List your company

Large Language Model Evaluation on HuggingFace Open LLM Leaderboard (Full Suite)

55.37GSM8KSynPO
31.949238.029644.1150.1904Jun 1, 2025Jul 2, 2025Aug 3, 2025Sep 4, 2025Oct 5, 2025Nov 6, 2025Dec 8, 2025

Evaluation results

MethodLinks
SynPOBackbone=LLama3-8B, Variant=Instruct2025.0655.37—83.19—54.3579.7136.0631.98—42.0855.4260.61
SynPOBackbone=Mistral-7B, Variant=Instruct2025.0655.11—83.76—54.5179.9235.8432.87—40.0855.3260.43
SynPOBackbone=Mistral-7B, Variant=Base2025.0654.95—81.26—50.5877.1430.8328.99—37.8452.155.24
DPOBackbone=LLama3-8B, Variant=Instruct2025.0654.83—82.85—52.8178.2234.5431.29—41.3254.4559.76
SynPOBackbone=LLama3-8B, Variant=Base2025.0654.81—80.88—53.0474.5531.4229.03—38.0352.1655.54
SimPOBackbone=LLama3-8B, Variant=Instruct2025.0654.72—82.29—53.1677.1535.0530.95—40.0954.3361.24
KTOBackbone=Mistral-7B, Variant=Base2025.0654.51—81.75—48.3477.1727.0327.66—35.5150.7954.37
DPOPBackbone=Mistral-7B, Variant=Base2025.0654.32—80.49—49.0776.2330.7526.24—36.1250.9954.73
SimPOBackbone=Mistral-7B, Variant=Instruct2025.0654.25—82.04—52.1978.7129.5232.7—39.153.5259.69
IPOBackbone=LLama3-8B, Variant=Instruct2025.0654.23—81.95—51.0278.5833.2730.31—40.9753.7859.95
DPOBackbone=Mistral-7B, Variant=Instruct2025.0654.22—83.22—51.1578.4633.7229.34—39.5353.7160.04
IPOBackbone=Mistral-7B, Variant=Base2025.0654.14—79.15—47.2574.1525.5225.59—34.8749.3153.84
KTOBackbone=LLama3-8B, Variant=Instruct2025.0654.01—82.7—53.6377.134.1231.15—41.754.3760.57
DPOPBackbone=Mistral-7B, Variant=Instruct2025.0653.65—82.04—52.9580.1334.5329.24—39.6954.0259.9
DPOPBackbone=LLama3-8B, Variant=Instruct2025.0653.57—82.52—52.2979.136.5130.55—41.8954.4659.26
DPOBackbone=Mistral-7B, Variant=Base2025.0653.47—81.4—48.7276.6629.1826.85—35.6850.7954.32
DPOBackbone=LLama3-8B, Variant=Base2025.0653.45—80.02—49.1575.4630.9727.34—36.5350.9454.61
SimPOBackbone=LLama3-8B, Variant=Base2025.0653.45—78.13—49.1573.4627.7326.88—36.1349.6952.61
CPOBackbone=LLama3-8B, Variant=Instruct2025.0653.4—81.08—52.7576.8135.0830.51—39.5653.4458.29
KTOBackbone=Mistral-7B, Variant=Instruct2025.0653.35—80.34—52.7778.1934.0230.62—40.4653.6959.8
SimPOBackbone=Mistral-7B, Variant=Base2025.0653.21—81.03—49.4976.6829.6326.94—35.1350.7253.63
CPOBackbone=Mistral-7B, Variant=Base2025.0653.06—80.45—49.1575.1526.3227.05—34.0449.9754.53
KTOBackbone=LLama3-8B, Variant=Base2025.0652.45—78.53—51.6775.1228.8827.09—35.8950.3152.84
IPOBackbone=Mistral-7B, Variant=Instruct2025.0652.42—81.61—52.0279.5531.8530.21—38.7553.0958.31
