mirror of
https://github.com/hanzoai/benchmarks.git
synced 2026-10-30 14:59:11 +00:00
1 line
12 KiB
JSON
1 line
12 KiB
JSON
{"meta":{"run 2026-09-27 10:08:23 +0000":{"host":"arm64-mac","machine":"aarch64","runtime":"bench (hanzo-ml)","load":[10.5673828125,13.26708984375,16.92626953125],"at":"2026-09-27 10:08:23 +0000","laya":{"runtime":"bench (hanzo-ml, kai::joint)","checkpoints":[{"id":"laya","repo":"hanzoai/kai-1","revision":"b50502c28537df49a3621f6fa543f9e8521e8a9c"},{"id":"laya-multilingual","repo":"hanzoai/kai-1-multilingual","revision":"3119843b5261e34903f93d03813467524c57982e"},{"id":"laya-agent","repo":"hanzoai/kai-1-agent","revision":"e3e416499b8a983d2aa0bcc684d5457ad6bfd6a1"}],"device":"metal","dtype":"f32"},"jev":{"model":"typesafe/jev-1.13","url":"https://openrouter.ai/api/alpha/decisions","price_per_input_token":4.2e-08},"cases":{"n":400,"families":["maintenance","trade"],"seed":13}},"run 2026-09-27 11:41:31 +0000":{"host":"arm64-mac","machine":"aarch64","runtime":"bench (hanzo-ml)","load":[8.63427734375,8.56640625,11.419921875],"at":"2026-09-27 11:41:31 +0000","pending":{},"cases":{"n":400,"families":["maintenance","trade"],"seed":13}},"run 2026-09-27 14:41:30 +0000":{"host":"arm64-mac","machine":"aarch64","runtime":"bench (hanzo-ml)","load":[7.23828125,8.83447265625,10.16259765625],"at":"2026-09-27 14:41:30 +0000","pending":{},"cases":{"n":400,"families":["maintenance","trade"],"seed":13}},"run 2026-09-28 03:23:08 +0000":{"host":"arm64-host","machine":"aarch64","runtime":"bench (hanzo-ml)","load":[2.4453125,2.50439453125,2.5205078125],"at":"2026-09-28 03:23:08 +0000","pending":{},"cases":{"n":400,"families":["maintenance","trade"],"seed":13}}},"keys":{"cap/joint/jev/acc":0.837,"cap/joint/jev/acc.action":0.78,"cap/joint/jev/acc.evidence":0.915,"cap/joint/jev/acc.fault":1.0,"cap/joint/jev/acc.option":0.965,"cap/joint/jev/acc.readiness":0.825,"cap/joint/jev/acc.review":0.6,"cap/joint/jev/acc.risk":0.93,"cap/joint/jev/acc.severity":0.93,"cap/joint/jev/exact":0.4375,"cap/joint/jev/maintenance.acc":0.826,"cap/joint/jev/maintenance.exact":0.435,"cap/joint/jev/maintenance.proj.acc":0.889,"cap/joint/jev/maintenance.proj.exact":0.545,"cap/joint/jev/maintenance.violate":0.29,"cap/joint/jev/proj.acc":0.8795,"cap/joint/jev/proj.acc.action":0.96,"cap/joint/jev/proj.acc.evidence":0.935,"cap/joint/jev/proj.acc.fault":1.0,"cap/joint/jev/proj.acc.option":0.975,"cap/joint/jev/proj.acc.readiness":0.92,"cap/joint/jev/proj.acc.review":0.6125,"cap/joint/jev/proj.acc.risk":0.9,"cap/joint/jev/proj.acc.severity":0.96,"cap/joint/jev/proj.exact":0.56,"cap/joint/jev/proj.violate":0.0,"cap/joint/jev/trade.acc":0.848,"cap/joint/jev/trade.exact":0.44,"cap/joint/jev/trade.proj.acc":0.87,"cap/joint/jev/trade.proj.exact":0.575,"cap/joint/jev/trade.violate":0.205,"cap/joint/jev/unanswered":0,"cap/joint/jev/violate":0.2475,"cap/joint/laya/acc":0.336,"cap/joint/laya/acc.action":0.28,"cap/joint/laya/acc.evidence":0.3,"cap/joint/laya/acc.fault":0.325,"cap/joint/laya/acc.option":0.315,"cap/joint/laya/acc.readiness":0.3275,"cap/joint/laya/acc.review":0.4825,"cap/joint/laya/acc.risk":0.26,"cap/joint/laya/acc.severity":0.26,"cap/joint/laya/en.acc":0.32,"cap/joint/laya/en.acc.action":0.1,"cap/joint/laya/en.acc.evidence":0.29,"cap/joint/laya/en.acc.fault":0.3,"cap/joint/laya/en.acc.option":0.275,"cap/joint/laya/en.acc.readiness":0.3525,"cap/joint/laya/en.acc.review":0.4825,"cap/joint/laya/en.acc.risk":0.28,"cap/joint/laya/en.acc.severity":0.285,"cap/joint/laya/en.exact":0.0,"cap/joint/laya/en.maintenance.acc":0.305,"cap/joint/laya/en.maintenance.exact":0.0,"cap/joint/laya/en.maintenance.proj.acc":0.354,"cap/joint/laya/en.maintenance.proj.exact":0.005,"cap/joint/laya/en.maintenance.violate":0.91,"cap/joint/laya/en.proj.acc":0.3445,"cap/joint/laya/en.proj.acc.action":0.32,"cap/joint/laya/en.proj.acc.evidence":0.29,"cap/joint/laya/en.proj.acc.fault":0.275,"cap/joint/laya/en.proj.acc.option":0.275,"cap/joint/laya/en.proj.acc.readiness":0.36,"cap/joint/laya/en.proj.acc.review":0.4825,"cap/joint/laya/en.proj.acc.risk":0.28,"cap/joint/laya/en.proj.acc.severity":0.32,"cap/joint/laya/en.proj.exact":0.0025,"cap/joint/laya/en.proj.violate":0.0,"cap/joint/laya/en.trade.acc":0.335,"cap/joint/laya/en.trade.exact":0.0,"cap/joint/laya/en.trade.proj.acc":0.335,"cap/joint/laya/en.trade.proj.exact":0.0,"cap/joint/laya/en.trade.violate":0.03,"cap/joint/laya/en.unanswered":0,"cap/joint/laya/en.violate":0.47,"cap/joint/laya/exact":0.005,"cap/joint/laya/maintenance.acc":0.329,"cap/joint/laya/maintenance.exact":0.0,"cap/joint/laya/maintenance.proj.acc":0.302,"cap/joint/laya/maintenance.proj.exact":0.07,"cap/joint/laya/maintenance.violate":0.96,"cap/joint/laya/proj.acc":0.3275,"cap/joint/laya/proj.acc.action":0.22,"cap/joint/laya/proj.acc.evidence":0.3,"cap/joint/laya/proj.acc.fault":0.3,"cap/joint/laya/proj.acc.option":0.315,"cap/joint/laya/proj.acc.readiness":0.335,"cap/joint/laya/proj.acc.review":0.4825,"cap/joint/laya/proj.acc.risk":0.285,"cap/joint/laya/proj.acc.severity":0.22,"cap/joint/laya/proj.exact":0.04,"cap/joint/laya/proj.violate":0.0,"cap/joint/laya/trade.acc":0.343,"cap/joint/laya/trade.exact":0.01,"cap/joint/laya/trade.proj.acc":0.353,"cap/joint/laya/trade.proj.exact":0.01,"cap/joint/laya/trade.violate":0.315,"cap/joint/laya/unanswered":0,"cap/joint/laya/violate":0.6375},"detail":{"laya":{"metrics":{"acc":0.336,"acc.fault":0.325,"acc.severity":0.26,"acc.readiness":0.3275,"acc.action":0.28,"acc.review":0.4825,"acc.option":0.315,"acc.evidence":0.3,"acc.risk":0.26,"exact":0.005,"violate":0.6375,"unanswered":0,"proj.acc":0.3275,"proj.acc.fault":0.3,"proj.acc.severity":0.22,"proj.acc.readiness":0.335,"proj.acc.action":0.22,"proj.acc.review":0.4825,"proj.acc.option":0.315,"proj.acc.evidence":0.3,"proj.acc.risk":0.285,"proj.exact":0.04,"proj.violate":0.0,"maintenance.acc":0.329,"maintenance.exact":0.0,"maintenance.violate":0.96,"maintenance.proj.acc":0.302,"maintenance.proj.exact":0.07,"trade.acc":0.343,"trade.exact":0.01,"trade.violate":0.315,"trade.proj.acc":0.353,"trade.proj.exact":0.01},"seconds":471.365,"dropped":0},"en.laya":{"metrics":{"acc":0.32,"acc.fault":0.3,"acc.severity":0.285,"acc.readiness":0.3525,"acc.action":0.1,"acc.review":0.4825,"acc.option":0.275,"acc.evidence":0.29,"acc.risk":0.28,"exact":0.0,"violate":0.47,"unanswered":0,"proj.acc":0.3445,"proj.acc.fault":0.275,"proj.acc.severity":0.32,"proj