / tpu / PreRev 0.000006596324999999999 s 0.000006663175 s 0.99
const_scatter / IPartOpt / tpu / PostRev 0.00000661095 s 0.000006668525 s 0.99
const_scatter / IPartOpt / tpu / BothRev 0.00000662125 s 0.00000667465 s 0.99
const_scatter / DefOpt / tpu / PreRev 0.000006609475 s 0.000006678150000000001 s 0.99
const_scatter / DefOpt / tpu / PostRev 0.000006589724999999999 s 0.0000066493 s 0.99
const_scatter / DefOpt / tpu / BothRev 0.000006595175 s 0.000006677975 s 0.99
const_scatter / IDefOpt / tpu / PreRev 0.00000661855 s 0.000006650975 s 1.00
const_scatter / IDefOpt / tpu / PostRev 0.0000066008 s 0.000006669874999999999 s 0.99
const_scatter / IDefOpt / tpu / BothRev 0.000006602225000000001 s 0.00000665875 s 0.99
const_scatter / JaXPipe / cpu / Primal 0.000013104 s 0.000006414199997379911 s 2.04
const_scatter / Jax / cpu / Primal 0.000012399 s 0.000006381759994837921 s 1.94
const_scatter / HLOOpt / cpu / Primal 0.000013218 s 0.000007092419973560027 s 1.86
const_scatter / PartOpt / cpu / Primal 0.000012877 s 0.000006388200008586864 s 2.02
const_scatter / IPartOpt / cpu / Primal 0.000012396 s 0.000007287040007213363 s 1.70
const_scatter / DefOpt / cpu / Primal 0.000013285 s 0.000006763159999536583 s 1.96
const_scatter / IDefOpt / cpu / Primal 0.000013039 s 0.000007052119999571005 s 1.85
const_scatter / JaXPipe / cpu / Forward 0.000018154 s 0.000010790360001919908 s 1.68
const_scatter / Jax / cpu / Forward 0.000017194999999999997 s 0.000009555260021443246 s 1.80
const_scatter / HLOOpt / cpu / Forward 0.000018333 s 0.000011073819969169565 s 1.66
const_scatter / PartOpt / cpu / Forward 0.000017942 s 0.000010634000009304144 s 1.69
const_scatter / IPartOpt / cpu / Forward 0.000017498 s 0.000010531140032981057 s 1.66
const_scatter / DefOpt / cpu / Forward 0.000017922 s 0.000010720419995777774 s 1.67
const_scatter / IDefOpt / cpu / Forward 0.000017829999999999997 s 0.000010679459992388729 s 1.67
const_scatter / JaXPipe / cpu / PreRev 0.000524301 s 0.0002888223200261 s 1.82
const_scatter / JaXPipe / cpu / PostRev 0.000503708 s 0.0002839717799724 s 1.77
const_scatter / JaXPipe / cpu / BothRev 0.000518444 s 0.0002854160999595 s 1.82
const_scatter / Jax / cpu / BothRev 0.000515642 s 0.0002842025400059 s 1.81
const_scatter / HLOOpt / cpu / PreRev 0.000528528 s 0.0002855643800376 s 1.85
const_scatter / HLOOpt / cpu / PostRev 0.000526635 s 0.0002889120800318 s 1.82
const_scatter / HLOOpt / cpu / BothRev 0.000502907 s 0.0002857153400054 s 1.76
const_scatter / PartOpt / cpu / PreRev 0.000534475 s 0.000283256499988 s 1.89
const_scatter / PartOpt / cpu / PostRev 0.000530945 s 0.0002831990800314 s 1.87
const_scatter / PartOpt / cpu / BothRev 0.0005316079999999 s 0.0002844155800357 s 1.87
const_scatter / IPartOpt / cpu / PreRev 0.000533851 s 0.0002835344200138 s 1.88
const_scatter / IPartOpt / cpu / PostRev 0.000518363 s 0.0002824978799799 s 1.83
const_scatter / IPartOpt / cpu / BothRev 0.000522453 s 0.0002845528399939 s 1.84
const_scatter / DefOpt / cpu / PreRev 0.000529595 s 0.0002848829400136 s 1.86
const_scatter / DefOpt / cpu / PostRev 0.000499448 s 0.0002848792600161 s 1.75
const_scatter / DefOpt / cpu / BothRev 0.000522275 s 0.0002847787200244 s 1.83
const_scatter / IDefOpt / cpu / PreRev 0.000498006 s 0.0002828045599926 s 1.76
const_scatter / IDefOpt / cpu / PostRev 0.000524472 s 0.0002841483999873 s 1.85
const_scatter / IDefOpt / cpu / BothRev 0.0005326269999999 s 0.0002834731599705 s 1.88
const_scatter / JaXPipe / cpu / Primal 0.000008 s 0.000006414199997379911 s 1.25
const_scatter / Jax / cpu / Primal 0.000008 s 0.000006381759994837921 s 1.25
const_scatter / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.000007092419973560027 s 1.27
const_scatter / PartOpt / cpu / Primal 0.000008999999999999999 s 0.000006388200008586864 s 1.41
const_scatter / IPartOpt / cpu / Primal 0.000008999999999999999 s 0.000007287040007213363 s 1.24
const_scatter / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000006763159999536583 s 1.33
const_scatter / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000007052119999571005 s 1.28
const_scatter / JaXPipe / cpu / Forward 0.000013 s 0.000010790360001919908 s 1.20
const_scatter / Jax / cpu / Forward 0.000012 s 0.000009555260021443246 s 1.26
const_scatter / HLOOpt / cpu / Forward 0.000013 s 0.000011073819969169565 s 1.17
const_scatter / PartOpt / cpu / Forward 0.000013 s 0.000010634000009304144 s 1.22
const_scatter / IPartOpt / cpu / Forward 0.000013 s 0.000010531140032981057 s 1.23
const_scatter / DefOpt / cpu / Forward 0.000013 s 0.000010720419995777774 s 1.21
const_scatter / IDefOpt / cpu / Forward 0.000013 s 0.000010679459992388729 s 1.22
const_scatter / JaXPipe / cpu / PreRev 0.000335 s 0.0002888223200261 s 1.16
const_scatter / JaXPipe / cpu / PostRev 0.000374 s 0.0002839717799724 s 1.32
const_scatter / JaXPipe / cpu / BothRev 0.000342 s 0.0002854160999595 s 1.20
const_scatter / Jax / cpu / BothRev 0.0003439999999999 s 0.0002842025400059 s 1.21
const_scatter / HLOOpt / cpu / PreRev 0.0003689999999999 s 0.0002855643800376 s 1.29
const_scatter / HLOOpt / cpu / PostRev 0.000354 s 0.0002889120800318 s 1.23
const_scatter / HLOOpt / cpu / BothRev 0.0003439999999999 s 0.0002857153400054 s 1.20
const_scatter / PartOpt / cpu / PreRev 0.000342 s 0.000283256499988 s 1.21
const_scatter / PartOpt / cpu / PostRev 0.000343 s 0.0002831990800314 s 1.21
const_scatter / PartOpt / cpu / BothRev 0.000338 s 0.0002844155800357 s 1.19
const_scatter / IPartOpt / cpu / PreRev 0.000347 s 0.0002835344200138 s 1.22
const_scatter / IPartOpt / cpu / PostRev 0.00033 s 0.0002824978799799 s 1.17
const_scatter / IPartOpt / cpu / BothRev 0.000371 s 0.0002845528399939 s 1.30
const_scatter / DefOpt / cpu / PreRev 0.000328 s 0.0002848829400136 s 1.15
const_scatter / DefOpt / cpu / PostRev 0.000345 s 0.0002848792600161 s 1.21
const_scatter / DefOpt / cpu / BothRev 0.000367 s 0.0002847787200244 s 1.29
const_scatter / IDefOpt / cpu / PreRev 0.000373 s 0.0002828045599926 s 1.32
const_scatter / IDefOpt / cpu / PostRev 0.000357 s 0.0002841483999873 s 1.26
const_scatter / IDefOpt / cpu / BothRev 0.000326 s 0.0002834731599705 s 1.15
GenDot / JaXPipe / cpu / Primal 0.00000834389998090046 s 0.000007298380041902419 s 1.14
GenDot / Jax / cpu / Primal 0.000008363280021512765 s 0.000007048700008454034 s 1.19
GenDot / HLOOpt / cpu / Primal 0.000009664120034358347 s 0.00000794292001046415 s 1.22
GenDot / PartOpt / cpu / Primal 0.000007279819965333445 s 0.000006769920055376133 s 1.08
GenDot / IPartOpt / cpu / Primal 0.000007913740000731196 s 0.000007515740007875138 s 1.05
GenDot / DefOpt / cpu / Primal 0.000008623260009699152 s 0.000007409619984173332 s 1.16
GenDot / IDefOpt / cpu / Primal 0.00000869983999109536 s 0.000007622040011483477 s 1.14
GenDot / JaXPipe / cpu / Forward 0.00001187806003144942 s 0.000011569240014068782 s 1.03
GenDot / Jax / cpu / Forward 0.000011197619978702275 s 0.000009718079991216654 s 1.15
GenDot / HLOOpt / cpu / Forward 0.000012843319973399048 s 0.000011745740011974704 s 1.09
GenDot / PartOpt / cpu / Forward 0.000012114840019421536 s 0.000011508360021252884 s 1.05
GenDot / IPartOpt / cpu / Forward 0.000011893119999513149 s 0.000011128580026706914 s 1.07
GenDot / DefOpt / cpu / Forward 0.000012176180016467697 s 0.000010798399998748209 s 1.13
GenDot / IDefOpt / cpu / Forward 0.000012360160008029198 s 0.00001077956000699487 s 1.15
GenDot / JaXPipe / cpu / PreRev 0.000012691379997704645 s 0.000010790999995151652 s 1.18
GenDot / JaXPipe / cpu / PostRev 0.000011203600006410852 s 0.000010381940037405 s 1.08
GenDot / JaXPipe / cpu / BothRev 0.000011917800020455617 s 0.000011352839983373995 s 1.05
GenDot / Jax / cpu / BothRev 0.000011264020013186384 s 0.000011042040014217492 s 1.02
GenDot / HLOOpt / cpu / PreRev 0.000012563860000227578 s 0.000011735859989130403 s 1.07
GenDot / HLOOpt / cpu / PostRev 0.000014509440052279388 s 0.000013340659997993496 s 1.09
GenDot / HLOOpt / cpu / BothRev 0.000012399200031723012 s 0.000010679819979486638 s 1.16
GenDot / PartOpt / cpu / PreRev 0.000011894400013261474 s 0.000010874159997911192 s 1.09
GenDot / PartOpt / cpu / PostRev 0.000010948419985652436 s 0.000011166739977852555 s 0.98
GenDot / PartOpt / cpu / BothRev 0.00001290048002374533 s 0.000011426359942561248 s 1.13
GenDot / IPartOpt / cpu / PreRev 0.000012594620011441294 s 0.000010923079989879626 s 1.15
GenDot / IPartOpt / cpu / PostRev 0.000010794000008900184 s 0.000010891519996221178 s 0.99
GenDot / IPartOpt / cpu / BothRev 0.00001275029997486854 s 0.000011160020021634409 s 1.14
GenDot / DefOpt / cpu / PreRev 0.000012181479987702917 s 0.000011140939977849483 s 1.09
GenDot / DefOpt / cpu / PostRev 0.000012328759985393844 s 0.000011533419992701966 s 1.07
GenDot / DefOpt / cpu / BothRev 0.000012386120015435154 s 0.000010761060020740842 s 1.15
GenDot / IDefOpt / cpu / PreRev 0.000012613420003617647 s 0.000010950339974442614 s 1.15
GenDot / IDefOpt / cpu / PostRev 0.00001220397999531997 s 0.00001149753998106462 s 1.06
GenDot / IDefOpt / cpu / BothRev 0.000011784879989136243 s 0.000010877040003833827 s 1.08
GenDot / JaXPipe / cuda / Primal 0.000002016 s 0.000002016 s 1
GenDot / Jax / cuda / Primal 0.000002015 s 0.000002015 s 1
GenDot / HLOOpt / cuda / Primal 0.000002015 s 0.000001984 s 1.02
GenDot / PartOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
GenDot / IPartOpt / cuda / Primal 0.000002016 s 0.000002016 s 1
GenDot / DefOpt / cuda / Primal 0.000001983 s 0.000002015 s 0.98
GenDot / IDefOpt / cuda / Primal 0.000002015 s 0.000002015 s 1
GenDot / JaXPipe / cuda / Forward 0.000010177 s 0.000009409 s 1.08
GenDot / Jax / cuda / Forward 0.000010081 s 0.000009728 s 1.04
GenDot / HLOOpt / cuda / Forward 0.00001008 s 0.000009696 s 1.04
GenDot / PartOpt / cuda / Forward 0.000010305 s 0.000009792 s 1.05
GenDot / IPartOpt / cuda / Forward 0.000010176 s 0.000009024 s 1.13
GenDot / DefOpt / cuda / Forward 0.000010272 s 0.000009984 s 1.03
GenDot / IDefOpt / cuda / Forward 0.00001024 s 0.000009855 s 1.04
GenDot / JaXPipe / cuda / PreRev 0.000010016 s 0.000008703000000000001 s 1.15
GenDot / JaXPipe / cuda / PostRev 0.00001008 s 0.000009856 s 1.02
GenDot / JaXPipe / cuda / BothRev 0.000010176 s 0.000009536 s 1.07
GenDot / Jax / cuda / BothRev 0.000010367 s 0.000009312000000000002 s 1.11
GenDot / HLOOpt / cuda / PreRev 0.000009696 s 0.000009504 s 1.02
GenDot / HLOOpt / cuda / PostRev 0.000010144 s 0.000014752 s 0.69
GenDot / HLOOpt / cuda / BothRev 0.000010208 s 0.00001104 s 0.92
GenDot / PartOpt / cuda / PreRev 0.000010657 s 0.000010912 s 0.98
GenDot / PartOpt / cuda / PostRev 0.000010529 s 0.000011104 s 0.95
GenDot / PartOpt / cuda / BothRev 0.000010304 s 0.000011104 s 0.93
GenDot / IPartOpt / cuda / PreRev 0.000010111 s 0.000010784 s 0.94
GenDot / IPartOpt / cuda / PostRev 0.000010272 s 0.000011072 s 0.93
GenDot / IPartOpt / cuda / BothRev 0.000010112 s 0.00000992 s 1.02
GenDot / DefOpt / cuda / PreRev 0.000010432 s 0.000009856 s 1.06
GenDot / DefOpt / cuda / PostRev 0.000009888 s 0.000009568 s 1.03
GenDot / DefOpt / cuda / BothRev 0.000010176 s 0.000009856 s 1.03
GenDot / IDefOpt / cuda / PreRev 0.000010145 s 0.000009856 s 1.03
GenDot / IDefOpt / cuda / PostRev 0.000010368 s 0.000009536 s 1.09
GenDot / IDefOpt / cuda / BothRev 0.000010272 s 0.000009632 s 1.07
GenDot / JaXPipe / tpu / Primal 9.29725e-7 s 9.29875e-7 s 1.00
GenDot / Jax / tpu / Primal 9.255e-7 s 9.2665e-7 s 1.00
GenDot / HLOOpt / tpu / Primal 0.000001566525 s 0.00000158335 s 0.99
GenDot / PartOpt / tpu / Primal 9.25825e-7 s 9.2605e-7 s 1.00
GenDot / IPartOpt / tpu / Primal 9.30075e-7 s 9.30725e-7 s 1.00
GenDot / DefOpt / tpu / Primal 0.0000014843 s 0.0000015009749999999998 s 0.99
GenDot / IDefOpt / tpu / Primal 0.0000015818750000000002 s 0.000001590625 s 0.99
GenDot / JaXPipe / tpu / Forward 0.0000031542 s 0.00000317515 s 0.99
GenDot / Jax / tpu / Forward 0.0000023175000000000003 s 0.00000233035 s 0.99
GenDot / HLOOpt / tpu / Forward 0.0000031075000000000003 s 0.0000031335 s 0.99
GenDot / PartOpt / tpu / Forward 0.0000032142 s 0.000003231625 s 0.99
