List your company

Tool Selection on APIGen sampled (test)

99.4Tool Selection F1 (2 calls)TokMem (w/ adapt)
6.8430.8754.978.93Oct 1, 2025

Evaluation results

MethodLinks
TokMem (w/ adapt)Model=Llama 3.1 8B, #Params=0.20M2025.1099.497.910099.1
TokMem (w/ adapt)Model=Llama 3.2 3B, #Params=0.15M2025.1099.298.210099.2
Fine-TuingModel=Llama 3.1 8B, #Params=3.41M2025.1098.897.298.298.1
Fine-TuingModel=Llama 3.2 3B, #Params=2.29M2025.1098.798.196.897.9
TokMem (w/ adapt)Model=Llama 3.2 1B, #Params=0.10M2025.1098.49898.998.4
TokMem (w/o adapt)Model=Llama 3.2 1B, #Params=0.10M2025.1086.880.990.886.2
TokMem (w/o adapt)Model=Llama 3.1 8B, #Params=0.20M2025.1084.98281.682.8
TokMem (w/o adapt)Model=Llama 3.2 3B, #Params=0.15M2025.1082.679.367.276.4
ICLModel=Llama 3.1 8B, #Params=–2025.1079.772.975.476
RAGModel=Llama 3.1 8B, #Params=–2025.1079.675.39382.6
RAGModel=Llama 3.2 3B, #Params=–2025.1078.171.269.372.8
ICLModel=Llama 3.2 3B, #Params=–2025.1066.859.259.661.9
RAGModel=Llama 3.2 1B, #Params=–2025.1029.510.810.516.9
ICLModel=Llama 3.2 1B, #Params=–2025.1027.611.110.516.4
Fine-TuingModel=Llama 3.2 1B, #Params=0.85M2025.1010.49.579