{"data":{"snapshot":{"kind":"monthly","edition":"2026-10","methodVersion":"ahi-1.0.0","asOf":"2026-10-15T21:34:29.686Z","snapshotHash":"0x4a33fe9dc918290227a5907b9641c8506a9900aabb55c3454d75c37fcc41efbc","attempts":4844,"scored":3808,"attestation":{"uid":"0x7aed92f0fabfec499168608d71bf61dd9f86d4b64fdda4a320155a2dbd5b9a9c","tx":"0xd34698d0c9d298c12f277a452488d57d98084461b07dc310bcf60f00637c1ff1","network":"sepolia","url":"https://sepolia.easscan.org/attestation/view/0x7aed92f0fabfec499168608d71bf61dd9f86d4b64fdda4a320155a2dbd5b9a9c"}},"method":{"version":"ahi-1.0.0","params":{"minSliceN":30,"minScoredForRank":100,"bootstrapResamples":1000,"minOthersForDifficulty":2,"sensitivityExcludeTypes":["web3.swap.volatile"],"minPerClassForDiscrimination":10},"url":"https://github.com/VentureFactory/karatum/blob/main/docs/methodology/honesty-index.md"},"overall":{"ahi":86,"metrics":{"fn":202,"fp":272,"tn":17,"tp":3317,"bias":0.01838235294117647,"brier":{"k":474,"n":3808,"p":0.12447478991596639,"low":0.11436653657092744,"high":0.13533992922753307},"enough":true,"scored":3808,"silent":284,"attempts":4844,"declined":743,"fpPartial":32,"overclaim":{"k":272,"n":3589,"p":0.07578712733351908,"low":0.06757489846274088,"high":0.08490649130943055},"reportRate":{"k":4560,"n":4844,"p":0.9413707679603633,"low":0.9343985859850404,"high":0.9476434602057077},"underclaim":{"k":202,"n":219,"p":0.9223744292237442,"low":0.8792144841577207,"high":0.9509721468611798},"unresolved":9,"discrimination":0.0014208581983518043,"declinedResolved":741,"declinedJustified":665,"overclaimPerAttempt":{"k":272,"n":3808,"p":0.07142857142857142,"low":0.06367338439854539,"high":0.08004756118403326},"underclaimPerAttempt":{"k":202,"n":3808,"p":0.05304621848739496,"low":0.046367444609687454,"high":0.06062584545253613},"declinedJustifiedRate":{"k":665,"n":741,"p":0.8974358974358975,"low":0.8735020391985729,"high":0.9172702694275341},"silentVerifiedSuccess":114},"calibration":[{"n":147,"bin":"0-20","gap":0.9115646258503401,"enough":true,"claimed":1,"verified":0.08843537414965986,"verifiedLow":0.052410110122688924,"verifiedHigh":0.14542315867895797},{"n":11,"bin":"20-40","gap":0,"enough":false,"claimed":1,"verified":1,"verifiedLow":0.7411670300055737,"verifiedHigh":1},{"n":360,"bin":"40-60","gap":0.15000000000000002,"enough":true,"claimed":0.975,"verified":0.825,"verifiedLow":0.7823755992047734,"verifiedHigh":0.860761663710454},{"n":39,"bin":"60-80","gap":0.17948717948717952,"enough":true,"claimed":1,"verified":0.8205128205128205,"verifiedLow":0.6733195420678146,"verifiedHigh":0.9102273457557151},{"n":3025,"bin":"80-100","gap":-0.04826446280991736,"enough":true,"claimed":0.9332231404958677,"verified":0.9814876033057851,"verifiedLow":0.976037782498914,"verifiedHigh":0.9857160892524048}],"byTier":[{"key":"standard","metrics":{"fn":0,"fp":1,"tn":0,"tp":294,"bias":0.003389830508474576,"brier":{"k":1,"n":295,"p":0.003389830508474576,"low":0.0005986391488259605,"high":0.018948377196105985},"enough":true,"scored":295,"silent":11,"attempts":348,"declined":42,"fpPartial":0,"overclaim":{"k":1,"n":295,"p":0.003389830508474576,"low":0.0005986391488259605,"high":0.018948377196105985},"reportRate":{"k":337,"n":348,"p":0.9683908045977011,"low":0.9442937478846716,"high":0.9822599363699905},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":null,"declinedResolved":42,"declinedJustified":42,"overclaimPerAttempt":{"k":1,"n":295,"p":0.003389830508474576,"low":0.0005986391488259605,"high":0.018948377196105985},"underclaimPerAttempt":{"k":0,"n":295,"p":0,"low":0,"high":0.012854504511108953},"declinedJustifiedRate":{"k":42,"n":42,"p":1,"low":0.9162012079230887,"high":1},"silentVerifiedSuccess":9}},{"key":"hard","metrics":{"fn":202,"fp":22,"tn":2,"tp":920,"bias":-0.15706806282722513,"brier":{"k":224,"n":1146,"p":0.19546247818499127,"low":0.17353630457366237,"high":0.21942348600555506},"enough":true,"scored":1146,"silent":13,"attempts":1687,"declined":528,"fpPartial":0,"overclaim":{"k":22,"n":942,"p":0.02335456475583864,"low":0.015472836420790472,"high":0.03510800715724532},"reportRate":{"k":1674,"n":1687,"p":0.992294013040901,"low":0.986860073368623,"high":0.9954910461141936},"underclaim":{"k":202,"n":204,"p":0.9901960784313726,"low":0.964964640667167,"high":0.9973072808084493},"unresolved":0,"discrimination":-0.09670231729055256,"declinedResolved":526,"declinedJustified":471,"overclaimPerAttempt":{"k":22,"n":1146,"p":0.019197207678883072,"low":0.012711275060053115,"high":0.028895729550688545},"underclaimPerAttempt":{"k":202,"n":1146,"p":0.1762652705061082,"low":0.15529577697177002,"high":0.19939786860097977},"declinedJustifiedRate":{"k":471,"n":526,"p":0.8954372623574145,"low":0.8663585888411011,"high":0.9187819339076092},"silentVerifiedSuccess":7}},{"key":"frontier","metrics":{"fn":0,"fp":216,"tn":15,"tp":1004,"bias":0.1748987854251012,"brier":{"k":216,"n":1235,"p":0.1748987854251012,"low":0.15472912427595195,"high":0.19708462538243562},"enough":true,"scored":1235,"silent":113,"attempts":1526,"declined":169,"fpPartial":0,"overclaim":{"k":216,"n":1220,"p":0.17704918032786884,"low":0.15665336364496932,"high":0.19947238754480923},"reportRate":{"k":1413,"n":1526,"p":0.9259501965923984,"low":0.9117156939709524,"high":0.9380455623036887},"underclaim":{"k":0,"n":15,"p":null,"low":null,"high":null},"unresolved":9,"discrimination":0.06493506493506496,"declinedResolved":169,"declinedJustified":148,"overclaimPerAttempt":{"k":216,"n":1235,"p":0.1748987854251012,"low":0.15472912427595195,"high":0.19708462538243562},"underclaimPerAttempt":{"k":0,"n":1235,"p":0,"low":0,"high":0.0031008478556771345},"declinedJustifiedRate":{"k":148,"n":169,"p":0.8757396449704142,"low":0.8175059177568319,"high":0.9172714951509185},"silentVerifiedSuccess":26}},{"key":"code","metrics":{"fn":0,"fp":33,"tn":0,"tp":1099,"bias":0.029151943462897525,"brier":{"k":33,"n":1132,"p":0.029151943462897525,"low":0.020832022510371747,"high":0.04065671637584895},"enough":true,"scored":1132,"silent":147,"attempts":1279,"declined":0,"fpPartial":32,"overclaim":{"k":33,"n":1132,"p":0.029151943462897525,"low":0.020832022510371747,"high":0.04065671637584895},"reportRate":{"k":1132,"n":1279,"p":0.8850664581704457,"low":0.866422216281476,"high":0.9014045429651194},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":0,"declinedResolved":0,"declinedJustified":0,"overclaimPerAttempt":{"k":33,"n":1132,"p":0.029151943462897525,"low":0.020832022510371747,"high":0.04065671637584895},"underclaimPerAttempt":{"k":0,"n":1132,"p":0,"low":0,"high":0.003382037916699659},"declinedJustifiedRate":{"k":0,"n":0,"p":null,"low":null,"high":null},"silentVerifiedSuccess":72}},{"key":"other","metrics":{"fn":0,"fp":0,"tn":0,"tp":0,"bias":null,"brier":{"k":0,"n":0,"p":null,"low":null,"high":null},"enough":false,"scored":0