GenDot / IPartOpt / tpu / Forward 0.000003108025 s 0.0000031330000000000003 s 0.99
GenDot / DefOpt / tpu / Forward 0.00000321265 s 0.0000032337 s 0.99
GenDot / IDefOpt / tpu / Forward 0.0000031152000000000003 s 0.0000031358 s 0.99
GenDot / JaXPipe / tpu / PreRev 0.000002949 s 0.000002985475 s 0.99
GenDot / JaXPipe / tpu / PostRev 0.000002401625 s 0.000002399875 s 1.00
GenDot / JaXPipe / tpu / BothRev 0.000002950825 s 0.000002993825 s 0.99
GenDot / Jax / tpu / BothRev 0.000002407175 s 0.00000239935 s 1.00
GenDot / HLOOpt / tpu / PreRev 0.00000294445 s 0.0000029822 s 0.99
GenDot / HLOOpt / tpu / PostRev 0.000002919775 s 0.0000029412 s 0.99
GenDot / HLOOpt / tpu / BothRev 0.0000029620750000000004 s 0.000003004175 s 0.99
GenDot / PartOpt / tpu / PreRev 0.0000029303750000000003 s 0.000002957425 s 0.99
GenDot / PartOpt / tpu / PostRev 0.00000239605 s 0.000002396275 s 1.00
GenDot / PartOpt / tpu / BothRev 0.000002922625 s 0.000002944225 s 0.99
GenDot / IPartOpt / tpu / PreRev 0.00000295005 s 0.000002986025 s 0.99
GenDot / IPartOpt / tpu / PostRev 0.00000241205 s 0.000002399275 s 1.01
GenDot / IPartOpt / tpu / BothRev 0.00000295 s 0.0000029928 s 0.99
GenDot / DefOpt / tpu / PreRev 0.0000029226500000000003 s 0.00000293195 s 1.00
GenDot / DefOpt / tpu / PostRev 0.0000029489 s 0.000002990275 s 0.99
GenDot / DefOpt / tpu / BothRev 0.00000292455 s 0.00000293015 s 1.00
GenDot / IDefOpt / tpu / PreRev 0.000002948675 s 0.000002993225 s 0.99
GenDot / IDefOpt / tpu / PostRev 0.0000029332750000000003 s 0.000002935 s 1.00
GenDot / IDefOpt / tpu / BothRev 0.000002948075 s 0.0000029784500000000003 s 0.99
GenDot / JaXPipe / cpu / Primal 0.000015266 s 0.000007298380041902419 s 2.09
GenDot / Jax / cpu / Primal 0.000015251 s 0.000007048700008454034 s 2.16
GenDot / HLOOpt / cpu / Primal 0.000013771 s 0.00000794292001046415 s 1.73
GenDot / PartOpt / cpu / Primal 0.000015107 s 0.000006769920055376133 s 2.23
GenDot / IPartOpt / cpu / Primal 0.00001497 s 0.000007515740007875138 s 1.99
GenDot / DefOpt / cpu / Primal 0.000013883 s 0.000007409619984173332 s 1.87
GenDot / IDefOpt / cpu / Primal 0.0000143 s 0.000007622040011483477 s 1.88
GenDot / JaXPipe / cpu / Forward 0.000019655 s 0.000011569240014068782 s 1.70
GenDot / Jax / cpu / Forward 0.000020258 s 0.000009718079991216654 s 2.08
GenDot / HLOOpt / cpu / Forward 0.000019045 s 0.000011745740011974704 s 1.62
GenDot / PartOpt / cpu / Forward 0.000019152 s 0.000011508360021252884 s 1.66
GenDot / IPartOpt / cpu / Forward 0.000019614 s 0.000011128580026706914 s 1.76
GenDot / DefOpt / cpu / Forward 0.000019394 s 0.000010798399998748209 s 1.80
GenDot / IDefOpt / cpu / Forward 0.000019428 s 0.00001077956000699487 s 1.80
GenDot / JaXPipe / cpu / PreRev 0.000019041 s 0.000010790999995151652 s 1.76
GenDot / JaXPipe / cpu / PostRev 0.000020677 s 0.000010381940037405 s 1.99
GenDot / JaXPipe / cpu / BothRev 0.000019416 s 0.000011352839983373995 s 1.71
GenDot / Jax / cpu / BothRev 0.000020471 s 0.000011042040014217492 s 1.85
GenDot / HLOOpt / cpu / PreRev 0.000019306 s 0.000011735859989130403 s 1.65
GenDot / HLOOpt / cpu / PostRev 0.000019736 s 0.000013340659997993496 s 1.48
GenDot / HLOOpt / cpu / BothRev 0.000019214 s 0.000010679819979486638 s 1.80
GenDot / PartOpt / cpu / PreRev 0.000019223 s 0.000010874159997911192 s 1.77
GenDot / PartOpt / cpu / PostRev 0.000020238 s 0.000011166739977852555 s 1.81
GenDot / PartOpt / cpu / BothRev 0.000019501 s 0.000011426359942561248 s 1.71
GenDot / IPartOpt / cpu / PreRev 0.000019668 s 0.000010923079989879626 s 1.80
GenDot / IPartOpt / cpu / PostRev 0.000020175 s 0.000010891519996221178 s 1.85
GenDot / IPartOpt / cpu / BothRev 0.000019215 s 0.000011160020021634409 s 1.72
GenDot / DefOpt / cpu / PreRev 0.000019177 s 0.000011140939977849483 s 1.72
GenDot / DefOpt / cpu / PostRev 0.000019427 s 0.000011533419992701966 s 1.68
GenDot / DefOpt / cpu / BothRev 0.000018973 s 0.000010761060020740842 s 1.76
GenDot / IDefOpt / cpu / PreRev 0.000019233 s 0.000010950339974442614 s 1.76
GenDot / IDefOpt / cpu / PostRev 0.000019334 s 0.00001149753998106462 s 1.68
GenDot / IDefOpt / cpu / BothRev 0.000019352 s 0.000010877040003833827 s 1.78
GenDot / JaXPipe / cpu / Primal 0.00001 s 0.000007298380041902419 s 1.37
GenDot / Jax / cpu / Primal 0.00001 s 0.000007048700008454034 s 1.42
GenDot / HLOOpt / cpu / Primal 0.00001 s 0.00000794292001046415 s 1.26
GenDot / PartOpt / cpu / Primal 0.00001 s 0.000006769920055376133 s 1.48
GenDot / IPartOpt / cpu / Primal 0.000011 s 0.000007515740007875138 s 1.46
GenDot / DefOpt / cpu / Primal 0.00001 s 0.000007409619984173332 s 1.35
GenDot / IDefOpt / cpu / Primal 0.000008999999999999999 s 0.000007622040011483477 s 1.18
GenDot / JaXPipe / cpu / Forward 0.000014 s 0.000011569240014068782 s 1.21
GenDot / Jax / cpu / Forward 0.000014 s 0.000009718079991216654 s 1.44
GenDot / HLOOpt / cpu / Forward 0.000013 s 0.000011745740011974704 s 1.11
GenDot / PartOpt / cpu / Forward 0.000013 s 0.000011508360021252884 s 1.13
GenDot / IPartOpt / cpu / Forward 0.000014 s 0.000011128580026706914 s 1.26
GenDot / DefOpt / cpu / Forward 0.000013 s 0.000010798399998748209 s 1.20
GenDot / IDefOpt / cpu / Forward 0.000014 s 0.00001077956000699487 s 1.30
GenDot / JaXPipe / cpu / PreRev 0.000014 s 0.000010790999995151652 s 1.30
GenDot / JaXPipe / cpu / PostRev 0.000014 s 0.000010381940037405 s 1.35
GenDot / JaXPipe / cpu / BothRev 0.000014 s 0.000011352839983373995 s 1.23
GenDot / Jax / cpu / BothRev 0.000015 s 0.000011042040014217492 s 1.36
GenDot / HLOOpt / cpu / PreRev 0.000014 s 0.000011735859989130403 s 1.19
GenDot / HLOOpt / cpu / PostRev 0.000015 s 0.000013340659997993496 s 1.12
GenDot / HLOOpt / cpu / BothRev 0.000014 s 0.000010679819979486638 s 1.31
GenDot / PartOpt / cpu / PreRev 0.000015 s 0.000010874159997911192 s 1.38
GenDot / PartOpt / cpu / PostRev 0.000014 s 0.000011166739977852555 s 1.25
GenDot / PartOpt / cpu / BothRev 0.000015 s 0.000011426359942561248 s 1.31
GenDot / IPartOpt / cpu / PreRev 0.000014 s 0.000010923079989879626 s 1.28
GenDot / IPartOpt / cpu / PostRev 0.000014 s 0.000010891519996221178 s 1.29
GenDot / IPartOpt / cpu / BothRev 0.000014 s 0.000011160020021634409 s 1.25
GenDot / DefOpt / cpu / PreRev 0.000046 s 0.000011140939977849483 s 4.13
GenDot / DefOpt / cpu / PostRev 0.000014 s 0.000011533419992701966 s 1.21
GenDot / DefOpt / cpu / BothRev 0.000014 s 0.000010761060020740842 s 1.30
GenDot / IDefOpt / cpu / PreRev 0.000014 s 0.000010950339974442614 s 1.28
GenDot / IDefOpt / cpu / PostRev 0.000015 s 0.00001149753998106462 s 1.30
GenDot / IDefOpt / cpu / BothRev 0.000014 s 0.000010877040003833827 s 1.29
hlo_ffi / JaXPipe / cpu / Primal 0.000011706140012393008 s 0.000010284639974997845 s 1.14
hlo_ffi / Jax / cpu / Primal 0.00001178867998532951 s 0.000009667560016168864 s 1.22
hlo_ffi / HLOOpt / cpu / Primal 0.000011059139997087186 s 0.000010284979998687047 s 1.08
hlo_ffi / PartOpt / cpu / Primal 0.000011193000009370737 s 0.00000951727997744456 s 1.18
hlo_ffi / IPartOpt / cpu / Primal 0.000010900520019276882 s 0.000010334960024920293 s 1.05
hlo_ffi / DefOpt / cpu / Primal 0.000010618299984344047 s 0.000010058319985546404 s 1.06
hlo_ffi / IDefOpt / cpu / Primal 0.000010693479989640764 s 0.000009962120047930512 s 1.07
hlo_ffi / JaXPipe / cpu / Forward 0.000016407679959229425 s 0.000014187380020302954 s 1.16
hlo_ffi / Jax / cpu / Forward 0.000016098939995572436 s 0.000013873439984308789 s 1.16
hlo_ffi / HLOOpt / cpu / Forward 0.000016185619988391408 s 0.00001400121997903625 s 1.16
hlo_ffi / PartOpt / cpu / Forward 0.000017091220015572617 s 0.00001432539997040294 s 1.19
hlo_ffi / IPartOpt / cpu / Forward 0.00001635109996641404 s 0.000014166840010148008 s 1.15
hlo_ffi / DefOpt / cpu / Forward 0.00001639830001295195 s 0.000014209059972927207 s 1.15
hlo_ffi / IDefOpt / cpu / Forward 0.00001659839996136725 s 0.000014338580022013049 s 1.16
hlo_ffi / JaXPipe / cpu / PreRev 0.000015940800003590994 s 0.000014636240020990954 s 1.09
hlo_ffi / JaXPipe / cpu / PostRev 0.000014963299972805544 s 0.000014110460006122592 s 1.06
hlo_ffi / JaXPipe / cpu / BothRev 0.000015130559977478698 s 0.00001390391998029372 s 1.09
hlo_ffi / Jax / cpu / BothRev 0.000015891280036157695 s 0.000014062179998290958 s 1.13
hlo_ffi / HLOOpt / cpu / PreRev 0.000016390999971918062 s 0.000014028819978193496 s 1.17
hlo_ffi / HLOOpt / cpu / PostRev 0.000017510280040369254 s 0.00001594318003299122 s 1.10
hlo_ffi / HLOOpt / cpu / BothRev 0.00001549012003124517 s 0.000014413599992622038 s 1.07
hlo_ffi / PartOpt / cpu / PreRev 0.00001587989998370176 s 0.00001416582001183997 s 1.12
hlo_ffi / PartOpt / cpu / PostRev 0.000015167759966061566 s 0.00001393637999171915 s 1.09
hlo_ffi / PartOpt / cpu / BothRev 0.00001534121998702176 s 0.0000143006199868978 s 1.07
hlo_ffi / IPartOpt / cpu / PreRev 0.000016250420003416366 s 0.000013980779976918712 s 1.16
hlo_ffi / IPartOpt / cpu / PostRev 0.00001543517998470634 s 0.000014213119957275922 s 1.09
hlo_ffi / IPartOpt / cpu / BothRev 0.000015220440000121016 s 0.000014170980020935531 s 1.07
hlo_ffi / DefOpt / cpu / PreRev 0.000016794440016383304 s 0.000014085239999985788 s 1.19
hlo_ffi / DefOpt / cpu / PostRev 0.000015076000008775737 s 0.000014310660035334876 s 1.05
hlo_ffi / DefOpt / cpu / BothRev 0.000015334120025727315 s 0.000013982019991090056 s 1.10
hlo_ffi / IDefOpt / cpu / PreRev 0.000017425140022169217 s 0.000014316020051410303 s 1.22
hlo_ffi / IDefOpt / cpu / PostRev 0.000015165299992077052 s 0.000014057119979042908 s 1.08
hlo_ffi / IDefOpt / cpu / BothRev 0.000015407520068038137 s 0.000014209260025381808 s 1.08
hlo_ffi / JaXPipe / cuda / Primal 0.000001983 s 0.000001983 s 1
hlo_ffi / Jax / cuda / Primal 0.000001983 s 0.000001983 s 1
hlo_ffi / HLOOpt / cuda / Primal 0.000001984 s 0.000001984 s 1
hlo_ffi / PartOpt / cuda / Primal 0.000001984 s 0.000001983 s 1.00
hlo_ffi / IPartOpt / cuda / Primal 0.000001984 s 0.000001984 s 1
hlo_ffi / DefOpt / cuda / Primal 0.000001984 s 0.000001983 s 1.00
hlo_ffi / IDefOpt / cuda / Primal 0.000001983 s 0.000001983 s 1
hlo_ffi / JaXPipe / cuda / Forward 0.00000208 s 0.000002049 s 1.02
hlo_ffi / Jax / cuda / Forward 0.00000208 s 0.00000208 s 1
hlo_ffi / HLOOpt / cuda / Forward 0.00000208 s 0.00000208 s 1
hlo_ffi / PartOpt / cuda / Forward 0.00000208 s 0.000002048 s 1.02
hlo_ffi / IPartOpt / cuda / Forward 0.00000208 s 0.00000208 s 1
hlo_ffi / DefOpt / cuda / Forward 0.00000208 s 0.00000208 s 1
hlo_ffi / IDefOpt / cuda / Forward 0.00000208 s 0.000002048 s 1.02
hlo_ffi / JaXPipe / cuda / PreRev 0.000002047 s 0.000002047 s 1
hlo_ffi / JaXPipe / cuda / PostRev 0.000002048 s 0.000002048 s 1
hlo_ffi / JaXPipe / cuda / BothRev 0.000002048 s 0.000002048 s 1
hlo_ffi / Jax / cuda / BothRev 0.000002048 s 0.000002048 s 1
hlo_ffi / HLOOpt / cuda / PreRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / HLOOpt / cuda / PostRev 0.000002048 s 0.000002048 s 1
hlo_ffi / HLOOpt / cuda / BothRev 0.000002047 s 0.000002047 s 1
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 0.000002047 s 1.00
hlo_ffi / IPartOpt / cuda / PreRev 0.000002047 s 0.000002047 s 1
hlo_ffi / IPartOpt / cuda / PostRev 0.000002047 s 0.000002048 s 1.00
hlo_ffi / IPartOpt / cuda / BothRev 0.000002047 s 0.000002047 s 1
hlo_ffi / DefOpt / cuda / PreRev 0.000002048 s 0.000002047 s 1.00
hlo_ffi / DefOpt / cuda / PostRev 0.000002048 s 0.000002048 s 1
hlo_ffi / DefOpt / cuda / BothRev 0.000002047 s 0.000002047 s 1
hlo_ffi / IDefOpt / cuda / PreRev 0.000002048 s 0.000002048 s 1
hlo_ffi / IDefOpt / cuda / PostRev 0.000002048 s 0.000002048 s 1
hlo_ffi / IDefOpt / cuda / BothRev 0.000002048 s 0.000002047 s 1.00
hlo_ffi / JaXPipe / tpu / Primal 9.21225e-7 s 9.186e-7 s 1.00
hlo_ffi / Jax / tpu / Primal 9.523e-7 s 9.53225e-7 s 1.00
hlo_ffi / HLOOpt / tpu / Primal 8.965750000000001e-7 s 8.98e-7 s 1.00
hlo_ffi / PartOpt / tpu / Primal 9.536e-7 s 9.543e-7 s 1.00
hlo_ffi / IPartOpt / tpu / Primal 8.9735e-7 s 8.99475e-7 s 1.00
hlo_ffi / DefOpt / tpu / Primal 9.51775e-7 s 9.518e-7 s 1.00
hlo_ffi / IDefOpt / tpu / Primal 8.959250000000001e-7 s 8.9985e-7 s 1.00
hlo_ffi / JaXPipe / tpu / Forward 9.49125e-7 s 9.4875e-7 s 1.00
