List your company

General Capability on Aggregate (GPQA-D, GSM8K, HumanEval, MATH-500, MBPP, MMLU-Pro)

75.9Average AccuracyFOREVER
35.8646.25556.6567.045May 10, 2026

Evaluation results

MethodLinks
FOREVERModel Scale=14B, Training Protocol=Capability-continual2026.0575.9——
FOREVERModel Scale=14B2026.0575.9——
FOREVERModel Scale=8B, Training Protocol=Capability-continual2026.0575.6——
MTLModel Scale=14B, Training Protocol=Joint-training2026.0574.6——
MTLModel Scale=14B2026.0574.6——
GCWM-DiamondScale=14B2026.0574.321.390.8333
GCWMModel Scale=14B, Training Protocol=Capability-continual2026.0574.3——
GCWMModel Scale=14B2026.0574.3——
EWCModel Scale=14B, Training Protocol=Capability-continual2026.0573.5——
EWCModel Scale=14B2026.0573.5——
MTLModel Scale=8B, Training Protocol=Joint-training2026.0573.3——
OPCMScale=14B2026.0572.93——
OPCMModel Scale=14B, Training Protocol=Capability-continual2026.0572.9——
OPCMModel Scale=14B2026.0572.9——
EWCModel Scale=8B, Training Protocol=Capability-continual2026.0572.6——
GCWM-DiamondScale=8B2026.0572.541.610.5
FOREVERModel Scale=4B, Training Protocol=Capability-continual2026.0572.5——
GCWMModel Scale=8B, Training Protocol=Capability-continual2026.0572.5——
AIMMergingModel Scale=14B, Training Protocol=Capability-continual2026.0572.2——
AIMMergingModel Scale=14B2026.0572.2——
Seq. SFTModel Scale=4B, Training Protocol=Capability-continual2026.0571.6——
L&SModel Scale=14B, Training Protocol=Capability-continual2026.0571.3——
L&SModel Scale=14B2026.0571.3——
OPCMScale=8B2026.0570.93——
OPCMModel Scale=8B, Training Protocol=Capability-continual2026.0570.9——
OPCMModel Scale=4B, Training Protocol=Capability-continual2026.0570.5——
OPCMScale=4B2026.0570.45——
Seq. SFTModel Scale=14B, Training Protocol=Capability-continual2026.0570.4——
Seq. SFTModel Scale=14B2026.0570.4——
GCWMModel Scale=4B, Training Protocol=Capability-continual2026.0570.3——
Seq. SFTModel Scale=8B, Training Protocol=Capability-continual2026.0570.3——
GCWM-DiamondScale=4B2026.0570.27-0.181
AIMMergingModel Scale=8B, Training Protocol=Capability-continual2026.0570.2——
L&SModel Scale=8B, Training Protocol=Capability-continual2026.0569.6——
MTLModel Scale=4B, Training Protocol=Joint-training2026.0568.7——
EWCModel Scale=4B, Training Protocol=Capability-continual2026.0568.6——
AIMMergingModel Scale=4B, Training Protocol=Capability-continual2026.0565.6——
L&SModel Scale=4B, Training Protocol=Capability-continual2026.0564.7——
FOREVERModel Scale=1.7B, Training Protocol=Capability-continual2026.0558.3——
GCWMModel Scale=1.7B, Training Protocol=Capability-continual2026.0558.3——
FOREVERModel Scale=1.7B2026.0558.3——
GCWMModel Scale=1.7B2026.0558.3——
GCWM-DiamondScale=1.7B2026.0558.261.441
MTLModel Scale=1.7B, Training Protocol=Joint-training2026.0557.1——
MTLModel Scale=1.7B2026.0557.1——
OPCMScale=1.7B2026.0556.82——
OPCMModel Scale=1.7B, Training Protocol=Capability-continual2026.0556.8——
OPCMModel Scale=1.7B2026.0556.8——
EWCModel Scale=1.7B, Training Protocol=Capability-continual2026.0554.7——
EWCModel Scale=1.7B2026.0554.7——
AIMMergingModel Scale=1.7B, Training Protocol=Capability-continual2026.0553.4——
AIMMergingModel Scale=1.7B2026.0553.4——
L&SModel Scale=1.7B, Training Protocol=Capability-continual2026.0552.4——
L&SModel Scale=1.7B2026.0552.4——
Seq. SFTModel Scale=1.7B, Training Protocol=Capability-continual2026.0551.9——
Seq. SFTModel Scale=1.7B2026.0551.9——
MTLModel Scale=0.6B, Training Protocol=Joint-training2026.0540.7——
AIMMergingModel Scale=0.6B, Training Protocol=Capability-continual2026.0540.1——
AIMMergingScale=0.6B2026.0540.07——
FOREVERModel Scale=0.6B, Training Protocol=Capability-continual2026.0540——
GCWMModel Scale=0.6B, Training Protocol=Capability-continual2026.0540——
GCWM-DiamondScale=0.6B2026.0539.96-0.110.5
L&SModel Scale=0.6B, Training Protocol=Capability-continual2026.0539.6——
EWCModel Scale=0.6B, Training Protocol=Capability-continual2026.0538.9——
OPCMModel Scale=0.6B, Training Protocol=Capability-continual2026.0538.8——
Seq. SFTModel Scale=0.6B, Training Protocol=Capability-continual2026.0537.4——