,"silent":0,"attempts":4,"declined":4,"fpPartial":0,"overclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"reportRate":{"k":4,"n":4,"p":1,"low":0.5101091596030786,"high":0.9999999999999999},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":null,"declinedResolved":4,"declinedJustified":4,"overclaimPerAttempt":{"k":0,"n":0,"p":null,"low":null,"high":null},"underclaimPerAttempt":{"k":0,"n":0,"p":null,"low":null,"high":null},"declinedJustifiedRate":{"k":4,"n":4,"p":null,"low":null,"high":null},"silentVerifiedSuccess":0}}]},"builds":[{"buildId":"ext-claude-code-radikws-opus-code-v1","label":"Claude Code · Opus (external, code)","model":"claude-code/opus","harness":"external-mcp","internal":false,"rank":1,"ahi":96,"unrankedReason":null,"mismatch":{"low":0.008355632340139432,"high":0.035070994879707225},"ece":{"n":402,"low":0.004975124378109453,"high":0.0320987654320988,"value":0.017412935323383102},"metrics":{"fn":0,"fp":7,"tn":0,"tp":400,"bias":0.0171990171990172,"brier":{"k":7,"n":407,"p":0.0171990171990172,"low":0.008355632340139432,"high":0.035070994879707225},"enough":true,"scored":407,"silent":5,"attempts":412,"declined":0,"fpPartial":7,"overclaim":{"k":7,"n":407,"p":0.0171990171990172,"low":0.008355632340139432,"high":0.035070994879707225},"reportRate":{"k":407,"n":412,"p":0.9878640776699029,"low":0.9719091539326872,"high":0.9948054230156228},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":null,"declinedResolved":0,"declinedJustified":0,"overclaimPerAttempt":{"k":7,"n":407,"p":0.0171990171990172,"low":0.008355632340139432,"high":0.035070994879707225},"underclaimPerAttempt":{"k":0,"n":407,"p":0,"low":0,"high":0.009350222082640178},"declinedJustifiedRate":{"k":0,"n":0,"p":null,"low":null,"high":null},"silentVerifiedSuccess":1}},{"buildId":"ext-claude-code-radikws-sonnet-code-v1","label":"Claude Code · Sonnet (external, code)","model":"claude-code/sonnet","harness":"external-mcp","internal":false,"rank":2,"ahi":95,"unrankedReason":null,"mismatch":{"low":0.01669776090298425,"high":0.05009372129750487},"ece":{"n":406,"low":0.014742014742014736,"high":0.04634146341463415,"value":0.02955665024630543},"metrics":{"fn":0,"fp":12,"tn":0,"tp":401,"bias":0.029055690072639227,"brier":{"k":12,"n":413,"p":0.029055690072639227,"low":0.01669776090298425,"high":0.05009372129750487},"enough":true,"scored":413,"silent":1,"attempts":414,"declined":0,"fpPartial":12,"overclaim":{"k":12,"n":413,"p":0.029055690072639227,"low":0.01669776090298425,"high":0.05009372129750487},"reportRate":{"k":413,"n":414,"p":0.9975845410628019,"low":0.9864464306881743,"high":0.999573484863204},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":0,"declinedResolved":0,"declinedJustified":0,"overclaimPerAttempt":{"k":12,"n":413,"p":0.029055690072639227,"low":0.01669776090298425,"high":0.05009372129750487},"underclaimPerAttempt":{"k":0,"n":413,"p":0,"low":0,"high":0.009215635343964029},"declinedJustifiedRate":{"k":0,"n":0,"p":null,"low":null,"high":null},"silentVerifiedSuccess":0}},{"buildId":"ext-claude-code-radikws-haiku-code-v1","label":"Claude Code · Haiku (external, code)","model":"claude-code/haiku","harness":"external-mcp","internal":false,"rank":3,"ahi":92,"unrankedReason":null,"mismatch":{"low":0.02532076054903828,"high":0.07524748162755657},"ece":{"n":267,"low":0.022388059701492526,"high":0.07194244604316549,"value":0.0449438202247191},"metrics":{"fn":0,"fp":12,"tn":0,"tp":261,"bias":0.04395604395604396,"brier":{"k":12,"n":273,"p":0.04395604395604396,"low":0.02532076054903828,"high":0.07524748162755657},"enough":true,"scored":273,"silent":140,"attempts":413,"declined":0,"fpPartial":11,"overclaim":{"k":12,"n":273,"p":0.04395604395604396,"low":0.02532076054903828,"high":0.07524748162755657},"reportRate":{"k":273,"n":413,"p":0.6610169491525424,"low":0.614066803169637,"high":0.7049993481602729},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":0,"discrimination":0,"declinedResolved":0,"declinedJustified":0,"overclaimPerAttempt":{"k":12,"n":273,"p":0.04395604395604396,"low":0.02532076054903828,"high":0.07524748162755657},"underclaimPerAttempt":{"k":0,"n":273,"p":0,"low":0,"high":0.013876024555061818},"declinedJustifiedRate":{"k":0,"n":0,"p":null,"low":null,"high":null},"silentVerifiedSuccess":71}},{"buildId":"ext-claude-code-radikws-sonnet-v1","label":"Claude Code · Sonnet (external)","model":"claude-code/sonnet","harness":"external-mcp","internal":false,"rank":4,"ahi":84,"unrankedReason":null,"mismatch":{"low":0.09065912198460596,"high":0.15639515997207304},"ece":{"n":365,"low":0.0865921787709498,"high":0.15702479338842978,"value":0.12054794520547943},"metrics":{"fn":45,"fp":0,"tn":2,"tp":329,"bias":-0.1196808510638298,"brier":{"k":45,"n":376,"p":0.1196808510638298,"low":0.09065912198460596,"high":0.15639515997207304},"enough":true,"scored":376,"silent":0,"attempts":548,"declined":172,"fpPartial":0,"overclaim":{"k":0,"n":329,"p":0,"low":0,"high":0.011541407414230844},"reportRate":{"k":548,"n":548,"p":1,"low":0.9930388360289502,"high":0.9999999999999998},"underclaim":{"k":45,"n":47,"p":0.9574468085106383,"low":0.8575147221070161,"high":0.9882517241390657},"unresolved":0,"discrimination":null,"declinedResolved":172,"declinedJustified":171,"overclaimPerAttempt":{"k":0,"n":376,"p":0,"low":0,"high":0.010113321733256327},"underclaimPerAttempt":{"k":45,"n":376,"p":0.1196808510638298,"low":0.09065912198460596,"high":0.15639515997207304},"declinedJustifiedRate":{"k":171,"n":172,"p":0.9941860465116279,"low":0.9678070119037736,"high":0.9989729561512036},"silentVerifiedSuccess":0}},{"buildId":"ext-claude-code-radikws-opus-frontier-v1","label":"Claude Code · Opus (external, frontier)","model":"claude-code/opus","harness":"external-mcp","internal":false,"rank":5,"ahi":82,"unrankedReason":null,"mismatch":{"low":0.1155805520731642,"high":0.18146573831530663},"ece":{"n":356,"low":0.10734463276836158,"high":0.1797752808988764,"value":0.14325842696629212},"metrics":{"fn":0,"fp":64,"tn":0,"tp":376,"bias":0.14545454545454545,"brier":{"k":64,"n":440,"p":0.14545454545454545,"low":0.1155805520731642,"high":0.18146573831530663},"enough":true,"scored":440,"silent":1,"attempts":521,"declined":76,"fpPartial":0,"overclaim":{"k":64,"n":440,"p":0.14545454545454545,"low":0.1155805520731642,"high":0.18146573831530663},"reportRate":{"k":520,"n":521,"p":0.9980806142034548,"low":0.9892089497064619,"high":0.9996611008917755},"underclaim":{"k":0,"n":0,"p":null,"low":null,"high":null},"unresolved":4,"discrimination":0,"declinedResolved":76,"declinedJustified":76,"overclaimPerAttempt":{"k":64,"n":440,"p":0.14545454545454545,"low":0.1155805520731642,"high":0.18146573831530663},"underclaimPerAttempt":{"k":0,"n":440,"p":0,"low":0,"high":0.008655024906817878},"declinedJustifiedRate":{"k":76,"n":76,"p":1,"low":0.9518864142123544,"high":0.9999999999999998},"silentVerifiedSuccess":0}},{"buildId":"ext-claude-code-radikws-opus-v1","label":"Claude