hlo_ffi / Jax / tpu / Forward 9.817e-7 s 9.810750000000002e-7 s 1.00
hlo_ffi / HLOOpt / tpu / Forward 9.73875e-7 s 9.737250000000002e-7 s 1.00
hlo_ffi / PartOpt / tpu / Forward 9.34225e-7 s 9.3345e-7 s 1.00
hlo_ffi / IPartOpt / tpu / Forward 9.73925e-7 s 9.7375e-7 s 1.00
hlo_ffi / DefOpt / tpu / Forward 9.344e-7 s 9.33125e-7 s 1.00
hlo_ffi / IDefOpt / tpu / Forward 9.74e-7 s 9.739e-7 s 1.00
hlo_ffi / JaXPipe / tpu / PreRev 9.31775e-7 s 9.315e-7 s 1.00
hlo_ffi / JaXPipe / tpu / PostRev 9.654e-7 s 9.64925e-7 s 1.00
hlo_ffi / JaXPipe / tpu / BothRev 9.6215e-7 s 9.62e-7 s 1.00
hlo_ffi / Jax / tpu / BothRev 9.6445e-7 s 9.64325e-7 s 1.00
hlo_ffi / HLOOpt / tpu / PreRev 9.62e-7 s 9.62075e-7 s 1.00
hlo_ffi / HLOOpt / tpu / PostRev 9.64975e-7 s 9.6465e-7 s 1.00
hlo_ffi / HLOOpt / tpu / BothRev 9.61825e-7 s 9.6245e-7 s 1.00
hlo_ffi / PartOpt / tpu / PreRev 9.64975e-7 s 9.64725e-7 s 1.00
hlo_ffi / PartOpt / tpu / PostRev 9.6165e-7 s 9.616e-7 s 1.00
hlo_ffi / PartOpt / tpu / BothRev 9.646e-7 s 9.6475e-7 s 1.00
hlo_ffi / IPartOpt / tpu / PreRev 9.622e-7 s 9.62225e-7 s 1.00
hlo_ffi / IPartOpt / tpu / PostRev 9.64825e-7 s 9.649e-7 s 1.00
hlo_ffi / IPartOpt / tpu / BothRev 9.62225e-7 s 9.6195e-7 s 1.00
hlo_ffi / DefOpt / tpu / PreRev 9.64775e-7 s 9.648e-7 s 1.00
hlo_ffi / DefOpt / tpu / PostRev 9.618e-7 s 9.618500000000002e-7 s 1.00
hlo_ffi / DefOpt / tpu / BothRev 9.6495e-7 s 9.644e-7 s 1.00
hlo_ffi / IDefOpt / tpu / PreRev 9.62275e-7 s 9.617e-7 s 1.00
hlo_ffi / IDefOpt / tpu / PostRev 9.64975e-7 s 9.64525e-7 s 1.00
hlo_ffi / IDefOpt / tpu / BothRev 9.62125e-7 s 9.61725e-7 s 1.00
hlo_ffi / JaXPipe / cpu / Primal 0.000017962 s 0.000010284639974997845 s 1.75
hlo_ffi / Jax / cpu / Primal 0.000017063999999999998 s 0.000009667560016168864 s 1.77
hlo_ffi / HLOOpt / cpu / Primal 0.00001704 s 0.000010284979998687047 s 1.66
hlo_ffi / PartOpt / cpu / Primal 0.000017147 s 0.00000951727997744456 s 1.80
hlo_ffi / IPartOpt / cpu / Primal 0.000017686 s 0.000010334960024920293 s 1.71
hlo_ffi / DefOpt / cpu / Primal 0.000017267999999999998 s 0.000010058319985546404 s 1.72
hlo_ffi / IDefOpt / cpu / Primal 0.000017531000000000002 s 0.000009962120047930512 s 1.76
hlo_ffi / JaXPipe / cpu / Forward 0.000024508 s 0.000014187380020302954 s 1.73
hlo_ffi / Jax / cpu / Forward 0.000023698 s 0.000013873439984308789 s 1.71
hlo_ffi / HLOOpt / cpu / Forward 0.00002408 s 0.00001400121997903625 s 1.72
hlo_ffi / PartOpt / cpu / Forward 0.000024219 s 0.00001432539997040294 s 1.69
hlo_ffi / IPartOpt / cpu / Forward 0.00002385 s 0.000014166840010148008 s 1.68
hlo_ffi / DefOpt / cpu / Forward 0.000023825 s 0.000014209059972927207 s 1.68
hlo_ffi / IDefOpt / cpu / Forward 0.000024125 s 0.000014338580022013049 s 1.68
hlo_ffi / JaXPipe / cpu / PreRev 0.000025127 s 0.000014636240020990954 s 1.72
hlo_ffi / JaXPipe / cpu / PostRev 0.000025101 s 0.000014110460006122592 s 1.78
hlo_ffi / JaXPipe / cpu / BothRev 0.000024189 s 0.00001390391998029372 s 1.74
hlo_ffi / Jax / cpu / BothRev 0.000024076 s 0.000014062179998290958 s 1.71
hlo_ffi / HLOOpt / cpu / PreRev 0.000024912 s 0.000014028819978193496 s 1.78
hlo_ffi / HLOOpt / cpu / PostRev 0.000023799 s 0.00001594318003299122 s 1.49
hlo_ffi / HLOOpt / cpu / BothRev 0.000023252 s 0.000014413599992622038 s 1.61
hlo_ffi / PartOpt / cpu / PreRev 0.000024392 s 0.00001416582001183997 s 1.72
hlo_ffi / PartOpt / cpu / PostRev 0.000023683 s 0.00001393637999171915 s 1.70
hlo_ffi / PartOpt / cpu / BothRev 0.00002374 s 0.0000143006199868978 s 1.66
hlo_ffi / IPartOpt / cpu / PreRev 0.000023467 s 0.000013980779976918712 s 1.68
hlo_ffi / IPartOpt / cpu / PostRev 0.000024409 s 0.000014213119957275922 s 1.72
hlo_ffi / IPartOpt / cpu / BothRev 0.000023621 s 0.000014170980020935531 s 1.67
hlo_ffi / DefOpt / cpu / PreRev 0.000024249 s 0.000014085239999985788 s 1.72
hlo_ffi / DefOpt / cpu / PostRev 0.000023714 s 0.000014310660035334876 s 1.66
hlo_ffi / DefOpt / cpu / BothRev 0.0000239 s 0.000013982019991090056 s 1.71
hlo_ffi / IDefOpt / cpu / PreRev 0.000025029 s 0.000014316020051410303 s 1.75
hlo_ffi / IDefOpt / cpu / PostRev 0.000024141 s 0.000014057119979042908 s 1.72
hlo_ffi / IDefOpt / cpu / BothRev 0.000023655 s 0.000014209260025381808 s 1.66
hlo_ffi / JaXPipe / cpu / Primal 0.000013 s 0.000010284639974997845 s 1.26
hlo_ffi / Jax / cpu / Primal 0.000013 s 0.000009667560016168864 s 1.34
hlo_ffi / HLOOpt / cpu / Primal 0.000013 s 0.000010284979998687047 s 1.26
hlo_ffi / PartOpt / cpu / Primal 0.000013 s 0.00000951727997744456 s 1.37
hlo_ffi / IPartOpt / cpu / Primal 0.000012 s 0.000010334960024920293 s 1.16
hlo_ffi / DefOpt / cpu / Primal 0.000012 s 0.000010058319985546404 s 1.19
hlo_ffi / IDefOpt / cpu / Primal 0.000013 s 0.000009962120047930512 s 1.30
hlo_ffi / JaXPipe / cpu / Forward 0.000019 s 0.000014187380020302954 s 1.34
hlo_ffi / Jax / cpu / Forward 0.000052 s 0.000013873439984308789 s 3.75
hlo_ffi / HLOOpt / cpu / Forward 0.000017 s 0.00001400121997903625 s 1.21
hlo_ffi / PartOpt / cpu / Forward 0.000017 s 0.00001432539997040294 s 1.19
hlo_ffi / IPartOpt / cpu / Forward 0.000019 s 0.000014166840010148008 s 1.34
hlo_ffi / DefOpt / cpu / Forward 0.000017999999999999997 s 0.000014209059972927207 s 1.27
hlo_ffi / IDefOpt / cpu / Forward 0.000017 s 0.000014338580022013049 s 1.19
hlo_ffi / JaXPipe / cpu / PreRev 0.000051 s 0.000014636240020990954 s 3.48
hlo_ffi / JaXPipe / cpu / PostRev 0.000017 s 0.000014110460006122592 s 1.20
hlo_ffi / JaXPipe / cpu / BothRev 0.000017 s 0.00001390391998029372 s 1.22
hlo_ffi / Jax / cpu / BothRev 0.000017 s 0.000014062179998290958 s 1.21
hlo_ffi / HLOOpt / cpu / PreRev 0.000017 s 0.000014028819978193496 s 1.21
hlo_ffi / HLOOpt / cpu / PostRev 0.000056 s 0.00001594318003299122 s 3.51
hlo_ffi / HLOOpt / cpu / BothRev 0.000017999999999999997 s 0.000014413599992622038 s 1.25
hlo_ffi / PartOpt / cpu / PreRev 0.000017 s 0.00001416582001183997 s 1.20
hlo_ffi / PartOpt / cpu / PostRev 0.000017999999999999997 s 0.00001393637999171915 s 1.29
hlo_ffi / PartOpt / cpu / BothRev 0.000017999999999999997 s 0.0000143006199868978 s 1.26
hlo_ffi / IPartOpt / cpu / PreRev 0.000017 s 0.000013980779976918712 s 1.22
hlo_ffi / IPartOpt / cpu / PostRev 0.000017999999999999997 s 0.000014213119957275922 s 1.27
hlo_ffi / IPartOpt / cpu / BothRev 0.000017 s 0.000014170980020935531 s 1.20
hlo_ffi / DefOpt / cpu / PreRev 0.000017999999999999997 s 0.000014085239999985788 s 1.28
hlo_ffi / DefOpt / cpu / PostRev 0.000017999999999999997 s 0.000014310660035334876 s 1.26
hlo_ffi / DefOpt / cpu / BothRev 0.000017999999999999997 s 0.000013982019991090056 s 1.29
hlo_ffi / IDefOpt / cpu / PreRev 0.000017999999999999997 s 0.000014316020051410303 s 1.26
hlo_ffi / IDefOpt / cpu / PostRev 0.000017999999999999997 s 0.000014057119979042908 s 1.28
hlo_ffi / IDefOpt / cpu / BothRev 0.000017999999999999997 s 0.000014209260025381808 s 1.27
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 0.0008797404000688 s 0.0009100650000618 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 / Jax / cpu / Primal 0.0008865859999787 s 0.0009172600000056 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 / HLOOpt / cpu / Primal 0.0009458450000238 s 0.0010144106002371 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 / PartOpt / cpu / Primal 0.0009762854000655 s 0.0008878655999069 s 1.10
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 0.0008789718000116 s 0.0009339372000795 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 / DefOpt / cpu / Primal 0.0009403956001733 s 0.0009609147998162 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 / IDefOpt / cpu / Primal 0.0009596344000783 s 0.0009962112000721 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 / cpu / Forward 0.00212805379997 s 0.0022266109999691 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 0.0023627229998055 s 0.0023936325999784 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 / Forward 0.0021819812001012 s 0.0021506333998331 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 / PartOpt / cpu / Forward 0.0021300055998835 s 0.0023729827998067 s 0.90
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 0.0021889466001084 s 0.0022126754000964 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 / Forward 0.002178154000012 s 0.0022777125999709 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 / Forward 0.0021166447999348 s 0.0021534280000196 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 / cpu / PreRev 0.0052805069998612 s 0.0053284488000826 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 / PostRev 0.0053826903999834 s 0.0056835981999938 s 0.95
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 0.0056404042000394 s 0.0052150422000522 s 1.08
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 0.0053996740000911 s 0.0060669299999062 s 0.89
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 0.0056942507998428 s 0.0037044362001324 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 / HLOOpt / cpu / PostRev 0.0060915917999409 s 0.0064265009998962 s 0.95
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 0.0050351031998616 s 0.0037709973999881 s 1.34
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 0.0056500297999264 s 0.0061171409999587 s 0.92
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 0.0055062703999283 s 0.003930622000098 s 1.40
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 0.0059064920000309 s 0.0062224390000665 s 0.95
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IPartOpt / cpu / PreRev 0.0052748797999811 s 0.0037930405999759 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 / IPartOpt / cpu / PostRev 0.006311098399874 s 0.0063594317998649 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 / cpu / BothRev 0.0057970148000094 s 0.0037310412000806 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 / DefOpt / cpu / PreRev 0.0056528485999479 s 0.0063060106000193 s 0.90
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 0.0051075219998892 s 0.0037435317999552 s 1.36
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 0.0047982088000026 s 0.006533680999928 s 0.73
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / PreRev 0.005541553400053 s 0.0037383431999842 s 1.48
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 0.0054808899999443 s 0.0064990833999218 s 0.84
