Molecular Property Prediction (Regression) on ESOL
0.515RMSEMolSight-Best
Evaluation Results
| Method | Links | |
|---|---|---|
| MolSight-BestCategory=Image-only, Configuration=best curriculum-trained architecture per task2026.05 | 0.515 | |
| DMMRL2026.03 | 0.535 | |
| SGGRL2026.03 | 0.575 | |
| MvMRL2026.03 | 0.601 | |
| Weave2026.03 | 0.61 | |
| Janus-Pro 7B (ICL)Category=VLM, requires LLM/VLM at inference=true, protocol=ICL2026.05 | 0.61 | |
| MolSight (SigLIP2, S6)Category=Image-only, Configuration=single-model results2026.05 | 0.627 | |
| GraSeq2026.03 | 0.652 | |
| FP-GNN2026.03 | 0.675 | |
| MPNN2026.03 | 0.702 | |
| RNN2026.03 | 0.743 | |
| Transformer2026.03 | 0.767 | |
| TrimNet2026.03 | 0.77 | |
| Uni-MolCategory=3D GNN, requires 3D conformer generation=true2026.05 | 0.788 | |
| Opus-4.72026.07 | 0.89 | |
| GROVER2026.03 | 0.911 | |
| MolCLRCMPNNModel Type=Graph-based2026.06 | 0.911 | |
| GCN2026.03 | 0.97 | |
| GPT-4o (ICL)Category=LLM, requires LLM/VLM at inference=true, protocol=ICL2026.05 | 0.98 | |
| SVMCategory=ML Models2026.02 | 1.01 | |
| SciReasoner2026.07 | 1.03 | |
| SciReasoner2026.07 | 1.03 | |
| GAT + FPCategory=Hybrid (GNN+FP)2026.02 | 1.04 | |
| SchNetModel Type=Graph-based2026.06 | 1.045 | |
| GIN + FPCategory=Hybrid (GNN+FP)2026.02 | 1.05 | |
| GCN + FPCategory=Hybrid (GNN+FP)2026.02 | 1.07 | |
| BLIP-2 (LoRA)Category=VLM, requires LLM/VLM at inference=true, protocol=LoRA2026.05 | 1.07 | |
| MolCLRGINModel Type=Graph-based2026.06 | 1.091 | |
| N-Gram2026.03 | 1.1 | |
| MolCLRGCNModel Type=Graph-based2026.06 | 1.102 | |
| GraphSAGE + FPCategory=Hybrid (GNN+FP)2026.02 | 1.11 | |
| MolCLR2026.07 | 1.11 | |
| GPT-4oStrategy=+MOLE-RAG2026.06 | 1.138 | |
| XGBoostCategory=ML Models2026.02 | 1.14 | |
| GraSPNet2026.02 | 1.161 | |
| GPT-4o-miniStrategy=+MOLE-RAG2026.06 | 1.171 | |
| SimSGT2026.02 | 1.213 | |
| GROVE2026.02 | 1.237 | |
| MGCNModel Type=Graph-based2026.06 | 1.266 | |
| GINCategory=GNN Models2026.02 | 1.28 | |
| ECFP + OverlapBPERegression Model=XGBoost2026.04 | 1.2972 | |
| GraphCL2026.02 | 1.39 | |
| GCNCategory=GNN Models2026.02 | 1.39 | |
| GraphSAGECategory=GNN Models2026.02 | 1.39 | |
| ChemDFM14BStrategy=+MOLE-RAG2026.06 | 1.467 | |
| GROVERModel Type=Graph-based2026.06 | 1.475 | |
| GATCategory=GNN Models2026.02 | 1.48 | |
| Random ForestCategory=ML Models2026.02 | 1.5 | |
| ECFPRegression Model=XGBoost2026.04 | 1.5668 | |
| GPT-4oStrategy=Baseline2026.06 | 1.723 | |
| Qwen34B-InstStrategy=+MOLE-RAG2026.06 | 1.852 | |
| JOAO2026.02 | 1.978 | |
| GraphFP2026.02 | 2.136 | |
| Llama3.2-3BStrategy=+MOLE-RAG2026.06 | 2.181 | |
| ContextPred2026.02 | 2.19 | |
| EdgePred2026.02 | 2.368 | |
| Qwen34B-InstStrategy=Baseline2026.06 | 2.537 | |
| GPT-4o-miniStrategy=Baseline2026.06 | 2.835 | |
| Mistral7B-InstStrategy=+MOLE-RAG2026.06 | 2.89 | |
| MGSSL2026.02 | 2.936 | |
| Infomax2026.02 | 2.953 | |
| AttrMasking2026.02 | 2.954 | |
| Llama3.2-3BStrategy=Baseline2026.06 | 3.175 | |
| Mistral7B-InstStrategy=Baseline2026.06 | 3.474 | |
| ChemDFM14BStrategy=Baseline2026.06 | 4.086 | |
| Linear RegressionCategory=ML Models2026.02 | 4.4 | |
| DeepSeek-V4-Pro2026.07 | 5.24 | |
| GPT-5.52026.07 | 8.72 | |
| Kimi-K2.62026.07 | 137.05 |