List your company

Chemistry Evaluation on SciBench

66.4AccuracyPoT + REFTOOL
23.44834.59945.7556.901May 27, 2025

Evaluation results

MethodLinks
PoT + REFTOOLModel Backbone=Gemini-1.5-Pro2025.0566.4—
TroVEModel Backbone=Gemini-1.5-Pro2025.0565.6—
ChemAgentModel Backbone=Gemini-1.5-Pro2025.0565.5—
PoT + RAGModel Backbone=Gemini-1.5-Pro2025.0563.7—
PoTModel Backbone=Gemini-1.5-Pro2025.0562.3—
PoT + REFTOOLModel Backbone=GPT-4o2025.0561.3—
CreatorModel Backbone=Gemini-1.5-Pro2025.0560—
ChemAgentModel Backbone=GPT-4o2025.0558.9—
PoTModel Backbone=GPT-4o2025.0558.9—
PoT + RAGModel Backbone=GPT-4o2025.0556.6—
PoT + RAGModel Backbone=GPT-42025.0554.1—
PoT + REFTOOLModel Backbone=GPT-42025.0553.4—
TroVEModel Backbone=GPT-4o2025.0552.7—
ChemAgentModel Backbone=GPT-42025.0552.5—
PoTModel Backbone=GPT-42025.0551.8—
StructChemModel Backbone=Gemini-1.5-Pro2025.0550.2—
PoT + REFTOOLModel Backbone=Llama-3.1-70B2025.0549.5—
ChemAgentModel Backbone=Llama-3.1-70B2025.0548.2—
PoT + RAGModel Backbone=Llama-3.1-70B2025.0548.1—
CreatorModel Backbone=GPT-42025.0546.9—
PoTModel Backbone=Llama-3.1-70B2025.0546.9—
LATMModel Backbone=GPT-42025.0545—
CreatorModel Backbone=GPT-4o2025.0543.3—
StructChemModel Backbone=GPT-4o2025.0540.5—
CreatorModel Backbone=Llama-3.1-70B2025.0540.1—
TroVEModel Backbone=GPT-42025.0539.2—
TroVEModel Backbone=Llama-3.1-70B2025.0538.6—
StructChemModel Backbone=Llama-3.1-70B2025.0537.9—
LATMModel Backbone=GPT-4o2025.0535.7—
LATMModel Backbone=Llama-3.1-70B2025.0531.4—
StructChemModel Backbone=GPT-42025.0529.7—
LATMModel Backbone=Gemini-1.5-Pro2025.0525.1—
ChemAgent2025.05—56.3
Creator2025.05—47.6
LATM2025.05—34.3
PoT2025.05—55
PoT + RAG2025.05—55.6
PoT + REFTOOL2025.05—57.7
StructChem2025.05—39.6
TroVE2025.05—49