.acc.readiness":0.36,"proj.acc.action":0.32,"proj.acc.review":0.4825,"proj.acc.option":0.275,"proj.acc.evidence":0.29,"proj.acc.risk":0.28,"proj.exact":0.0025,"proj.violate":0.0,"maintenance.acc":0.305,"maintenance.exact":0.0,"maintenance.violate":0.91,"maintenance.proj.acc":0.354,"maintenance.proj.exact":0.005,"trade.acc":0.335,"trade.exact":0.0,"trade.violate":0.03,"trade.proj.acc":0.335,"trade.proj.exact":0.0},"seconds":272.337,"dropped":0},"jev":{"metrics":{"acc":0.837,"acc.fault":1.0,"acc.severity":0.93,"acc.readiness":0.825,"acc.action":0.78,"acc.review":0.6,"acc.option":0.965,"acc.evidence":0.915,"acc.risk":0.93,"exact":0.4375,"violate":0.2475,"unanswered":0,"proj.acc":0.8795,"proj.acc.fault":1.0,"proj.acc.severity":0.96,"proj.acc.readiness":0.92,"proj.acc.action":0.96,"proj.acc.review":0.6125,"proj.acc.option":0.975,"proj.acc.evidence":0.935,"proj.acc.risk":0.9,"proj.exact":0.56,"proj.violate":0.0,"maintenance.acc":0.826,"maintenance.exact":0.435,"maintenance.violate":0.29,"maintenance.proj.acc":0.889,"maintenance.proj.exact":0.545,"trade.acc":0.848,"trade.exact":0.44,"trade.violate":0.205,"trade.proj.acc":0.87,"trade.proj.exact":0.575},"n_errors":0,"error_kinds":{},"cost_usd":0.017648,"latency_ms":[101.223292,101.891041,102.667542,103.474833,103.70500000000001,103.82395799999999,104.629833,104.74275,104.77591699999999,105.215875,105.623042,105.73579199999999,105.799083,106.2515,106.431625,106.504875,106.533291,106.679834,106.79958400000001,106.812791,106.94366600000001,107.09837499999999,107.238417,107.278167,107.386584,107.854375,108.028083,108.22375,108.563542,108.58283300000001,108.72325,108.922958,109.63225,109.769666,109.844125,110.030083,110.13675,110.336833,110.509875,110.6695,110.81125,110.956125,111.09929100000001,111.229042,111.229833,111.386125,111.464167,111.52737499999999,111.796375,111.885291,112.100542,112.322,112.621792,112.64699999999999,112.790875,112.797166,112.897917,113.051125,113.14616600000001,113.308,113.436083,113.493084,113.650167,114.036125,114.336542,114.57458299999999,114.70916700000001,114.83125,114.852333,115.00862500000001,115.105834,115.43029200000001,115.823542,115.83425,115.91125,116.34025000000001,116.42554200000001,116.488667,116.979625,117.244083,117.310416,117.62379100000001,117.959209,118.07270799999999,118.208542,118.37995799999999,118.400666,118.404042,118.41095899999999,118.54974999999999,118.63395899999999,118.73204199999999,118.984208,119.2555,119.606459,119.610334,119.914917,120.14066700000001,120.151834,120.535375,120.67504199999999,121.08429199999999,121.172667,121.45975,121.572333,121.62487499999999,121.871417,121.88520899999999,121.893875,122.016208,122.04433300000001,122.071958,122.121167,122.144917,122.230333,122.511666,122.70325,123.393709,123.486417,123.49291699999999,123.5285,123.568792,123.576458,123.747167,123.81,123.89595800000001,123.906958,124.181083,124.323667,124.37087500000001,125.21325,125.33787499999998,125.58233299999999,125.718292,126.32358399999998,126.82950000000001,127.09962499999999,127.432459,127.89437500000001,127.94112499999999,127.95925,128.310291,128.811542,129.13945800000002,129.334583,129.370167,129.549292,129.71170800000002,129.839833,129.89075,129.92