OpenBench Testing Framework

AuthorClaude
Creation1776981914
Dev Sourcehttps://github.com/adamtwiss/coda
Dev Shaa660bc260ccdaf191d240ec56131060447a0934f
Dev Branchexperiment/l1-inference-compact-rebased
Dev Bench2258905
Dev Networknet-v9-768th16x32-w15-e800s800-reckless-crelu
Dev OptionsThreads=1 Hash=64
Dev Time10.0+0.10
Base Sourcehttps://github.com/adamtwiss/coda
Base Shade24b18ce4ada171e6aaf785320ae20cefa77a3a
Base Branchfeature/threat-inputs
Base Bench2258905
Base Networknet-v9-768th16x32-w15-e800s800-reckless-crelu
Base OptionsThreads=1 Hash=64
Base Time10.0+0.10
Opening Book4moves_noob.epd
Upload PGNsFALSE
Syzygy WDLOPTIONAL
Syzygy ADJ.OPTIONAL
Win ADJ.movecount=3 score=500
Draw ADJ.movenumber=20 movecount=10 score=10
ScalingBASE 250000 NPS

experiment/l1-inference-... vs feature/threat-inputs

ApproveRestartDelete
Copy Stat Block
Elo   | -1.63 +- 2.44 (95%)
SPRT  | 10.0+0.10s Threads=1 Hash=64MB
LLR   | -2.97 (-2.94, 2.94) [-3.00, 3.00]
Games | N: 17244 W: 1994 L: 2075 D: 13175
Penta | [111, 1513, 5450, 1442, 106]
#UserUpdatedGamesWonLostDrawnTimeCrash