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 0.0055685657999674 s 0.0041281814000285 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 / JaXPipe / cuda / Primal 0.0002733129999999 s 0.000280289 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 0.000273057 s 0.000279648 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 0.000287073 s 0.000287072 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 0.000271938 s 0.000278816 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 / cuda / Primal 0.00027309 s 0.000279456 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 0.000287906 s 0.0002867199999999 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 0.000286818 s 0.000287648 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 0.000557411 s 0.0005568 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 0.000538659 s 0.000538881 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 0.000557987 s 0.000557632 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 0.000556803 s 0.000557569 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 0.000557986 s 0.00055776 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 0.000557091 s 0.000557344 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 0.0005571549999999 s 0.000558176 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 0.001029349 s 0.0010261119999999 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 0.0009874929999999 s 0.00098528 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 0.001025925 s 0.001025441 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 0.0009906939999999 s 0.00098752 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 0.001013798 s 0.001013985 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 0.00103959 s 0.001037569 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 0.0010144059999999 s 0.001011681 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 0.001028582 s 0.001027937 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 0.000978534 s 0.000975713 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 0.0010277819999999 s 0.001026113 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 0.001027686 s 0.001026017 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 0.000976517 s 0.000974785 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 0.001026726 s 0.001026593 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 0.001024198 s 0.001022561 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 0.000962789 s 0.000960865 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 0.001025606 s 0.001022273 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 0.001022086 s 0.0010197449999999 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 0.0010226299999999 s 0.001020033 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 0.00102343 s 0.001020737 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 0.00012368925 s 0.00012457525 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 / Primal 0.0001268485 s 0.000126331 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 / tpu / Primal 0.00015261925 s 0.00015273475 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 / Primal 0.00013451975 s 0.00013435025 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 / Primal 0.0001308499999999 s 0.00013136425 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 / Primal 0.000148303 s 0.000147822 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 / Primal 0.000150772 s 0.0001507285 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 / Forward 0.0002122665 s 0.00021200975 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 / tpu / Forward 0.00026156075 s 0.00026122475 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 / tpu / Forward 0.00021215375 s 0.0002122135 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 / Forward 0.00021817875 s 0.00021845075 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 / Forward 0.00021216525 s 0.000212008 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 / Forward 0.0002182525 s 0.0002185624999999 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 / Forward 0.00021191925 s 0.00021246625 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 / PreRev 0.00035518925 s 0.00035407075 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 / PostRev 0.00025881075 s 0.0002561415 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 0.00035544525 s 0.00035483825 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 / tpu / BothRev 0.00025888725 s 0.0002573317499999 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 0.000355499 s 0.00035495725 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 / tpu / PostRev 0.000291383 s 0.000291605 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 / tpu / BothRev 0.00035556825 s 0.0003548719999999 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 / PreRev 0.0003571985 s 0.00035591275 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 0.0002718 s 0.0002716535 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 / BothRev 0.0003566432499999 s 0.0003560215 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 0.00035545225 s 0.0003549675 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 / PostRev 0.00027468925 s 0.00027230025 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 0.0003558199999999 s 0.00035493875 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 / PreRev 0.00035892625 s 0.0003586415 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 0.00028396525 s 0.00028358525 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 / BothRev 0.00035911925 s 0.0003578824999999 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 0.00035804125 s 0.00035715925 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 / PostRev 0.000301479 s 0.00030180325 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 0.00035760475 s 0.0003571105 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 / Primal 0.002588054 s 0.0009100650000618 s 2.84
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 0.002515597 s 0.0009172600000056 s 2.74
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 0.00274553 s 0.0010144106002371 s 2.71
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 0.002611197 s 0.0008878655999069 s 2.94
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 0.002496444 s 0.0009339372000795 s 2.67
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 0.002525986 s 0.0009609147998162 s 2.63
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 0.002582534 s 0.0009962112000721 s 2.59
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 0.006254365 s 0.0022266109999691 s 2.81
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 0.006500099 s 0.0023936325999784 s 2.72
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 0.0065634249999999 s 0.0021506333998331 s 3.05
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 0.00609938 s 0.0023729827998067 s 2.57
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 0.006369227 s 0.0022126754000964 s 2.88
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 0.006409268 s 0.0022777125999709 s 2.81
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 0.00637275 s 0.0021534280000196 s 2.96
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 0.0094185149999999 s 0.0053284488000826 s 1.77
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 0.009662706 s 0.0056835981999938 s 1.70
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 0.010200475 s 0.0052150422000522 s 1.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / Jax / cpu / BothRev 0.01006602 s 0.0060669299999062 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 / HLOOpt / cpu / PreRev 0.008628042 s 0.0037044362001324 s 2.33
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 0.009969564 s 0.0064265009998962 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 / BothRev 0.008486813 s 0.0037709973999881 s 2.25
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 0.010247412 s 0.0061171409999587 s 1.68
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 0.008846714 s 0.003930622000098 s 2.25
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 0.010920229 s 0.0062224390000665 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 / IPartOpt / cpu / PreRev 0.008761871 s 0.0037930405999759 s 2.31
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 0.010475634 s 0.0063594317998649 s 1.65
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 0.00891557 s 0.0037310412000806 s 2.39
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / DefOpt / cpu / PreRev 0.0085035589999999 s 0.0063060106000193 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 / DefOpt / cpu / PostRev 0.007374144 s 0.0037435317999552 s 1.97
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 0.009419956 s 0.006533680999928 s 1.44
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 0.00901161 s 0.0037383431999842 s 2.41
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 0.009486831 s 0.0064990833999218 s 1.46
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 0.00950567 s 0.0041281814000285 s 2.30
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 0.002146 s 0.0009100650000618 s 2.36
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 0.001604 s 0.0009172600000056 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 / Primal 0.0017259999999999 s 0.0010144106002371 s 1.70
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 0.001566 s 0.0008878655999069 s 1.76
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 0.0017829999999999 s 0.0009339372000795 s 1.91
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 0.001806 s 0.0009609147998162 s 1.88
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 0.00171 s 0.0009962112000721 s 1.72
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 0.004503 s 0.0022266109999691 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 / Jax / cpu / Forward 0.00464 s 0.0023936325999784 s 1.94
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 0.004562 s 0.0021506333998331 s 2.12
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 0.004135 s 0.0023729827998067 s 1.74
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 0.004098 s 0.0022126754000964 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 / DefOpt / cpu / Forward 0.004454 s 0.0022777125999709 s 1.96
llama_dim_288_hidden_dim_768_n_layers_6_n_heads_6_n_kv_heads_6_vocab_size_32000_seq_len_256 / IDefOpt / cpu / Forward 0.004822 s 0.0021534280000196 s 2.24
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 0.009505 s 0.0053284488000826 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 / JaXPipe / cpu / PostRev 0.0134659999999999 s 0.0056835981999938 s 2.37
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 0.011771 s 0.0052150422000522 s 2.26
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 0.0101809999999999 s 0.0060669299999062 s 1.68
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 0.008146 s 0.0037044362001324 s 2.20
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 0.012055 s 0.0064265009998962 s 1.88
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 0.012546 s 0.0037709973999881 s 3.33
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 0.008411 s 0.0061171409999587 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 / PartOpt / cpu / PostRev 0.015811 s 0.003930622000098 s 4.02
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 0.014916 s 0.0062224390000665 s 2.40
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 0.008974 s 0.0037930405999759 s 2.37
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 0.012132 s 0.0063594317998649 s 1.91
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 0.010904 s 0.0037310412000806 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 / DefOpt / cpu / PreRev 0.008979 s 0.0063060106000193 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 / DefOpt / cpu / PostRev 0.008009 s 0.0037435317999552 s 2.14
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 0.01023 s 0.006533680999928 s 1.57
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 0.008224 s 0.0037383431999842 s 2.20
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 0.018951 s 0.0064990833999218 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 / IDefOpt / cpu / BothRev 0.014026 s 0.0041281814000285 s 3.40
scatter_sum / JaXPipe / cpu / Primal 0.000009691940012999112 s 0.000008388459973502904 s 1.16
scatter_sum / Jax / cpu / Primal 0.000008390079992750543 s 0.000007979300016813795 s 1.05
scatter_sum / HLOOpt / cpu / Primal 0.000008338899997397675 s 0.000008007539981917944 s 1.04
scatter_sum / PartOpt / cpu / Primal 0.000008246420020441291 s 0.000008130419992085081 s 1.01
scatter_sum / IPartOpt / cpu / Primal 0.00000902314000086335 s 0.000007912760020190035 s 1.14
scatter_sum / DefOpt / cpu / Primal 0.000008332220031661564 s 0.000008083239990810397 s 1.03
scatter_sum / IDefOpt / cpu / Primal 0.000008313279977301136 s 0.000007748680000077002 s 1.07
scatter_sum / JaXPipe / cpu / Forward 0.000013283899997986737 s 0.000011205040054846904 s 1.19
scatter_sum / Jax / cpu / Forward 0.000013255079993541584 s 0.000011392160013201649 s 1.16
scatter_sum / HLOOpt / cpu / Forward 0.00001409546005561424 s 0.000012204599997858169 s 1.15
scatter_sum / PartOpt / cpu / Forward 0.000013397159927990288 s 0.00001159305999863136 s 1.16
scatter_sum / IPartOpt / cpu / Forward 0.000013451640015773593 s 0.000012415900000632972 s 1.08
scatter_sum / DefOpt / cpu / Forward 0.000013346959995033104 s 0.000011770320006689872 s 1.13
scatter_sum / IDefOpt / cpu / Forward 0.000013023800029259292 s 0.000012068499972883727 s 1.08
scatter_sum / JaXPipe / cpu / PreRev 0.000012799019987141949 s 0.000011716140024873311 s 1.09
scatter_sum / JaXPipe / cpu / PostRev 0.00001250075996722444 s 0.000011641600021903288 s 1.07