570800000001,130.10866700000003,130.64283400000002,130.741333,130.805083,131.046625,131.117833,131.158875,131.49670899999998,131.555334,131.72500000000002,131.988,132.078042,132.144375,132.17175,132.389458,132.400791,132.410416,132.76637499999998,132.877417,132.928917,132.99325000000002,133.24425000000002,133.29670900000002,133.30529199999998,133.354,133.38975,133.470959,133.655458,133.754458,133.9855,134.026375,134.092666,134.177,134.390083,134.70825,135.18533299999999,135.29312499999997,135.382208,135.61024999999998,135.64816599999997,135.912084,136.296417,136.403458,136.697125,136.845792,137.217583,137.503208,137.591375,137.595959,137.828042,138.04262500000002,138.273584,138.36041600000001,138.499958,138.578333,139.01625,139.05825,139.147291,139.30958399999997,139.448875,139.673667,139.920708,140.37941700000002,140.795208,140.83950000000002,140.842625,140.887875,140.95125,141.098375,141.344791,141.45174999999998,143.13620799999998,143.746083,143.951042,144.16799999999998,144.541625,144.56912499999999,144.716708,144.98695899999998,145.126917,145.520667,145.698667,145.806583,146.10620799999998,146.128791,146.176209,146.22037500000002,146.291833,146.31295799999998,146.549959,146.79608299999998,146.887541,147.054542,147.1045,147.166917,147.433375,147.458,147.487042,147.522333,147.798875,147.951417,148.311459,148.406041,148.64350000000002,148.786125,149.177292,149.275291,149.47475,149.77083399999998,149.874292,150.176584,150.334459,150.367375,150.872583,150.888708,151.054375,151.097375,151.719209,152.150583,152.321083,152.4435,152.58925000000002,152.61179099999998,153.185208,153.307042,153.353833,153.446834,154.138417,154.195167,154.565333,156.301167,156.567291,157.041667,157.16329100000002,157.290875,157.6975,157.78129099999998,158.38633299999998,158.55433399999998,158.705,159.12420899999998,159.855042,159.996375,160.328292,160.36375,160.580083,160.713625,161.09495900000002,161.299083,161.311458,161.46829200000002,161.56737500000003,161.588667,161.598167,161.612792,161.7875,161.8475,161.91387500000002,162.848125,163.86012499999998,163.873292,164.33670899999998,164.413208,165.17891600000002,166.305708,166.308833,166.424791,166.8615,167.08075,167.80283300000002,168.567667,168.712375,169.040458,170.36641699999998,170.84329100000002,172.728375,174.187542,174.189042,174.809417,175.548,175.658791,176.20974999999999,176.45,177.167416,177.427417,178.178375,178.41179200000002,178.46575,178.941958,179.55170900000002,181.343792,181.669916,182.387791,183.204959,183.583542,183.74325,185.286958,185.484958,187.19945900000002,187.366583,187.476792,187.852209,190.098625,190.40975,191.17362500000002,191.99887500000003,195.015041,195.30895800000002,195.563042,196.141917,196.944542,197.75458400000002,199.215,203.318208,209.71725,210.99625,211.45708399999998,211.48470799999998,214.116708,214.7695,215.716542,219.212417,219.43687500000001,220.96687500000002,223.7305,225.359125,226.781375,230.827583,231.030958,235.27958299999997,235.48525,236.56845800000002,236.57225000000003,247.901791,250.37504199999998,259.88025,271.415083,275.457375,286.342791,316.9625,365.003417,375.417,379.581167,394.773958,395.755,398.90700000000004,420.493833,426.577333,430.880625]}}} |