DPOPBackbone=LLama3-8B, Variant=Base2025.0652.35—79.62—50.1274.5329.3826.96—36.9250.3853.2
CPOBackbone=Mistral-7B, Variant=Instruct2025.0652.28—80.01—50.2879.1527.8132.66—38.8552.4758.74
IPOBackbone=LLama3-8B, Variant=Base2025.0651.76—79.48—48.0374.8725.7627.01—35.4749.2851.83
CPOBackbone=LLama3-8B, Variant=Base2025.0651.52—79.42—49.0473.3226.8628.14—36.1849.8254.11
SPACEIteration=Iter4, Backbone=Mistral-7B2025.1246.0265.8784.4458.551.8674.5135.945.2830.531.4152.43—
SPACEIteration=Iter3, Backbone=Mistral-7B2025.1245.8465.9684.3958.5351.3474.1933.1945.3829.931.4452.03—
SPACEIteration=Iter2, Backbone=Mistral-7B2025.1245.5565.7884.2558.5750.7173.9533.0145.3829.5231.5951.83—
SPACEIteration=Iter1, Backbone=Mistral-7B2025.1245.4164.8583.8658.8748.9975.1431.5145.7329.4630.9451.48—
S-SimPOIteration=Iter1, Backbone=Mistral-7B2025.1241.5161.6983.7757.4848.9874.923.0141.8629.930.2249.33—
SPACEIteration=Iter0, Backbone=Mistral-7B2025.1241.3262.7183.7958.6747.1574.926.9645.8729.6630.7350.18—
S-IPOIteration=Iter3, Backbone=Mistral-7B2025.1241.0560.6783.4257.0344.3574.272341.7329.3229.6948.45—
S-IPOIteration=Iter0, Backbone=Mistral-7B2025.1240.9660.4983.5857.1745.4573.8820.8542.3330.7130.0148.54—
SPINIteration=Iter2, Backbone=Mistral-7B2025.1240.2662.283.658.0346.9274.4325.0643.528.9530.3749.33—
S-IPOIteration=Iter4, Backbone=Mistral-7B2025.1240.1161.4383.5656.6145.5674.1123.1141.0529.3129.3648.42—
S-SimPOIteration=Iter0, Backbone=Mistral-7B2025.1239.6761.3583.7657.2947.4474.6624.8841.8630.130.0449.11—
S-IPOIteration=Iter2, Backbone=Mistral-7B2025.1239.3560.0783.1957.2943.3573.823.2941.6930.529.8648.24—
S-IPOIteration=Iter1, Backbone=Mistral-7B2025.1238.9160.0783.4757.2943.6273.821.9842.0830.1929.948.13—
SPINIteration=Iter0, Backbone=Mistral-7B2025.1238.2961.1883.4957.8843.7374.1122.0944.629.6230.1548.51—
Mistral-7BBackbone=Mistral-7B-Base2025.1237.6861.0183.2457.8642.6274.0323.6344.2629.8629.9948.42—
S-SimPOIteration=Iter2, Backbone=Mistral-7B2025.1237.0961.5283.7757.1649.5874.5122.4443.0429.1729.9748.83—
S-SimPOIteration=Iter3, Backbone=Mistral-7B2025.1235.2261.2683.757.3549.8874.8221.1542.829.0430.0148.52—
SPINIteration=Iter3, Backbone=Mistral-7B2025.1234.8762.2984.0158.0945.7275.1424.6443.0328.6829.6248.61—
SPINIteration=Iter4, Backbone=Mistral-7B2025.1234.761.8683.9957.994675.1425.7343.3427.129.8248.57—
S-SimPOIteration=Iter4, Backbone=Mistral-7B2025.1234.3861.3583.6857.4850.0274.1921.3142.6729.1639.449.36—
SPINIteration=Iter1, Backbone=Mistral-7B2025.1232.8561.5284.0257.3247.6973.9521.9240.5928.2230.1247.82—