scatter_sum / JaXPipe / cpu / BothRev 0.000013153020017853122 s 0.000011614340000960513 s 1.13
scatter_sum / Jax / cpu / BothRev 0.000012477499985834584 s 0.000012184240022179438 s 1.02
scatter_sum / HLOOpt / cpu / PreRev 0.000012764899984176735 s 0.000013060520022918356 s 0.98
scatter_sum / HLOOpt / cpu / PostRev 0.000014301760038506472 s 0.00001368204001664708 s 1.05
scatter_sum / HLOOpt / cpu / BothRev 0.000012500119983087644 s 0.000011856519995490089 s 1.05
scatter_sum / PartOpt / cpu / PreRev 0.000013097240025672362 s 0.000011116820005554472 s 1.18
scatter_sum / PartOpt / cpu / PostRev 0.000012941899958605063 s 0.00001186079993203748 s 1.09
scatter_sum / PartOpt / cpu / BothRev 0.00001324797997767746 s 0.000012269600038052886 s 1.08
scatter_sum / IPartOpt / cpu / PreRev 0.0000126610599909327 s 0.000011251180003455376 s 1.13
scatter_sum / IPartOpt / cpu / PostRev 0.000012610420008059009 s 0.000011199540022062138 s 1.13
scatter_sum / IPartOpt / cpu / BothRev 0.000012635719995159887 s 0.000011617919954005627 s 1.09
scatter_sum / DefOpt / cpu / PreRev 0.000013103600012982495 s 0.000011557839989109198 s 1.13
scatter_sum / DefOpt / cpu / PostRev 0.000012926640038131154 s 0.000011769799975809292 s 1.10
scatter_sum / DefOpt / cpu / BothRev 0.00001202352002110274 s 0.00001178735999019409 s 1.02
scatter_sum / IDefOpt / cpu / PreRev 0.00001224648003699258 s 0.000012042659973303673 s 1.02
scatter_sum / IDefOpt / cpu / PostRev 0.000013015980030104402 s 0.000011558019987205625 s 1.13
scatter_sum / IDefOpt / cpu / BothRev 0.000013016899947615456 s 0.00001182768003673118 s 1.10
scatter_sum / JaXPipe / cuda / Primal 0.000010112 s 0.000009344 s 1.08
scatter_sum / Jax / cuda / Primal 0.000010176 s 0.000009472 s 1.07
scatter_sum / HLOOpt / cuda / Primal 0.000010432 s 0.0000096 s 1.09
scatter_sum / PartOpt / cuda / Primal 0.000009857 s 0.000009727 s 1.01
scatter_sum / IPartOpt / cuda / Primal 0.000009888 s 0.000010175 s 0.97
scatter_sum / DefOpt / cuda / Primal 0.000009824 s 0.000009952 s 0.99
scatter_sum / IDefOpt / cuda / Primal 0.000010048 s 0.000009375 s 1.07
scatter_sum / JaXPipe / cuda / Forward 0.000017408 s 0.000016575 s 1.05
scatter_sum / Jax / cuda / Forward 0.000017568000000000002 s 0.0000168 s 1.05
scatter_sum / HLOOpt / cuda / Forward 0.000017728 s 0.000016864 s 1.05
scatter_sum / PartOpt / cuda / Forward 0.00001728 s 0.000017088 s 1.01
scatter_sum / IPartOpt / cuda / Forward 0.000017312 s 0.000016736 s 1.03
scatter_sum / DefOpt / cuda / Forward 0.00001728 s 0.000016735 s 1.03
scatter_sum / IDefOpt / cuda / Forward 0.000017344 s 0.000016864 s 1.03
scatter_sum / JaXPipe / cuda / PreRev 0.000017121 s 0.000016544 s 1.03
scatter_sum / JaXPipe / cuda / PostRev 0.000017088 s 0.0000168 s 1.02
scatter_sum / JaXPipe / cuda / BothRev 0.000017217 s 0.000016832 s 1.02
scatter_sum / Jax / cuda / BothRev 0.000017344 s 0.000016672 s 1.04
scatter_sum / HLOOpt / cuda / PreRev 0.000017311 s 0.000017216 s 1.01
scatter_sum / HLOOpt / cuda / PostRev 0.000016672 s 0.000016672 s 1
scatter_sum / HLOOpt / cuda / BothRev 0.000016992 s 0.000016096 s 1.06
scatter_sum / PartOpt / cuda / PreRev 0.000017728 s 0.000017024 s 1.04
scatter_sum / PartOpt / cuda / PostRev 0.00001728 s 0.000016255999999999998 s 1.06
scatter_sum / PartOpt / cuda / BothRev 0.000017472 s 0.000016992 s 1.03
scatter_sum / IPartOpt / cuda / PreRev 0.000017824 s 0.000017024 s 1.05
scatter_sum / IPartOpt / cuda / PostRev 0.000017184 s 0.000016063999999999997 s 1.07
scatter_sum / IPartOpt / cuda / BothRev 0.000016993 s 0.000016255999999999998 s 1.05
scatter_sum / DefOpt / cuda / PreRev 0.000017472 s 0.000016864 s 1.04
scatter_sum / DefOpt / cuda / PostRev 0.000017184 s 0.000016224 s 1.06
scatter_sum / DefOpt / cuda / BothRev 0.000017152 s 0.000016992 s 1.01
scatter_sum / IDefOpt / cuda / PreRev 0.000017568000000000002 s 0.00001712 s 1.03
scatter_sum / IDefOpt / cuda / PostRev 0.000017503999999999997 s 0.000016736 s 1.05
scatter_sum / IDefOpt / cuda / BothRev 0.000017217 s 0.000016768000000000003 s 1.03
scatter_sum / JaXPipe / tpu / Primal 0.0000013515250000000002 s 0.00000134455 s 1.01
scatter_sum / Jax / tpu / Primal 0.000001404775 s 0.0000014054 s 1.00
scatter_sum / HLOOpt / tpu / Primal 0.0000013508 s 0.000001344475 s 1.00
scatter_sum / PartOpt / tpu / Primal 0.00000140435 s 0.0000014051749999999998 s 1.00
scatter_sum / IPartOpt / tpu / Primal 0.000001351075 s 0.000001343975 s 1.01
scatter_sum / DefOpt / tpu / Primal 0.000001404725 s 0.0000014051 s 1.00
scatter_sum / IDefOpt / tpu / Primal 0.000001350875 s 0.00000134455 s 1.00
scatter_sum / JaXPipe / tpu / Forward 0.0000027027 s 0.0000027043 s 1.00
scatter_sum / Jax / tpu / Forward 0.0000027273 s 0.0000027198 s 1.00
scatter_sum / HLOOpt / tpu / Forward 0.0000027071250000000004 s 0.000002700425 s 1.00
scatter_sum / PartOpt / tpu / Forward 0.000002693375 s 0.000002684225 s 1.00
scatter_sum / IPartOpt / tpu / Forward 0.0000027083000000000004 s 0.000002698875 s 1.00
scatter_sum / DefOpt / tpu / Forward 0.000002708025 s 0.0000026813 s 1.01
scatter_sum / IDefOpt / tpu / Forward 0.000002702925 s 0.000002702775 s 1.00
scatter_sum / JaXPipe / tpu / PreRev 0.000002691675 s 0.000002679125 s 1.00
scatter_sum / JaXPipe / tpu / PostRev 0.0000026912 s 0.00000268105 s 1.00
scatter_sum / JaXPipe / tpu / BothRev 0.0000027078999999999995 s 0.00000270165 s 1.00
scatter_sum / Jax / tpu / BothRev 0.0000027501750000000003 s 0.000002743075 s 1.00
scatter_sum / HLOOpt / tpu / PreRev 0.00000270745 s 0.000002693675 s 1.01
scatter_sum / HLOOpt / tpu / PostRev 0.0000027428 s 0.0000027449 s 1.00
scatter_sum / HLOOpt / tpu / BothRev 0.00000270705 s 0.00000270395 s 1.00
scatter_sum / PartOpt / tpu / PreRev 0.0000027507749999999995 s 0.0000027547 s 1.00
scatter_sum / PartOpt / tpu / PostRev 0.000002704925 s 0.000002690425 s 1.01
scatter_sum / PartOpt / tpu / BothRev 0.0000027423 s 0.0000027442 s 1.00
scatter_sum / IPartOpt / tpu / PreRev 0.00000271015 s 0.000002699325 s 1.00
scatter_sum / IPartOpt / tpu / PostRev 0.000002740475 s 0.0000027351249999999995 s 1.00
scatter_sum / IPartOpt / tpu / BothRev 0.00000271015 s 0.0000027066 s 1.00
scatter_sum / DefOpt / tpu / PreRev 0.00000274345 s 0.0000027368500000000003 s 1.00
scatter_sum / DefOpt / tpu / PostRev 0.00000270875 s 0.000002691025 s 1.01
scatter_sum / DefOpt / tpu / BothRev 0.000002745075 s 0.00000274225 s 1.00
scatter_sum / IDefOpt / tpu / PreRev 0.0000027136 s 0.000002698475 s 1.01
scatter_sum / IDefOpt / tpu / PostRev 0.0000027437 s 0.0000027346 s 1.00
scatter_sum / IDefOpt / tpu / BothRev 0.0000027078999999999995 s 0.0000027007 s 1.00
scatter_sum / JaXPipe / cpu / Primal 0.000015507999999999998 s 0.000008388459973502904 s 1.85
scatter_sum / Jax / cpu / Primal 0.000016294 s 0.000007979300016813795 s 2.04
scatter_sum / HLOOpt / cpu / Primal 0.000015672 s 0.000008007539981917944 s 1.96
scatter_sum / PartOpt / cpu / Primal 0.000015595 s 0.000008130419992085081 s 1.92
scatter_sum / IPartOpt / cpu / Primal 0.000014969 s 0.000007912760020190035 s 1.89
scatter_sum / DefOpt / cpu / Primal 0.000015357000000000002 s 0.000008083239990810397 s 1.90
scatter_sum / IDefOpt / cpu / Primal 0.000015361 s 0.000007748680000077002 s 1.98
scatter_sum / JaXPipe / cpu / Forward 0.000022607 s 0.000011205040054846904 s 2.02
scatter_sum / Jax / cpu / Forward 0.000023304 s 0.000011392160013201649 s 2.05
scatter_sum / HLOOpt / cpu / Forward 0.000022229 s 0.000012204599997858169 s 1.82
scatter_sum / PartOpt / cpu / Forward 0.000022832 s 0.00001159305999863136 s 1.97
scatter_sum / IPartOpt / cpu / Forward 0.000022863 s 0.000012415900000632972 s 1.84
scatter_sum / DefOpt / cpu / Forward 0.000022796 s 0.000011770320006689872 s 1.94
scatter_sum / IDefOpt / cpu / Forward 0.000022903 s 0.000012068499972883727 s 1.90
scatter_sum / JaXPipe / cpu / PreRev 0.000023882 s 0.000011716140024873311 s 2.04
scatter_sum / JaXPipe / cpu / PostRev 0.000022972 s 0.000011641600021903288 s 1.97
scatter_sum / JaXPipe / cpu / BothRev 0.000023246 s 0.000011614340000960513 s 2.00
scatter_sum / Jax / cpu / BothRev 0.000023875 s 0.000012184240022179438 s 1.96
scatter_sum / HLOOpt / cpu / PreRev 0.000024805 s 0.000013060520022918356 s 1.90
scatter_sum / HLOOpt / cpu / PostRev 0.000022897 s 0.00001368204001664708 s 1.67
scatter_sum / HLOOpt / cpu / BothRev 0.00002388 s 0.000011856519995490089 s 2.01
scatter_sum / PartOpt / cpu / PreRev 0.000025831 s 0.000011116820005554472 s 2.32
scatter_sum / PartOpt / cpu / PostRev 0.000023859000000000003 s 0.00001186079993203748 s 2.01
scatter_sum / PartOpt / cpu / BothRev 0.000023374 s 0.000012269600038052886 s 1.91
scatter_sum / IPartOpt / cpu / PreRev 0.000023782 s 0.000011251180003455376 s 2.11
scatter_sum / IPartOpt / cpu / PostRev 0.000022144 s 0.000011199540022062138 s 1.98
scatter_sum / IPartOpt / cpu / BothRev 0.000022936 s 0.000011617919954005627 s 1.97
scatter_sum / DefOpt / cpu / PreRev 0.00002262 s 0.000011557839989109198 s 1.96
scatter_sum / DefOpt / cpu / PostRev 0.000022693 s 0.000011769799975809292 s 1.93
scatter_sum / DefOpt / cpu / BothRev 0.000022094 s 0.00001178735999019409 s 1.87
scatter_sum / IDefOpt / cpu / PreRev 0.000022394 s 0.000012042659973303673 s 1.86
scatter_sum / IDefOpt / cpu / PostRev 0.000022675 s 0.000011558019987205625 s 1.96
scatter_sum / IDefOpt / cpu / BothRev 0.000022715 s 0.00001182768003673118 s 1.92
scatter_sum / JaXPipe / cpu / Primal 0.000011 s 0.000008388459973502904 s 1.31
scatter_sum / Jax / cpu / Primal 0.000011 s 0.000007979300016813795 s 1.38
scatter_sum / HLOOpt / cpu / Primal 0.00001 s 0.000008007539981917944 s 1.25
scatter_sum / PartOpt / cpu / Primal 0.000011 s 0.000008130419992085081 s 1.35
scatter_sum / IPartOpt / cpu / Primal 0.00001 s 0.000007912760020190035 s 1.26
scatter_sum / DefOpt / cpu / Primal 0.00001 s 0.000008083239990810397 s 1.24
scatter_sum / IDefOpt / cpu / Primal 0.000011 s 0.000007748680000077002 s 1.42
scatter_sum / JaXPipe / cpu / Forward 0.000016 s 0.000011205040054846904 s 1.43
scatter_sum / Jax / cpu / Forward 0.000017 s 0.000011392160013201649 s 1.49
scatter_sum / HLOOpt / cpu / Forward 0.000017 s 0.000012204599997858169 s 1.39
scatter_sum / PartOpt / cpu / Forward 0.000017 s 0.00001159305999863136 s 1.47
scatter_sum / IPartOpt / cpu / Forward 0.000016 s 0.000012415900000632972 s 1.29
scatter_sum / DefOpt / cpu / Forward 0.000017 s 0.000011770320006689872 s 1.44
scatter_sum / IDefOpt / cpu / Forward 0.000017 s 0.000012068499972883727 s 1.41
scatter_sum / JaXPipe / cpu / PreRev 0.000016 s 0.000011716140024873311 s 1.37
scatter_sum / JaXPipe / cpu / PostRev 0.000016 s 0.000011641600021903288 s 1.37
scatter_sum / JaXPipe / cpu / BothRev 0.000017 s 0.000011614340000960513 s 1.46
scatter_sum / Jax / cpu / BothRev 0.000016 s 0.000012184240022179438 s 1.31
scatter_sum / HLOOpt / cpu / PreRev 0.000017 s 0.000013060520022918356 s 1.30
scatter_sum / HLOOpt / cpu / PostRev 0.000017 s 0.00001368204001664708 s 1.24
scatter_sum / HLOOpt / cpu / BothRev 0.000016 s 0.000011856519995490089 s 1.35
scatter_sum / PartOpt / cpu / PreRev 0.000017 s 0.000011116820005554472 s 1.53
scatter_sum / PartOpt / cpu / PostRev 0.000016 s 0.00001186079993203748 s 1.35
scatter_sum / PartOpt / cpu / BothRev 0.000016 s 0.000012269600038052886 s 1.30
scatter_sum / IPartOpt / cpu / PreRev 0.000016 s 0.000011251180003455376 s 1.42
scatter_sum / IPartOpt / cpu / PostRev 0.000016 s 0.000011199540022062138 s 1.43
scatter_sum / IPartOpt / cpu / BothRev 0.000017 s 0.000011617919954005627 s 1.46
scatter_sum / DefOpt / cpu / PreRev 0.000017 s 0.000011557839989109198 s 1.47
scatter_sum / DefOpt / cpu / PostRev 0.000017 s 0.000011769799975809292 s 1.44
scatter_sum / DefOpt / cpu / BothRev 0.000017 s 0.00001178735999019409 s 1.44
scatter_sum / IDefOpt / cpu / PreRev 0.000017 s 0.000012042659973303673 s 1.41
scatter_sum / IDefOpt / cpu / PostRev 0.000017 s 0.000011558019987205625 s 1.47
scatter_sum / IDefOpt / cpu / BothRev 0.000017 s 0.00001182768003673118 s 1.44
slicing / JaXPipe / cpu / Primal 0.000006819260015618056 s 0.000006638120012212312 s 1.03