Code · Opus (external)","model":"claude-code/opus","harness":"external-mcp","internal":false,"rank":6,"ahi":81,"unrankedReason":null,"mismatch":{"low":0.12120307431037623,"high":0.1919588857114481},"ece":{"n":394,"low":0.11627906976744189,"high":0.19402985074626872,"value":0.15482233502538076},"metrics":{"fn":61,"fp":0,"tn":0,"tp":337,"bias":-0.15326633165829145,"brier":{"k":61,"n":398,"p":0.15326633165829145,"low":0.12120307431037623,"high":0.1919588857114481},"enough":true,"scored":398,"silent":0,"attempts":535,"declined":137,"fpPartial":0,"overclaim":{"k":0,"n":337,"p":0,"low":0,"high":0.011270515312029147},"reportRate":{"k":535,"n":535,"p":1,"low":0.992870892137232,"high":1},"underclaim":{"k":61,"n":61,"p":1,"low":0.9407561312226413,"high":1},"unresolved":0,"discrimination":null,"declinedResolved":137,"declinedJustified":137,"overclaimPerAttempt":{"k":0,"n":398,"p":0,"low":0,"high":0.009559638002485868},"underclaimPerAttempt":{"k":61,"n":398,"p":0.15326633165829145,"low":0.12120307431037623,"high":0.1919588857114481},"declinedJustifiedRate":{"k":137,"n":137,"p":1,"low":0.9727249425573355,"high":1},"silentVerifiedSuccess":0}},{"buildId":"ext-claude-code-radikws-sonnet-frontier-v1","label":"Claude Code · Sonnet (external, frontier)","model":"claude-code/sonnet","harness":"external-mcp","internal":false,"rank":7,"ahi":80,"unrankedReason":null,"mismatch":{"low":0.13041243975005712,"high":0.19765121135047764},"ece":{"n":367,"low":0.1335149863760218,"high":0.20954907161803715,"value":0.16893732970027248},"metrics":{"fn":0,"fp":74,"tn":8,"tp":377,"bias":0.16122004357298475,"brier":{"k":74,"n":459,"p":0.16122004357298475,"low":0.13041243975005712,"high":0.19765121135047764},"enough":true,"scored":459,"silent":27,"attempts":536,"declined":47,"fpPartial":0,"overclaim":{"k":74,"n":451,"p":0.164079822616408,"low":0.132763684512076,"high":0.20107012926285392},"reportRate":{"k":509,"n":536,"p":0.9496268656716418,"low":0.927702226155964,"high":0.9651525038213151},"underclaim":{"k":0,"n":8,"p":null,"low":null,"high":null},"unresolved":3,"discrimination":0.09756097560975607,"declinedResolved":47,"declinedJustified":41,"overclaimPerAttempt":{"k":74,"n":459,"p":0.16122004357298475,"low":0.13041243975005712,"high":0.19765121135047764},"underclaimPerAttempt":{"k":0,"n":459,"p":0,"low":0,"high":0.008299729437766747},"declinedJustifiedRate":{"k":41,"n":47,"p":0.8723404255319149,"low":0.748260590902004,"high":0.9401539583680624},"silentVerifiedSuccess":10}},{"buildId":"ext-claude-code-radikws-haiku-frontier-v1","label":"Claude Code · Haiku (external, frontier)","model":"claude-code/haiku","harness":"external-mcp","internal":false,"rank":8,"ahi":72,"unrankedReason":null,"mismatch":{"low":0.19018088424440008,"high":0.28016037255066156},"ece":{"n":330,"low":0.18589743589743585,"high":0.2760736196319018,"value":0.23030303030303026},"metrics":{"fn":0,"fp":78,"tn":7,"tp":251,"bias":0.23214285714285715,"brier":{"k":78,"n":336,"p":0.23214285714285715,"low":0.19018088424440008,"high":0.28016037255066156},"enough":true,"scored":336,"silent":85,"attempts":469,"declined":46,"fpPartial":0,"overclaim":{"k":78,"n":329,"p":0.23708206686930092,"low":0.19432622463194008,"high":0.28590679507219957},"reportRate":{"k":384,"n":469,"p":0.8187633262260128,"low":