Reagent Prediction on ChemCoTBench
51.2Top-1 SimilarityChemMAS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ChemMAS2025.09 | 51.2 | 59.1 | 67.7 | |
| GPT5Evaluation Protocol=Zero-shot2025.09 | 40.1 | 55.1 | 61.1 | |
| Gemini2.5-ProEvaluation Protocol=Zero-shot2025.09 | 38.9 | 52.1 | 59.8 | |
| Qwen3-235B-A22BEvaluation Protocol=Zero-shot2025.09 | 36.4 | 50.2 | 58.7 | |
| DeepSeek-R1Evaluation Protocol=Zero-shot2025.09 | 35.2 | 47.6 | 55.4 | |
| Claude3.7-SonnetEvaluation Protocol=Zero-shot2025.09 | 34.3 | 48 | 54.3 |