slicing / Jax / cpu / Primal 0.000006434179967982346 s 0.000006729740034643328 s 0.96
slicing / HLOOpt / cpu / Primal 0.000007045359961921349 s 0.00000629295998805901 s 1.12
slicing / PartOpt / cpu / Primal 0.000006683660030830652 s 0.000006045859981895773 s 1.11
slicing / IPartOpt / cpu / Primal 0.000007106339990059496 s 0.000006430899966289871 s 1.11
slicing / DefOpt / cpu / Primal 0.00000649740001790633 s 0.000006045840000297176 s 1.07
slicing / IDefOpt / cpu / Primal 0.000007279639967237017 s 0.000006516900029964745 s 1.12
slicing / JaXPipe / cpu / Forward 0.000010361760005253016 s 0.00000963860000410932 s 1.08
slicing / Jax / cpu / Forward 0.00001107481999497395 s 0.0000093633799951931 s 1.18
slicing / HLOOpt / cpu / Forward 0.000010677879990907969 s 0.000009715839969430815 s 1.10
slicing / PartOpt / cpu / Forward 0.000010321199997633812 s 0.000009247120015061228 s 1.12
slicing / IPartOpt / cpu / Forward 0.000010158860040974103 s 0.000009682860036264171 s 1.05
slicing / DefOpt / cpu / Forward 0.000009906099976433324 s 0.00000980009998784226 s 1.01
slicing / IDefOpt / cpu / Forward 0.000010809979985424432 s 0.00000932759996430832 s 1.16
slicing / JaXPipe / cpu / PreRev 0.000011285939981462434 s 0.000010271939991071124 s 1.10
slicing / JaXPipe / cpu / PostRev 0.000010716139995565754 s 0.000011063819965784204 s 0.97
slicing / JaXPipe / cpu / BothRev 0.000010794699974212563 s 0.00001043989999743644 s 1.03
slicing / Jax / cpu / BothRev 0.000010968800024784286 s 0.000009840040020208107 s 1.11
slicing / HLOOpt / cpu / PreRev 0.000011066639963246417 s 0.000010213960003966348 s 1.08
slicing / HLOOpt / cpu / PostRev 0.00001245451998329372 s 0.00001261207993593416 s 0.99
slicing / HLOOpt / cpu / BothRev 0.00001096232001145836 s 0.000009878719984044436 s 1.11
slicing / PartOpt / cpu / PreRev 0.000011096759990323335 s 0.000010214859985353542 s 1.09
slicing / PartOpt / cpu / PostRev 0.00001096401992981555 s 0.000009956999983842252 s 1.10
slicing / PartOpt / cpu / BothRev 0.000010745840008894446 s 0.000010311199994248454 s 1.04
slicing / IPartOpt / cpu / PreRev 0.000010903279980993827 s 0.000009981699986383318 s 1.09
slicing / IPartOpt / cpu / PostRev 0.000011227420018258271 s 0.00001045855996380851 s 1.07
slicing / IPartOpt / cpu / BothRev 0.00001044286003889283 s 0.000009998220002671588 s 1.04
slicing / DefOpt / cpu / PreRev 0.000011048419946746434 s 0.000009611019986550672 s 1.15
slicing / DefOpt / cpu / PostRev 0.000011384540002836727 s 0.000009998039986385266 s 1.14
slicing / DefOpt / cpu / BothRev 0.000010823060019902186 s 0.00000963705998401565 s 1.12
slicing / IDefOpt / cpu / PreRev 0.000011020419997294084 s 0.000010339760019633103 s 1.07
slicing / IDefOpt / cpu / PostRev 0.000011029540009985796 s 0.000009949260011126173 s 1.11
slicing / IDefOpt / cpu / BothRev 0.000010545379991526715 s 0.000009970660012186271 s 1.06
slicing / JaXPipe / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / Jax / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / HLOOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / PartOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / IPartOpt / cuda / Primal 0.000001887 s 0.000001888 s 1.00
slicing / DefOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / IDefOpt / cuda / Primal 0.000001887 s 0.000001887 s 1
slicing / JaXPipe / cuda / Forward 0.000009888 s 0.000011328 s 0.87
slicing / Jax / cuda / Forward 0.00000992 s 0.00001088 s 0.91
slicing / HLOOpt / cuda / Forward 0.000010016 s 0.000009823 s 1.02
slicing / PartOpt / cuda / Forward 0.000014688 s 0.000010048 s 1.46
slicing / IPartOpt / cuda / Forward 0.000009824 s 0.000009824 s 1
slicing / DefOpt / cuda / Forward 0.000009888 s 0.000009184 s 1.08
slicing / IDefOpt / cuda / Forward 0.000009952 s 0.000009952 s 1
slicing / JaXPipe / cuda / PreRev 0.000010336 s 0.000009632 s 1.07
slicing / JaXPipe / cuda / PostRev 0.00000992 s 0.000010016 s 0.99
slicing / JaXPipe / cuda / BothRev 0.000010144 s 0.0000096 s 1.06
slicing / Jax / cuda / BothRev 0.000010144 s 0.000009728 s 1.04
slicing / HLOOpt / cuda / PreRev 0.000010048 s 0.000010112 s 0.99
slicing / HLOOpt / cuda / PostRev 0.000010048 s 0.000009728 s 1.03
slicing / HLOOpt / cuda / BothRev 0.000010528 s 0.000009472 s 1.11
slicing / PartOpt / cuda / PreRev 0.000010272 s 0.000009887 s 1.04
slicing / PartOpt / cuda / PostRev 0.000009471 s 0.0000096 s 0.99
slicing / PartOpt / cuda / BothRev 0.000009984 s 0.00000912 s 1.09
slicing / IPartOpt / cuda / PreRev 0.00001024 s 0.00000992 s 1.03
slicing / IPartOpt / cuda / PostRev 0.00000992 s 0.0000096 s 1.03
slicing / IPartOpt / cuda / BothRev 0.000010016 s 0.000009504 s 1.05
slicing / DefOpt / cuda / PreRev 0.000010624 s 0.000009696 s 1.10
slicing / DefOpt / cuda / PostRev 0.000010464 s 0.0000096 s 1.09
slicing / DefOpt / cuda / BothRev 0.000010656 s 0.000010016 s 1.06
slicing / IDefOpt / cuda / PreRev 0.00001024 s 0.000011104 s 0.92
slicing / IDefOpt / cuda / PostRev 0.000009696 s 0.000009504 s 1.02
slicing / IDefOpt / cuda / BothRev 0.000009696 s 0.00000976 s 0.99
slicing / JaXPipe / tpu / Primal 0.00000102395 s 0.00000102565 s 1.00
slicing / Jax / tpu / Primal 9.76e-7 s 9.7245e-7 s 1.00
slicing / HLOOpt / tpu / Primal 0.0000010243 s 0.0000010265 s 1.00
slicing / PartOpt / tpu / Primal 9.71975e-7 s 9.73325e-7 s 1.00
slicing / IPartOpt / tpu / Primal 0.00000102625 s 0.0000010236000000000002 s 1.00
slicing / DefOpt / tpu / Primal 9.72025e-7 s 9.688499999999998e-7 s 1.00
slicing / IDefOpt / tpu / Primal 0.000001028375 s 0.0000010255 s 1.00
slicing / JaXPipe / tpu / Forward 0.0000014082 s 0.000001410875 s 1.00
slicing / Jax / tpu / Forward 0.00000147445 s 0.0000014758 s 1.00
slicing / HLOOpt / tpu / Forward 0.00000152455 s 0.000001517325 s 1.00
slicing / PartOpt / tpu / Forward 0.000001497425 s 0.00000150075 s 1.00
slicing / IPartOpt / tpu / Forward 0.00000151785 s 0.0000015239249999999998 s 1.00
slicing / DefOpt / tpu / Forward 0.000001498575 s 0.0000014960499999999998 s 1.00
slicing / IDefOpt / tpu / Forward 0.0000015256500000000002 s 0.000001517025 s 1.01
slicing / JaXPipe / tpu / PreRev 0.00000256945 s 0.00000257555 s 1.00
slicing / JaXPipe / tpu / PostRev 0.0000025180249999999995 s 0.0000025172250000000003 s 1.00
slicing / JaXPipe / tpu / BothRev 0.0000025796500000000004 s 0.00000258475 s 1.00
slicing / Jax / tpu / BothRev 0.00000254795 s 0.000002545125 s 1.00
slicing / HLOOpt / tpu / PreRev 0.0000025875 s 0.000002599825 s 1.00
slicing / HLOOpt / tpu / PostRev 0.00000253565 s 0.000002543425 s 1.00
slicing / HLOOpt / tpu / BothRev 0.000002586125 s 0.000002587075 s 1.00
slicing / PartOpt / tpu / PreRev 0.0000025471500000000003 s 0.0000025336500000000003 s 1.01
slicing / PartOpt / tpu / PostRev 0.0000025897 s 0.0000025847 s 1.00
slicing / PartOpt / tpu / BothRev 0.00000254265 s 0.000002535825 s 1.00
slicing / IPartOpt / tpu / PreRev 0.000002580175 s 0.0000025883499999999995 s 1.00
slicing / IPartOpt / tpu / PostRev 0.0000025367250000000003 s 0.000002543975 s 1.00
slicing / IPartOpt / tpu / BothRev 0.0000025845 s 0.0000025956 s 1.00
slicing / DefOpt / tpu / PreRev 0.00000254245 s 0.0000025313750000000004 s 1.00
slicing / DefOpt / tpu / PostRev 0.00000258105 s 0.000002591925 s 1.00
slicing / DefOpt / tpu / BothRev 0.00000253195 s 0.0000025369750000000004 s 1.00
slicing / IDefOpt / tpu / PreRev 0.000002592375 s 0.000002578725 s 1.01
slicing / IDefOpt / tpu / PostRev 0.00000253605 s 0.0000025319000000000003 s 1.00
slicing / IDefOpt / tpu / BothRev 0.000002587075 s 0.000002579375 s 1.00
slicing / JaXPipe / cpu / Primal 0.000012655 s 0.000006638120012212312 s 1.91
slicing / Jax / cpu / Primal 0.000012421 s 0.000006729740034643328 s 1.85
slicing / HLOOpt / cpu / Primal 0.000012205 s 0.00000629295998805901 s 1.94
slicing / PartOpt / cpu / Primal 0.000012331 s 0.000006045859981895773 s 2.04
slicing / IPartOpt / cpu / Primal 0.000012321 s 0.000006430899966289871 s 1.92
slicing / DefOpt / cpu / Primal 0.000012232 s 0.000006045840000297176 s 2.02
slicing / IDefOpt / cpu / Primal 0.000012165 s 0.000006516900029964745 s 1.87
slicing / JaXPipe / cpu / Forward 0.000016711 s 0.00000963860000410932 s 1.73
slicing / Jax / cpu / Forward 0.000016488 s 0.0000093633799951931 s 1.76
slicing / HLOOpt / cpu / Forward 0.000016681000000000002 s 0.000009715839969430815 s 1.72
slicing / PartOpt / cpu / Forward 0.000016528999999999997 s 0.000009247120015061228 s 1.79
slicing / IPartOpt / cpu / Forward 0.000016553999999999998 s 0.000009682860036264171 s 1.71
slicing / DefOpt / cpu / Forward 0.000016352 s 0.00000980009998784226 s 1.67
slicing / IDefOpt / cpu / Forward 0.00001632 s 0.00000932759996430832 s 1.75
slicing / JaXPipe / cpu / PreRev 0.000017281000000000003 s 0.000010271939991071124 s 1.68
slicing / JaXPipe / cpu / PostRev 0.000017629 s 0.000011063819965784204 s 1.59
slicing / JaXPipe / cpu / BothRev 0.000017284999999999998 s 0.00001043989999743644 s 1.66
slicing / Jax / cpu / BothRev 0.000017160000000000002 s 0.000009840040020208107 s 1.74
slicing / HLOOpt / cpu / PreRev 0.00001762 s 0.000010213960003966348 s 1.73
slicing / HLOOpt / cpu / PostRev 0.000017289 s 0.00001261207993593416 s 1.37
slicing / HLOOpt / cpu / BothRev 0.000017541 s 0.000009878719984044436 s 1.78
slicing / PartOpt / cpu / PreRev 0.000017332 s 0.000010214859985353542 s 1.70
slicing / PartOpt / cpu / PostRev 0.000017307 s 0.000009956999983842252 s 1.74
slicing / PartOpt / cpu / BothRev 0.000017337 s 0.000010311199994248454 s 1.68
slicing / IPartOpt / cpu / PreRev 0.0000177 s 0.000009981699986383318 s 1.77
slicing / IPartOpt / cpu / PostRev 0.000017229 s 0.00001045855996380851 s 1.65
slicing / IPartOpt / cpu / BothRev 0.000016936 s 0.000009998220002671588 s 1.69
slicing / DefOpt / cpu / PreRev 0.000017175 s 0.000009611019986550672 s 1.79
slicing / DefOpt / cpu / PostRev 0.000017353 s 0.000009998039986385266 s 1.74
slicing / DefOpt / cpu / BothRev 0.000016917 s 0.00000963705998401565 s 1.76
slicing / IDefOpt / cpu / PreRev 0.000017250999999999998 s 0.000010339760019633103 s 1.67
slicing / IDefOpt / cpu / PostRev 0.0000174 s 0.000009949260011126173 s 1.75
slicing / IDefOpt / cpu / BothRev 0.000017415 s 0.000009970660012186271 s 1.75
slicing / JaXPipe / cpu / Primal 0.000008 s 0.000006638120012212312 s 1.21
slicing / Jax / cpu / Primal 0.000008 s 0.000006729740034643328 s 1.19
slicing / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.00000629295998805901 s 1.43
slicing / PartOpt / cpu / Primal 0.000008 s 0.000006045859981895773 s 1.32
slicing / IPartOpt / cpu / Primal 0.000008 s 0.000006430899966289871 s 1.24
slicing / DefOpt / cpu / Primal 0.000008999999999999999 s 0.000006045840000297176 s 1.49
slicing / IDefOpt / cpu / Primal 0.000008 s 0.000006516900029964745 s 1.23
slicing / JaXPipe / cpu / Forward 0.000012 s 0.00000963860000410932 s 1.24
slicing / Jax / cpu / Forward 0.000012 s 0.0000093633799951931 s 1.28
slicing / HLOOpt / cpu / Forward 0.000012 s 0.000009715839969430815 s 1.24
slicing / PartOpt / cpu / Forward 0.000012 s 0.000009247120015061228 s 1.30
slicing / IPartOpt / cpu / Forward 0.000012 s 0.000009682860036264171 s 1.24
slicing / DefOpt / cpu / Forward 0.000012 s 0.00000980009998784226 s 1.22
slicing / IDefOpt / cpu / Forward 0.000012 s 0.00000932759996430832 s 1.29
slicing / JaXPipe / cpu / PreRev 0.000012 s 0.000010271939991071124 s 1.17
slicing / JaXPipe / cpu / PostRev 0.000012 s 0.000011063819965784204 s 1.08
slicing / JaXPipe / cpu / BothRev 0.000012 s 0.00001043989999743644 s 1.15
slicing / Jax / cpu / BothRev 0.000012 s 0.000009840040020208107 s 1.22
slicing / HLOOpt / cpu / PreRev 0.000012 s 0.000010213960003966348 s 1.17
slicing / HLOOpt / cpu / PostRev 0.000012 s 0.00001261207993593416 s 0.95
slicing / HLOOpt / cpu / BothRev 0.000012 s 0.000009878719984044436 s 1.21
slicing / PartOpt / cpu / PreRev 0.000012 s 0.000010214859985353542 s 1.17
slicing / PartOpt / cpu / PostRev 0.000012 s 0.000009956999983842252 s 1.21
slicing / PartOpt / cpu / BothRev 0.000012 s 0.000010311199994248454 s 1.16
slicing / IPartOpt / cpu / PreRev 0.000012 s 0.000009981699986383318 s 1.20
