Advanced Property Reasoning on PolyBench (test)
0.39RgLPolyLM (Qwen-2.5-14B)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PolyLM (Qwen-2.5-14B)category=PolyBench, CoT=true2026.01 | 0.39 | 5.1 | |
| PolyLM (Phi-4-14B)category=PolyBench, CoT=true2026.01 | 0.38 | 5.21 | |
| PolyLM (ChemLLM-7B)category=PolyBench, CoT=false2026.01 | 0.35 | 4.61 | |
| PolyLM (ChemLLM-7B)category=PolyBench, CoT=true2026.01 | 0.35 | 5.18 | |
| PolyLM (Phi-4-14B)category=PolyBench, CoT=false2026.01 | 0.33 | 5.08 | |
| GPT-4ocategory=Closed-source2026.01 | 0.33 | 4.73 | |
| PolyLM (Qwen-2.5-14B)category=PolyBench, CoT=false2026.01 | 0.31 | 4.86 | |
| 4o-Minicategory=Closed-source2026.01 | 0.3 | 4.44 | |
| Sonnetcategory=Closed-source2026.01 | 0.3 | 4.88 | |
| Phi-4-14Bcategory=Off-the-shelf2026.01 | 0.28 | 5.19 | |
| Qwen-2.5-7Bcategory=Off-the-shelf2026.01 | 0.27 | 4.04 | |
| PolyLM (Qwen-2.5-7B)category=PolyBench, CoT=true2026.01 | 0.26 | 4.67 | |
| Qwen-2.5-14Bcategory=Off-the-shelf2026.01 | 0.25 | 4.56 | |
| PolyLM (Qwen-2.5-7B)category=PolyBench, CoT=false2026.01 | 0.25 | 4.96 | |
| LlaSMol-8Bcategory=Domain aligned2026.01 | 0.24 | 3.56 | |
| ChemLLM-7Bcategory=Domain aligned2026.01 | 0.24 | 3.56 | |
| Llamole-8Bcategory=Domain aligned2026.01 | 0.22 | 2.9 | |
| GPT-5category=Closed-source2026.01 | 0.17 | 5.58 | |
| ether0-24Bcategory=Domain aligned2026.01 | 0.15 | 3.19 |