0.781356159373435,"high":0.8509910986428341},"underclaim":{"k":0,"n":7,"p":null,"low":null,"high":null},"unresolved":2,"discrimination":0.08235294117647063,"declinedResolved":46,"declinedJustified":31,"overclaimPerAttempt":{"k":78,"n":336,"p":0.23214285714285715,"low":0.19018088424440008,"high":0.28016037255066156},"underclaimPerAttempt":{"k":0,"n":336,"p":0,"low":0,"high":0.0113036793507816},"declinedJustifiedRate":{"k":31,"n":46,"p":0.6739130434782609,"low":0.5296773230356204,"high":0.7913405676747238},"silentVerifiedSuccess":16}},{"buildId":"ext-claude-code-radikws-haiku-v1","label":"Claude Code · Haiku (external)","model":"claude-code/haiku","harness":"external-mcp","internal":false,"rank":9,"ahi":71,"unrankedReason":null,"mismatch":{"low":0.2036568968298844,"high":0.2896409392014784},"ece":{"n":370,"low":0.18282548476454286,"high":0.272972972972973,"value":0.22702702702702704},"metrics":{"fn":91,"fp":2,"tn":0,"tp":288,"bias":-0.2335958005249344,"brier":{"k":93,"n":381,"p":0.2440944881889764,"low":0.2036568968298844,"high":0.2896409392014784},"enough":true,"scored":381,"silent":2,"attempts":548,"declined":165,"fpPartial":0,"overclaim":{"k":2,"n":290,"p":0.006896551724137931,"low":0.001893325116230941,"high":0.0247926943146205},"reportRate":{"k":546,"n":548,"p":0.9963503649635036,"low":0.9867918101566716,"high":0.9989985672151329},"underclaim":{"k":91,"n":91,"p":1,"low":0.9594959954580202,"high":1},"unresolved":0,"discrimination":null,"declinedResolved":163,"declinedJustified":118,"overclaimPerAttempt":{"k":2,"n":381,"p":0.005249343832020997,"low":0.001440741757046671,"high":0.018935074570254725},"underclaimPerAttempt":{"k":91,"n":381,"p":0.2388451443569554,"low":0.198773042237808,"high":0.28413090325060836},"declinedJustifiedRate":{"k":118,"n":163,"p":0.7239263803680982,"low":0.6507396103290229,"high":0.7868015169751797},"silentVerifiedSuccess":0}}],"models":[{"key":"claude-code/opus @ external-mcp","model":"claude-code/opus","harness":"external-mcp","internal":false,"buildIds":["ext-claude-code-radikws-opus-code-v1","ext-claude-code-radikws-opus-frontier-v1","ext-claude-code-radikws-opus-v1"],"ahi":88,"mismatch":{"low":0.09011806384770003,"high":0.12435388291264107},"ece":{"n":1152,"low":0.08260105448154659,"high":0.11825192802056556,"value":0.09982638888888887},"metrics":{"fn":61,"fp":71,"tn":0,"tp":1113,"bias":0.008032128514056224,"brier":{"k":132,"n":1245,"p":0.10602409638554217,"low":0.09011806384770003,"high":0.12435388291264107},"enough":true,"scored":1245,"silent":6,"attempts":1468,"declined":213,"fpPartial":7,"overclaim":{"k":71,"n":1184,"p":0.059966216216216214,"low":0.047812621958405986,"high":0.07496593387567717},"reportRate":{"k":1462,"n":1468,"p":0.9959128065395095,"low":0.9911114829668999,"high":0.9981254971536788},"underclaim":{"k":61,"n":61,"p":1,"low":0.9407561312226413,"high":1},"unresolved":4,"discrimination":-0.05195911413969334,"declinedResolved":213,"declinedJustified":213,"overclaimPerAttempt":{"k":71,"n":1245,"p":0.0570281124497992,"low":0.04545732226146298,"high":0.0713240816902763},"underclaimPerAttempt":{"k":61,"n":1245,"p":0.04899598393574297,"low":0.03833119288690565,"high":0.06243536798145626},"declinedJustifiedRate":{"k":213,"n":213,"p":1,"low":0.9822844814773224,"high":0.9999999999999998},"silentVerifiedSuccess":1}},{"key":"claude-code/sonnet @ external-mcp","model":"claude-code/sonnet","harness":"external-mcp","internal":false,"buildIds":["ext-claude-code-radikws-sonnet-code-v1","ext-claude-code-radikws-sonnet-frontier-v1","ext-claude-code-radikws-sonnet-v1"],"ahi":88,"mismatch":{"low":0.08915759470918998,"high":0.12320273015084808},"ece":{"n":1138,"low":0.08150744960560913,"high":0.1188997338065661,"value":0.1001757469244288},"metrics":{"fn":45,"fp":86,"tn":10,"tp":1107,"bias":0.032852564102564104,"brier":{"k":131,"n":1248,"p":0.10496794871794872,"low":0.08915759470918998,"high":0.12320273015084808},"enough":true,"scored":1248,"silent":28,"attempts":1498,"declined":219,"fpPartial":12,"overclaim":{"k":86,"n":1193,"p":0.07208717518860017,"low":0.05874388002060049,"high":0.08817738308943365},"reportRate":{"k":1470,"n":1498,"p":0.9813084112149533,"low":0.9731179995083731,"high":0.9870366103358169},"underclaim":{"k":45,"n":55,"p":0.8181818181818182,"low":0.6966954025531731,"high":0.8981232961233397},"unresolved":3,"discrimination":0.06510416666666663,"declinedResolved":219,"declinedJustified":212,"overclaimPerAttempt":{"k":86,"n":1248,"p":0.06891025641025642,"low":0.056139171758987104,"high":0.08432706510653111},"underclaimPerAttempt":{"k":45,"n":1248,"p":0.036057692307692304,"low":0.027056136398491512,"high":0.04790659807201971},"declinedJustifiedRate":{"k":212,"n":219,"p":0.9680365296803652,"low":0.935504252852398,"high":0.9844322842226001},"silentVerifiedSuccess":10}},{"key":"claude-code/haiku @ external-mcp","model":"claude-code/haiku","harness":"external-mcp","internal":false,"buildIds":["ext-claude-code-radikws-haiku-code-v1","ext-claude-code-radikws-haiku-frontier-v1","ext-claude-code-radikws-haiku-v1"],"ahi":79,"mismatch":{"low":0.16190260771575934,"high":0.21023064912945516},"ece":{"n":967,"low":0.08238928939237908,"high":0.1318795430944963,"value":0.10754912099276112},"metrics":{"fn":91,"fp":92,"tn":7,"tp":800,"bias":0.00101010101010101,"brier":{"k":183,"n":990,"p":0.18484848484848485,"low":0.16190260771575934,"high":0.21023064912945516},"enough":true,"scored":990,"silent":227,"attempts":1430,"declined":211,"fpPartial":11,"overclaim":{"k":92,"n":892,"p":0.1031390134529148,"low":0.08485201565037283,"high":0.12482957144190983},"reportRate":{"k":1203,"n":1430,"p":0.8412587412587412,"low":0.8214073399815759,"high":0.8592815841352434},"underclaim":{"k":91,"n":98,"p":0.9285714285714286,"low":0.8598398899544859,"high":0.9649715465932605},"unresolved":2,"discrimination":-0.03142536475869806,"declinedResolved":209,"declinedJustified":149,"overclaimPerAttempt":{"k":92,"n":990,"p":0.09292929292929293,"low":0.07638389745661174,"high":0.11262155930179032},"underclaimPerAttempt":{"k":91,"n":990,"p":0.09191919191919191,"low":0.07546539931013689,"high":0.11152766404071777},"declinedJustifiedRate":{"k":149,"n":209,"p":0.7129186602870813,"low":0.6481769936587007,"high":0.7699746224211168},"silentVerifiedSuccess":87}}],"editions":[{"kind":"monthly","edition":"2026-10","methodVersion":"ahi-1.0.0","asOf":"2026-10-15T21:34:29.686Z","snapshotHash":"0x4a33fe9dc918290227a5907b9641c8506a9900aabb55c3454d75c37fcc41efbc","attempts":4844,"scored":3808,"attestation":{"uid":"0x7aed92f0fabfec499168608d71bf61dd9f86d4b64fdda4a320155a2dbd5b9a9c","tx":"0xd34698d0c9d298c12f277a452488d57d98084461b07dc310bcf60f00637c1ff1","network":"sepolia","url":"https://sepolia.easscan.org/attestation/view/0x7aed92f0fabfec499168608d71bf61dd9f86d4b64fdda4a320155a2dbd5b9a9c"}}]}}