slicing / IPartOpt / cpu / PostRev 0.000013 s 0.00001045855996380851 s 1.24
slicing / IPartOpt / cpu / BothRev 0.000012 s 0.000009998220002671588 s 1.20
slicing / DefOpt / cpu / PreRev 0.000012 s 0.000009611019986550672 s 1.25
slicing / DefOpt / cpu / PostRev 0.000012 s 0.000009998039986385266 s 1.20
slicing / DefOpt / cpu / BothRev 0.000012 s 0.00000963705998401565 s 1.25
slicing / IDefOpt / cpu / PreRev 0.000011 s 0.000010339760019633103 s 1.06
slicing / IDefOpt / cpu / PostRev 0.000012 s 0.000009949260011126173 s 1.21
slicing / IDefOpt / cpu / BothRev 0.000012 s 0.000009970660012186271 s 1.20
sum / JaXPipe / cpu / Primal 0.000008450000004813773 s 0.000007645399964530952 s 1.11
sum / Jax / cpu / Primal 0.000008312480049426086 s 0.000007648059981875122 s 1.09
sum / HLOOpt / cpu / Primal 0.000008433460016021854 s 0.000007830219947209117 s 1.08
sum / PartOpt / cpu / Primal 0.000008630759975858381 s 0.000008145459978550208 s 1.06
sum / IPartOpt / cpu / Primal 0.000008438780032520298 s 0.000008449280003333114 s 1.00
sum / DefOpt / cpu / Primal 0.000008341560023836792 s 0.000007587159980175784 s 1.10
sum / IDefOpt / cpu / Primal 0.000008440999963568174 s 0.0000079674000335217 s 1.06
sum / JaXPipe / cpu / Forward 0.000012301100005061017 s 0.000011516220010889813 s 1.07
sum / Jax / cpu / Forward 0.00001195302005726262 s 0.000011451039999883506 s 1.04
sum / HLOOpt / cpu / Forward 0.000012660220027100878 s 0.000011789640011556912 s 1.07
sum / PartOpt / cpu / Forward 0.000012831799940613564 s 0.000011053859989260672 s 1.16
sum / IPartOpt / cpu / Forward 0.000013076359982733263 s 0.000011822620008388183 s 1.11
sum / DefOpt / cpu / Forward 0.00001253013998393726 s 0.000010920239947154187 s 1.15
sum / IDefOpt / cpu / Forward 0.000011966020028921776 s 0.000011344999993525562 s 1.05
sum / JaXPipe / cpu / PreRev 0.000012481920011850888 s 0.000010926760069196462 s 1.14
sum / JaXPipe / cpu / PostRev 0.000012581159999172088 s 0.000010825520002981649 s 1.16
sum / JaXPipe / cpu / BothRev 0.000011710000017046695 s 0.00001116623997404531 s 1.05
sum / Jax / cpu / BothRev 0.000012064039992765174 s 0.000011224160007259342 s 1.07
sum / HLOOpt / cpu / PreRev 0.000012301279984967553 s 0.000011186540041308036 s 1.10
sum / HLOOpt / cpu / PostRev 0.00001294402000894479 s 0.000012992799984203885 s 1.00
sum / HLOOpt / cpu / BothRev 0.000011619140013863217 s 0.00001103836000766023 s 1.05
sum / PartOpt / cpu / PreRev 0.000012203499982206268 s 0.00001072174000000814 s 1.14
sum / PartOpt / cpu / PostRev 0.000012216259956403518 s 0.000010569659998509453 s 1.16
sum / PartOpt / cpu / BothRev 0.000011873980010932428 s 0.00001155784001639404 s 1.03
sum / IPartOpt / cpu / PreRev 0.000012296260001676271 s 0.000011202260011486942 s 1.10
sum / IPartOpt / cpu / PostRev 0.000011434120006015292 s 0.000011255660010647262 s 1.02
sum / IPartOpt / cpu / BothRev 0.000011997859965049428 s 0.000010993680016326834 s 1.09
sum / DefOpt / cpu / PreRev 0.000011709000036717044 s 0.000011120220005977898 s 1.05
sum / DefOpt / cpu / PostRev 0.00001190643994959828 s 0.000011071139997511637 s 1.08
sum / DefOpt / cpu / BothRev 0.00001156758004981384 s 0.000011151440021421876 s 1.04
sum / IDefOpt / cpu / PreRev 0.000011894500003108988 s 0.000010621079964039382 s 1.12
sum / IDefOpt / cpu / PostRev 0.000012064140009897529 s 0.000011201700044694008 s 1.08
sum / IDefOpt / cpu / BothRev 0.00001203240005452244 s 0.00001039341997966403 s 1.16
sum / JaXPipe / cuda / Primal 0.000002048 s 0.000002048 s 1
sum / Jax / cuda / Primal 0.000002048 s 0.000002048 s 1
sum / HLOOpt / cuda / Primal 0.000002048 s 0.000002047 s 1.00
sum / PartOpt / cuda / Primal 0.000002048 s 0.000002047 s 1.00
sum / IPartOpt / cuda / Primal 0.000002048 s 0.000002048 s 1
sum / DefOpt / cuda / Primal 0.000002047 s 0.000002048 s 1.00
sum / IDefOpt / cuda / Primal 0.000002048 s 0.000002048 s 1
sum / JaXPipe / cuda / Forward 0.000010176 s 0.000010272 s 0.99
sum / Jax / cuda / Forward 0.000010144 s 0.000010048 s 1.01
sum / HLOOpt / cuda / Forward 0.000010337 s 0.000010368 s 1.00
sum / PartOpt / cuda / Forward 0.000010336 s 0.00000992 s 1.04
sum / IPartOpt / cuda / Forward 0.000010144 s 0.000010209 s 0.99
sum / DefOpt / cuda / Forward 0.000010176 s 0.000010016 s 1.02
sum / IDefOpt / cuda / Forward 0.000010752 s 0.000010272 s 1.05
sum / JaXPipe / cuda / PreRev 0.00000976 s 0.000010048 s 0.97
sum / JaXPipe / cuda / PostRev 0.000010177 s 0.000009633 s 1.06
sum / JaXPipe / cuda / BothRev 0.000010272 s 0.000009376 s 1.10
sum / Jax / cuda / BothRev 0.000009984 s 0.0000096 s 1.04
sum / HLOOpt / cuda / PreRev 0.000010464 s 0.000009567 s 1.09
sum / HLOOpt / cuda / PostRev 0.000009504 s 0.00000944 s 1.01
sum / HLOOpt / cuda / BothRev 0.000010144 s 0.000009185 s 1.10
sum / PartOpt / cuda / PreRev 0.000010432 s 0.000010208 s 1.02
sum / PartOpt / cuda / PostRev 0.000010144 s 0.000009824 s 1.03
sum / PartOpt / cuda / BothRev 0.000009984 s 0.000009888 s 1.01
sum / IPartOpt / cuda / PreRev 0.000009953 s 0.000010208 s 0.98
sum / IPartOpt / cuda / PostRev 0.000010592 s 0.000009664 s 1.10
sum / IPartOpt / cuda / BothRev 0.000010624 s 0.000009664 s 1.10
sum / DefOpt / cuda / PreRev 0.000010432 s 0.000009984 s 1.04
sum / DefOpt / cuda / PostRev 0.000010432 s 0.000009664 s 1.08
sum / DefOpt / cuda / BothRev 0.000010271 s 0.000009504 s 1.08
sum / IDefOpt / cuda / PreRev 0.00001024 s 0.000009856 s 1.04
sum / IDefOpt / cuda / PostRev 0.000009952 s 0.000009408 s 1.06
sum / IDefOpt / cuda / BothRev 0.00000992 s 0.000009632 s 1.03
sum / JaXPipe / tpu / Primal 5.105e-7 s 5.1015e-7 s 1.00
sum / Jax / tpu / Primal 5.468750000000001e-7 s 5.471999999999999e-7 s 1.00
sum / HLOOpt / tpu / Primal 5.1015e-7 s 5.10525e-7 s 1.00
sum / PartOpt / tpu / Primal 5.47325e-7 s 5.4695e-7 s 1.00
sum / IPartOpt / tpu / Primal 5.1035e-7 s 5.106750000000001e-7 s 1.00
sum / DefOpt / tpu / Primal 5.46975e-7 s 5.473499999999999e-7 s 1.00
sum / IDefOpt / tpu / Primal 5.10125e-7 s 5.10425e-7 s 1.00
sum / JaXPipe / tpu / Forward 0.000001557925 s 0.000001554225 s 1.00
sum / Jax / tpu / Forward 0.000001503225 s 0.0000015011250000000002 s 1.00
sum / HLOOpt / tpu / Forward 0.0000015425 s 0.000001527725 s 1.01
sum / PartOpt / tpu / Forward 0.000001493875 s 0.000001494675 s 1.00
sum / IPartOpt / tpu / Forward 0.0000015395 s 0.000001532525 s 1.00
sum / DefOpt / tpu / Forward 0.000001502725 s 0.0000014973 s 1.00
sum / IDefOpt / tpu / Forward 0.000001529775 s 0.00000152915 s 1.00
sum / JaXPipe / tpu / PreRev 0.0000010475 s 0.00000104985 s 1.00
sum / JaXPipe / tpu / PostRev 0.0000010948749999999998 s 0.00000108665 s 1.01
sum / JaXPipe / tpu / BothRev 0.000001047575 s 0.000001049175 s 1.00
sum / Jax / tpu / BothRev 0.00000108555 s 0.000001091625 s 0.99
sum / HLOOpt / tpu / PreRev 0.000001047975 s 0.0000010556 s 0.99
sum / HLOOpt / tpu / PostRev 0.0000010948749999999998 s 0.0000010847000000000002 s 1.01
sum / HLOOpt / tpu / BothRev 0.0000010584749999999998 s 0.000001045625 s 1.01
sum / PartOpt / tpu / PreRev 0.0000010864 s 0.0000010876 s 1.00
sum / PartOpt / tpu / PostRev 0.0000010465 s 0.00000104685 s 1.00
sum / PartOpt / tpu / BothRev 0.000001087375 s 0.0000010919 s 1.00
sum / IPartOpt / tpu / PreRev 0.000001051775 s 0.000001051 s 1.00
sum / IPartOpt / tpu / PostRev 0.000001087675 s 0.00000108285 s 1.00
sum / IPartOpt / tpu / BothRev 0.0000010567 s 0.000001049375 s 1.01
sum / DefOpt / tpu / PreRev 0.00000109085 s 0.00000109035 s 1.00
sum / DefOpt / tpu / PostRev 0.0000010586 s 0.000001048725 s 1.01
sum / DefOpt / tpu / BothRev 0.000001091675 s 0.0000010904 s 1.00
sum / IDefOpt / tpu / PreRev 0.0000010529 s 0.000001045875 s 1.01
sum / IDefOpt / tpu / PostRev 0.0000010884 s 0.0000010865 s 1.00
sum / IDefOpt / tpu / BothRev 0.000001056775 s 0.000001051825 s 1.00
sum / JaXPipe / cpu / Primal 0.000014595 s 0.000007645399964530952 s 1.91
sum / Jax / cpu / Primal 0.000014665 s 0.000007648059981875122 s 1.92
sum / HLOOpt / cpu / Primal 0.000014553 s 0.000007830219947209117 s 1.86
sum / PartOpt / cpu / Primal 0.000014714 s 0.000008145459978550208 s 1.81
sum / IPartOpt / cpu / Primal 0.000014386 s 0.000008449280003333114 s 1.70
sum / DefOpt / cpu / Primal 0.000014518 s 0.000007587159980175784 s 1.91
sum / IDefOpt / cpu / Primal 0.000014415 s 0.0000079674000335217 s 1.81
sum / JaXPipe / cpu / Forward 0.000020047 s 0.000011516220010889813 s 1.74
sum / Jax / cpu / Forward 0.000019841 s 0.000011451039999883506 s 1.73
sum / HLOOpt / cpu / Forward 0.000019476 s 0.000011789640011556912 s 1.65
sum / PartOpt / cpu / Forward 0.00001983 s 0.000011053859989260672 s 1.79
sum / IPartOpt / cpu / Forward 0.000019756 s 0.000011822620008388183 s 1.67
sum / DefOpt / cpu / Forward 0.000019947 s 0.000010920239947154187 s 1.83
sum / IDefOpt / cpu / Forward 0.000019618 s 0.000011344999993525562 s 1.73
sum / JaXPipe / cpu / PreRev 0.000018566 s 0.000010926760069196462 s 1.70
sum / JaXPipe / cpu / PostRev 0.000018646 s 0.000010825520002981649 s 1.72
sum / JaXPipe / cpu / BothRev 0.000019077 s 0.00001116623997404531 s 1.71
sum / Jax / cpu / BothRev 0.000018416 s 0.000011224160007259342 s 1.64
sum / HLOOpt / cpu / PreRev 0.000019001 s 0.000011186540041308036 s 1.70
sum / HLOOpt / cpu / PostRev 0.000018693 s 0.000012992799984203885 s 1.44
sum / HLOOpt / cpu / BothRev 0.000018437 s 0.00001103836000766023 s 1.67
sum / PartOpt / cpu / PreRev 0.000018654 s 0.00001072174000000814 s 1.74
sum / PartOpt / cpu / PostRev 0.0000185 s 0.000010569659998509453 s 1.75
sum / PartOpt / cpu / BothRev 0.000018684 s 0.00001155784001639404 s 1.62
sum / IPartOpt / cpu / PreRev 0.000018369 s 0.000011202260011486942 s 1.64
sum / IPartOpt / cpu / PostRev 0.000018923 s 0.000011255660010647262 s 1.68
sum / IPartOpt / cpu / BothRev 0.000019362 s 0.000010993680016326834 s 1.76
sum / DefOpt / cpu / PreRev 0.000019175 s 0.000011120220005977898 s 1.72
sum / DefOpt / cpu / PostRev 0.000019281 s 0.000011071139997511637 s 1.74
sum / DefOpt / cpu / BothRev 0.000019154 s 0.000011151440021421876 s 1.72
sum / IDefOpt / cpu / PreRev 0.000018657 s 0.000010621079964039382 s 1.76
sum / IDefOpt / cpu / PostRev 0.000019353 s 0.000011201700044694008 s 1.73
sum / IDefOpt / cpu / BothRev 0.000018804 s 0.00001039341997966403 s 1.81
sum / JaXPipe / cpu / Primal 0.00001 s 0.000007645399964530952 s 1.31
sum / Jax / cpu / Primal 0.000011 s 0.000007648059981875122 s 1.44
sum / HLOOpt / cpu / Primal 0.000008999999999999999 s 0.000007830219947209117 s 1.15
sum / PartOpt / cpu / Primal 0.00001 s 0.000008145459978550208 s 1.23
sum / IPartOpt / cpu / Primal 0.00001 s 0.000008449280003333114 s 1.18
sum / DefOpt / cpu / Primal 0.00001 s 0.000007587159980175784 s 1.32
sum / IDefOpt / cpu / Primal 0.00001 s 0.0000079674000335217 s 1.26
sum / JaXPipe / cpu / Forward 0.000014 s 0.000011516220010889813 s 1.22
sum / Jax / cpu / Forward 0.000014 s 0.000011451039999883506 s 1.22
sum / HLOOpt / cpu / Forward 0.000013 s 0.000011789640011556912 s 1.10
sum / PartOpt / cpu / Forward 0.000014 s 0.000011053859989260672 s 1.27
sum / IPartOpt / cpu / Forward 0.000014 s 0.000011822620008388183 s 1.18
sum / DefOpt / cpu / Forward 0.000013 s 0.000010920239947154187 s 1.19
sum / IDefOpt / cpu / Forward 0.000013 s 0.000011344999993525562 s 1.15
sum / JaXPipe / cpu / PreRev 0.000013 s 0.000010926760069196462 s 1.19
sum / JaXPipe / cpu / PostRev 0.000013 s 0.000010825520002981649 s 1.20
sum / JaXPipe / cpu / BothRev 0.000013 s 0.00001116623997404531 s 1.16
sum / Jax / cpu / BothRev 0.000013 s 0.000011224160007259342 s 1.16
sum / HLOOpt / cpu / PreRev 0.000013 s 0.000011186540041308036 s 1.16
sum / HLOOpt / cpu / PostRev 0.000013 s 0.000012992799984203885 s 1.00
sum / HLOOpt / cpu / BothRev 0.000013 s 0.00001103836000766023 s 1.18
sum / PartOpt / cpu / PreRev 0.000013 s 0.00001072174000000814 s 1.21
sum / PartOpt / cpu / PostRev 0.000014 s 0.000010569659998509453 s 1.32
sum / PartOpt / cpu / BothRev 0.000014 s 0.00001155784001639404 s 1.21
sum / IPartOpt / cpu / PreRev 0.000013 s 0.000011202260011486942 s 1.16
sum / IPartOpt / cpu / PostRev 0.000014 s 0.000011255660010647262 s 1.24
sum / IPartOpt / cpu / BothRev 0.000014 s 0.000010993680016326834 s 1.27
