← BBW graph · Download audit JSON

UnbiasedStrata: all 86 task outcomes

Provisional engine result. The known parallel tool-call defect makes this unsuitable as a certified clean benchmark. The original ledger and archives are retained. Corrected grading logs were added to the Minesweeper task bundle.

All six timeouts: raw 10/10, minus five points, recorded 5/10. Minesweeper was regraded from its saved submission after correcting the evaluator import: 0 to 10/10. One client timeout was recovered by Pi; its attempt did not reach the serving proxy. The bundled Pi launcher bypassed Regina’s bash timeout patch: rows record 120 seconds, but no default cap was active. The 900/1800-second task limits and their five-point deductions were verified.

TaskStatusRaw gradeTimeout deductionCorrected scoreSeconds / budgetAPI errorsChecks
api-contract-driftcompleted5.0/5.00.05.043.7 / 9000passed
comfyui-run-workflowcompleted10.0/10.00.010.047.5 / 9000passed
comfyui-sdcompleted10.0/10.00.010.066.8 / 9000passed
comfyui-sd-loracompleted3.0/10.00.03.0399.4 / 9000passed
comfyui-sd-sweepcompleted10.0/10.00.010.0156.2 / 9000passed
comfyui-sd-twostagecompleted10.0/10.00.010.0452.1 / 9000passed
comfyui-sweepcompleted10.0/10.00.010.0132.0 / 9000passed
conv-dim-buildcompleted1.0/4.00.01.034.5 / 9000passed
conv-dim-fixcompleted3.0/3.00.03.030.2 / 9000passed
conv-dim-stridecompleted4.0/4.00.04.033.1 / 9000passed
countdown-gamecompleted3.0/3.00.03.010.6 / 9000passed
countdown-hardcompleted4.0/4.00.04.09.0 / 9000passed
crypto-encrypt-blobcompleted7.0/7.00.07.019.6 / 9000passed
crypto-encrypt-blob-no-appendcompleted11.0/11.00.011.043.3 / 9000passed
crypto-password-storecompleted6.0/6.00.06.056.5 / 9000passed
file-ringcompleted3.0/3.00.03.012.9 / 9000passed
file-ring-checksumcompleted9.0/9.00.09.028.6 / 9000passed
file-ring-creativecompleted4.0/4.00.04.011.6 / 9000passed
file-ring-creative-actcuecompleted4.0/4.00.04.09.9 / 9000passed
file-ring-creative-noexamplecompleted4.0/4.00.04.09.2 / 9000passed
file-ring-creative-v2completed4.0/4.00.04.06.3 / 9000passed
file-ring-impossiblecompleted3.0/3.00.03.020.9 / 9000passed
file-ring-resolvecompleted6.0/6.00.06.024.0 / 9000passed
file-ring-traversecompleted6.0/6.00.06.024.3 / 9000passed
fmtspec-minicompleted10.0/10.00.010.0167.3 / 9000passed
framed-kvcompleted10.0/10.00.010.0327.5 / 9000passed
jsoncanoncompleted10.0/10.00.010.0136.2 / 9000passed
kv-stress-distractor-primecompleted4.0/4.00.04.019.6 / 9000passed
kv-stress-longctx-logiccompleted2.0/2.00.02.051.7 / 9000passed
kv-stress-longctx-pricingcompleted8.0/8.00.08.046.2 / 9000passed
kv-stress-resume-calccompleted5.0/5.00.05.022.9 / 9000passed
memcached-bisect-buriedcompleted10.0/10.00.010.057.0 / 9000passed
memcached-bisect-deepcompleted10.0/10.00.010.0129.3 / 9000passed
memcached-bisect-easycompleted10.0/10.00.010.052.3 / 9000passed
memcached-op-isprimecompleted10.0/10.00.010.0337.4 / 9000passed
memcached-op-multcompleted2.0/10.00.02.019.2 / 9000passed
memcached-op-sayhicompleted10.0/10.00.010.039.6 / 9000passed
minesweeper-servercompleted0.0/10.00.010.0187.7 / 9000passed
minesweeper3d-servercompleted10.0/10.00.010.0181.2 / 9000passed
no-op-trapcompleted4.0/4.00.04.09.3 / 9000passed
numba-stencil-modescompleted10.0/10.00.010.01364.5 / 18000passed
patmatch-basiccompleted10.0/10.00.010.059.4 / 9000passed
patmatch-classestimeout10.0/10.05.05.0901.7 / 9000passed
patmatch-rulescompleted10.0/10.00.010.0189.1 / 18000passed
prime-add-testcompleted2.0/2.00.02.012.2 / 9000passed
prime-buildcompleted8.0/8.00.08.033.7 / 9000passed
prime-factorcompleted8.0/8.00.08.034.4 / 9000passed
prime-factor-cachetimeout10.0/10.05.05.0903.0 / 9000passed
prime-factor-fastcompleted10.0/10.00.010.054.8 / 9000passed
prime-frontendcompleted10.0/10.00.010.041.8 / 9000passed
prime-test-suitecompleted7.0/7.00.07.027.7 / 9000passed
prime-test-suite-faircompleted6.0/6.00.06.028.3 / 9000passed
prime-test-suite-minimalcompleted6.0/6.00.06.0139.8 / 9000passed
prime-test-suite-oldcompleted6.0/6.00.06.0130.7 / 9000passed
prime-test-suite-opuscompleted6.0/6.00.06.0106.0 / 9000passed
psim-arccompleted10.0/10.00.010.0116.5 / 9000passed
psim-basiccompleted10.0/10.00.010.077.8 / 9000passed
psim-blackboxcompleted10.0/10.00.010.0619.5 / 18000passed
psim-blackbox-sealedcompleted8.667/10.00.08.6671647.9 / 18000passed
psim-blackbox-xlcompleted10.0/10.00.010.01637.2 / 18000passed
psim-collidetimeout10.0/10.05.05.0902.1 / 9000passed
psim-fixbugscompleted10.0/10.00.010.0211.3 / 9000passed
psim-nbodycompleted10.0/10.00.010.091.6 / 18000passed
psim-nbody-tersetimeout10.0/10.05.05.0901.9 / 9000passed
py2c-calccompleted10.0/10.00.010.01671.1 / 18000passed
py2c-calcvarstimeout10.0/10.05.05.01801.7 / 18000passed
py2c-minivmcompleted10.0/10.00.010.0568.7 / 9000passed
refactor-renamecompleted6.0/6.00.06.029.2 / 9000passed
regex-golfcompleted20.0/20.00.020.018.2 / 9000passed
rex-blackbox-sealedtimeout10.0/10.05.05.01802.1 / 18000passed
sql-join-ordercompleted5.0/5.00.05.026.7 / 9000passed
statefile-migratecompleted10.0/10.00.010.061.0 / 9000passed
strided-conv-inversecompleted4.0/4.00.04.018.1 / 9000passed
swebench-django-acreatecompleted4.67/10.00.04.67120.7 / 9001passed
swebench-django-escapeseqcompleted6.0/10.00.06.0115.1 / 9000passed
swebench-django-serialm2mcompleted8.440000000000001/10.00.08.440000000000001887.6 / 18000passed
swebench-pytest-caplogcompleted10.0/10.00.010.0221.9 / 9000passed
swebench-pytest-mrocompleted2.0/10.00.02.046.2 / 9000passed
swebench-pytest-skippdbcompleted3.0/10.00.03.016.1 / 9000passed
tcp-renocompleted2.0/10.00.02.0181.7 / 18000passed
transformer-shapecompleted5.0/5.00.05.019.3 / 9000passed
udiff-applycompleted10.0/10.00.010.0180.1 / 9000passed
verify-before-claimcompleted4.0/4.00.04.011.5 / 9000passed
vm-blackbox-sealedcompleted10.0/10.00.010.01689.5 / 18000passed
write-hellocompleted1.0/1.00.01.06.2 / 9000passed
write-hello-wordcompleted1.0/1.00.01.08.6 / 9000passed