Design & Synthesis on PolyBench (test)
82Similarity ScorePolyLM (Phi-4-14B)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PolyLM (Phi-4-14B)category=PolyBench, CoT=false2026.01 | 82 | 66 | |
| PolyLM (Phi-4-14B)category=PolyBench, CoT=true2026.01 | 75 | 82 | |
| PolyLM (ChemLLM-7B)category=PolyBench, CoT=false2026.01 | 74 | 95 | |
| PolyLM (Qwen-2.5-14B)category=PolyBench, CoT=true2026.01 | 70 | 94 | |
| PolyLM (Qwen-2.5-14B)category=PolyBench, CoT=false2026.01 | 68 | 81 | |
| PolyLM (Qwen-2.5-7B)category=PolyBench, CoT=false2026.01 | 64 | 72 | |
| PolyLM (ChemLLM-7B)category=PolyBench, CoT=true2026.01 | 63 | 66 | |
| PolyLM (Qwen-2.5-7B)category=PolyBench, CoT=true2026.01 | 43 | 62 | |
| Sonnetcategory=Closed-source2026.01 | 26 | 29 | |
| GPT-5category=Closed-source2026.01 | 26 | 21 | |
| Llamole-8Bcategory=Domain aligned2026.01 | 24 | 62 | |
| Phi-4-14Bcategory=Off-the-shelf2026.01 | 23 | 65 | |
| 4o-Minicategory=Closed-source2026.01 | 23 | 9 | |
| ChemLLM-7Bcategory=Domain aligned2026.01 | 20 | 25 | |
| Qwen-2.5-14Bcategory=Off-the-shelf2026.01 | 18 | 81 | |
| ether0-24Bcategory=Domain aligned2026.01 | 18 | 52 | |
| Qwen-2.5-7Bcategory=Off-the-shelf2026.01 | 17 | 68 | |
| LlaSMol-8Bcategory=Domain aligned2026.01 | 16 | 31 | |
| GPT-4ocategory=Closed-source2026.01 | 16 | 11 |