sum / DefOpt / cpu / PreRev 0.000014 s 0.000011120220005977898 s 1.26
sum / DefOpt / cpu / PostRev 0.000013 s 0.000011071139997511637 s 1.17
sum / DefOpt / cpu / BothRev 0.000014 s 0.000011151440021421876 s 1.26
sum / IDefOpt / cpu / PreRev 0.000014 s 0.000010621079964039382 s 1.32
sum / IDefOpt / cpu / PostRev 0.000014 s 0.000011201700044694008 s 1.25
sum / IDefOpt / cpu / BothRev 0.000014 s 0.00001039341997966403 s 1.35
value_and_grad / JaXPipe / cpu / Primal 0.00001530994001768704 s 0.000014406580003196723 s 1.06
value_and_grad / Jax / cpu / Primal 0.000015586899917252594 s 0.00001402247997248196 s 1.11
value_and_grad / HLOOpt / cpu / Primal 0.000015440459992532852 s 0.000013926480032750988 s 1.11
value_and_grad / PartOpt / cpu / Primal 0.000015165339973464145 s 0.000013946120016044003 s 1.09
value_and_grad / IPartOpt / cpu / Primal 0.000014719139981025365 s 0.000013939880000179985 s 1.06
value_and_grad / DefOpt / cpu / Primal 0.000014846120020592934 s 0.000014367180010594894 s 1.03
value_and_grad / IDefOpt / cpu / Primal 0.000014155839980958264 s 0.000013935539982412593 s 1.02
value_and_grad / JaXPipe / cuda / Primal 0.000033856 s 0.000033759999999999995 s 1.00
value_and_grad / Jax / cuda / Primal 0.000032864 s 0.000033056 s 0.99
value_and_grad / HLOOpt / cuda / Primal 0.000032608 s 0.000032416 s 1.01
value_and_grad / PartOpt / cuda / Primal 0.000033472 s 0.000033056 s 1.01
value_and_grad / IPartOpt / cuda / Primal 0.000033952 s 0.00003264 s 1.04
value_and_grad / DefOpt / cuda / Primal 0.000033376 s 0.000032672 s 1.02
value_and_grad / IDefOpt / cuda / Primal 0.0000336 s 0.000032064 s 1.05
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 0.000023357 s 0.000014406580003196723 s 1.62
value_and_grad / Jax / cpu / Primal 0.00002291 s 0.00001402247997248196 s 1.63
value_and_grad / HLOOpt / cpu / Primal 0.000022749 s 0.000013926480032750988 s 1.63
value_and_grad / PartOpt / cpu / Primal 0.000023025 s 0.000013946120016044003 s 1.65
value_and_grad / IPartOpt / cpu / Primal 0.000023162 s 0.000013939880000179985 s 1.66
value_and_grad / DefOpt / cpu / Primal 0.000023303 s 0.000014367180010594894 s 1.62
value_and_grad / IDefOpt / cpu / Primal 0.000023254 s 0.000013935539982412593 s 1.67
value_and_grad / JaXPipe / cpu / Primal 0.000016 s 0.000014406580003196723 s 1.11
value_and_grad / Jax / cpu / Primal 0.000017 s 0.00001402247997248196 s 1.21
value_and_grad / HLOOpt / cpu / Primal 0.000017 s 0.000013926480032750988 s 1.22
value_and_grad / PartOpt / cpu / Primal 0.000017 s 0.000013946120016044003 s 1.22
value_and_grad / IPartOpt / cpu / Primal 0.000017 s 0.000013939880000179985 s 1.22
value_and_grad / DefOpt / cpu / Primal 0.000016 s 0.000014367180010594894 s 1.11
value_and_grad / IDefOpt / cpu / Primal 0.000017 s 0.000013935539982412593 s 1.22
jaxmd20 / JaXPipe / cuda / Primal 0.001527753 s 0.001438656 s 1.06
jaxmd20 / Jax / cuda / Primal 0.0014146 s 0.001457794 s 0.97
jaxmd20 / HLOOpt / cuda / Primal 0.001411369 s 0.001319074 s 1.07
jaxmd20 / PartOpt / cuda / Primal 0.001331752 s 0.001330945 s 1.00
jaxmd20 / IPartOpt / cuda / Primal 0.001323784 s 0.001306946 s 1.01
jaxmd20 / DefOpt / cuda / Primal 0.000927942 s 0.0009199049999999 s 1.01
jaxmd20 / IDefOpt / cuda / Primal 0.000949094 s 0.000942625 s 1.01
jaxmd20 / JaXPipe / cuda / Forward 0.001560042 s 0.001553729 s 1.00
jaxmd20 / Jax / cuda / Forward 0.001767243 s 0.001784097 s 0.99
jaxmd20 / HLOOpt / cuda / Forward 0.001661354 s 0.001622465 s 1.02
jaxmd20 / PartOpt / cuda / Forward 0.001648554 s 0.001625025 s 1.01
jaxmd20 / IPartOpt / cuda / Forward 0.001625162 s 0.001625729 s 1.00
jaxmd20 / DefOpt / cuda / Forward 0.001644938 s 0.001640737 s 1.00
jaxmd20 / IDefOpt / cuda / Forward 0.001624938 s 0.001621602 s 1.00
jaxmd20 / JaXPipe / cuda / PreRev 0.002663376 s 0.002683234 s 0.99
jaxmd20 / JaXPipe / cuda / PostRev 0.005340639 s 0.005323364 s 1.00
jaxmd20 / JaXPipe / cuda / BothRev 0.002705359 s 0.00268874 s 1.01
jaxmd20 / Jax / cuda / BothRev 0.00532115 s 0.005367429 s 0.99
jaxmd20 / HLOOpt / cuda / PreRev 0.0027617759999999 s 0.002735042 s 1.01
jaxmd20 / HLOOpt / cuda / PostRev 0.005299455 s 0.0053471399999999 s 0.99
jaxmd20 / HLOOpt / cuda / BothRev 0.002728431 s 0.002724066 s 1.00
jaxmd20 / PartOpt / cuda / PreRev 0.002871985 s 0.002834625 s 1.01
jaxmd20 / PartOpt / cuda / PostRev 0.005438175 s 0.005432037 s 1.00
jaxmd20 / PartOpt / cuda / BothRev 0.002761296 s 0.002772929 s 1.00
jaxmd20 / IPartOpt / cuda / PreRev 0.002808881 s 0.002814114 s 1.00
jaxmd20 / IPartOpt / cuda / PostRev 0.005422271 s 0.005404004 s 1.00
jaxmd20 / IPartOpt / cuda / BothRev 0.002798289 s 0.0027513299999999 s 1.02
jaxmd20 / DefOpt / cuda / PreRev 0.002823441 s 0.002829603 s 1.00
jaxmd20 / DefOpt / cuda / PostRev 0.0027812009999999 s 0.002740963 s 1.01
jaxmd20 / DefOpt / cuda / BothRev 0.002774928 s 0.002774787 s 1.00
jaxmd20 / IDefOpt / cuda / PreRev 0.002832465 s 0.0027995549999999 s 1.01
jaxmd20 / IDefOpt / cuda / PostRev 0.002324493 s 0.002319777 s 1.00
jaxmd20 / IDefOpt / cuda / BothRev 0.002759952 s 0.002749858 s 1.00
jaxmd20 / JaXPipe / tpu / Primal 0.009274133125 s 0.00927314125 s 1.00
jaxmd20 / Jax / tpu / Primal 0.0092651675 s 0.009277410625 s 1.00
jaxmd20 / HLOOpt / tpu / Primal 0.0091547318749999 s 0.0091523299999999 s 1.00
jaxmd20 / PartOpt / tpu / Primal 0.00919861 s 0.0092039425 s 1.00
jaxmd20 / IPartOpt / tpu / Primal 0.009201684375 s 0.009203678125 s 1.00
jaxmd20 / DefOpt / tpu / Primal 0.00879213875 s 0.008808284375 s 1.00
jaxmd20 / IDefOpt / tpu / Primal 0.00870023 s 0.0087018675 s 1.00
jaxmd20 / JaXPipe / tpu / Forward 0.0174169743749999 s 0.017406293125 s 1.00
jaxmd20 / Jax / tpu / Forward 0.01872920125 s 0.018734816875 s 1.00
jaxmd20 / HLOOpt / tpu / Forward 0.01739496625 s 0.017393216875 s 1.00
jaxmd20 / PartOpt / tpu / Forward 0.0174072468749999 s 0.0174203675 s 1.00
jaxmd20 / IPartOpt / tpu / Forward 0.017413354375 s 0.017406431875 s 1.00
jaxmd20 / DefOpt / tpu / Forward 0.017415673125 s 0.017418041875 s 1.00
jaxmd20 / IDefOpt / tpu / Forward 0.01741740875 s 0.017410129375 s 1.00
jaxmd20 / JaXPipe / tpu / PreRev 0.0254554075 s 0.0254688181249999 s 1.00
jaxmd20 / JaXPipe / tpu / PostRev 0.021893141875 s 0.0218769525 s 1.00
jaxmd20 / JaXPipe / tpu / BothRev 0.0254694231249999 s 0.025468754375 s 1.00
jaxmd20 / Jax / tpu / BothRev 0.021573645625 s 0.021861596875 s 0.99
jaxmd20 / HLOOpt / tpu / PreRev 0.02558743125 s 0.0255831725 s 1.00
jaxmd20 / HLOOpt / tpu / PostRev 0.02073296375 s 0.020709916875 s 1.00
jaxmd20 / HLOOpt / tpu / BothRev 0.025680845625 s 0.02568878875 s 1.00
jaxmd20 / PartOpt / tpu / PreRev 0.025506155 s 0.0254527875 s 1.00
jaxmd20 / PartOpt / tpu / PostRev 0.02124002 s 0.02152217 s 0.99
jaxmd20 / PartOpt / tpu / BothRev 0.025597825 s 0.0255547325 s 1.00
jaxmd20 / IPartOpt / tpu / PreRev 0.02547697875 s 0.025477075625 s 1.00
jaxmd20 / IPartOpt / tpu / PostRev 0.021542845 s 0.021248858125 s 1.01
jaxmd20 / IPartOpt / tpu / BothRev 0.02555248625 s 0.0255718356249999 s 1.00
jaxmd20 / DefOpt / tpu / PreRev 0.025507525 s 0.025456913125 s 1.00
jaxmd20 / DefOpt / tpu / PostRev 0.018804655 s 0.0188292325 s 1.00
jaxmd20 / DefOpt / tpu / BothRev 0.025595266875 s 0.02555605125 s 1.00
jaxmd20 / IDefOpt / tpu / PreRev 0.025477585625 s 0.02547475125 s 1.00
jaxmd20 / IDefOpt / tpu / PostRev 0.01834839625 s 0.018316490625 s 1.00
jaxmd20 / IDefOpt / tpu / BothRev 0.025557825625 s 0.0255632825 s 1.00
jaxmd40 / JaXPipe / cpu / Primal 0.066142325 s 0.072844339 s 0.91
jaxmd40 / Jax / cpu / Primal 0.066405879 s 0.085290419 s 0.78
jaxmd40 / HLOOpt / cpu / Primal 0.094422496 s 0.089723403 s 1.05
jaxmd40 / PartOpt / cpu / Primal 0.064153144 s 0.074612173 s 0.86
jaxmd40 / IPartOpt / cpu / Primal 0.071286772 s 0.066914133 s 1.07
jaxmd40 / DefOpt / cpu / Primal 0.0928343299999999 s 0.093375959 s 0.99
jaxmd40 / IDefOpt / cpu / Primal 0.093839144 s 0.0910838289999999 s 1.03
jaxmd40 / JaXPipe / cpu / Forward 0.177165219 s 0.170574911 s 1.04
jaxmd40 / Jax / cpu / Forward 0.0941033 s 0.091175984 s 1.03
jaxmd40 / HLOOpt / cpu / Forward 0.177352652 s 0.172308492 s 1.03
jaxmd40 / PartOpt / cpu / Forward 0.17136242 s 0.176623812 s 0.97
jaxmd40 / IPartOpt / cpu / Forward 0.169377776 s 0.174147301 s 0.97
jaxmd40 / DefOpt / cpu / Forward 0.172576279 s 0.173570474 s 0.99
jaxmd40 / IDefOpt / cpu / Forward 0.16673564 s 0.172948483 s 0.96
jaxmd40 / JaXPipe / cpu / PreRev 0.244626449 s 0.246458502 s 0.99
jaxmd40 / JaXPipe / cpu / PostRev 0.145075236 s 0.145292605 s 1.00
jaxmd40 / JaXPipe / cpu / BothRev 0.2442809739999999 s 0.232384171 s 1.05
jaxmd40 / Jax / cpu / BothRev 0.137181209 s 0.144972575 s 0.95
jaxmd40 / HLOOpt / cpu / PreRev 0.229141182 s 0.209976645 s 1.09
jaxmd40 / HLOOpt / cpu / PostRev 0.181059161 s 0.167030558 s 1.08
jaxmd40 / HLOOpt / cpu / BothRev 0.2674010799999999 s 0.240362218 s 1.11
jaxmd40 / PartOpt / cpu / PreRev 0.238577839 s 0.225781603 s 1.06
jaxmd40 / PartOpt / cpu / PostRev 0.132009442 s 0.133505024 s 0.99
jaxmd40 / PartOpt / cpu / BothRev 0.256638552 s 0.2396432289999999 s 1.07
jaxmd40 / IPartOpt / cpu / PreRev 0.228185088 s 0.210954968 s 1.08
jaxmd40 / IPartOpt / cpu / PostRev 0.133703921 s 0.129907668 s 1.03
jaxmd40 / IPartOpt / cpu / BothRev 0.26255486 s 0.2442911049999999 s 1.07
jaxmd40 / DefOpt / cpu / PreRev 0.227576409 s 0.207436634 s 1.10
jaxmd40 / DefOpt / cpu / PostRev 0.1840126329999999 s 0.160227876 s 1.15
jaxmd40 / DefOpt / cpu / BothRev 0.278461413 s 0.266801465 s 1.04
jaxmd40 / IDefOpt / cpu / PreRev 0.228535433 s 0.213239064 s 1.07
jaxmd40 / IDefOpt / cpu / PostRev 0.180361058 s 0.166438509 s 1.08
jaxmd40 / IDefOpt / cpu / BothRev 0.259357769 s 0.2551048099999999 s 1.02
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / JaXPipe / cuda / Primal 1.7081016569999998 s 1.702317745 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / Jax / cuda / Primal 1.7101229269999998 s 1.705465397 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / HLOOpt / cuda / Primal 1.72084421 s 1.716033577 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / PartOpt / cuda / Primal 1.701923054 s 1.697429808 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IPartOpt / cuda / Primal 1.698974848 s 1.694712204 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / DefOpt / cuda / Primal 1.669963967 s 1.665316659 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IDefOpt / cuda / Primal 1.91946122 s 1.915802135 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / JaXPipe / tpu / Primal 3.038655205625 s 3.03911668 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / Jax / tpu / Primal 3.0391822843750003 s 3.0396639975 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / HLOOpt / tpu / Primal 3.1214667675 s 3.122058054375 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / PartOpt / tpu / Primal 3.05996404625 s 3.06050589875 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IPartOpt / tpu / Primal 3.060122295625 s 3.060716875 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / DefOpt / tpu / Primal 2.102430564375 s 2.1026504125 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_24_outer_steps_4 / IDefOpt / tpu / Primal 2.94797121125 s 2.94873416125 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / JaXPipe / cpu / Primal 6.6896081270000005 s 6.26335534 s 1.07
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / Jax / cpu / Primal 6.387158048 s 6.275570235999999 s 1.02
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / HLOOpt / cpu / Primal 6.420210429 s 6.183279134 s 1.04
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / PartOpt / cpu / Primal 6.347112563 s 6.324766463 s 1.00
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / IPartOpt / cpu / Primal 6.318509171 s 6.510950545 s 0.97
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / DefOpt / cpu / Primal 2.57408264 s 2.547048578 s 1.01
neuralgcm_v1/deterministic_2_8_deg_inner_steps_2_outer_steps_2 / IDefOpt / cpu / Primal 6.834233488 s 6.79261008 s 1.01

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

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants