Functional Group Reasoning (Binary Classification) on FGBench Comparison
81AccuracyLFM2-2.6B-MMAI
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LFM2-2.6B-MMAITraining=ST, Reasoning=/think2026.03 | 81 | 100 | |
| LFM2-2.6B-MMAITraining=MT, Reasoning=/think2026.03 | 80.1 | 100 | |
| LFM2-2.6B-MMAITraining=ST, Reasoning=/no think2026.03 | 79.6 | 100 | |
| LFM2-2.6B-MMAITraining=MT, Reasoning=/no think2026.03 | 78.9 | 100 | |
| Opus-4-5Class=Proprietary LLM2026.03 | 77.7 | 100 | |
| Sonnet-4-5Class=Proprietary LLM2026.03 | 74.1 | 100 | |
| LFM2-2.6BClass=Liquid Foundation Model2026.03 | 71.2 | 100 | |
| GPT-5-1Class=Proprietary LLM2026.03 | 69.3 | 100 | |
| Qwen2.5-7BClass=Open-weight LLM2026.03 | 66.4 | 100 | |
| GPT-5-2Class=Proprietary LLM2026.03 | 63.7 | 100 | |
| DeepSeek-3-2Class=Open-weight LLM2026.03 | 49.4 | 99.4 | |
| Llama-3.1Parameters=8B2026.03 | 47.4 | 99.1 | |
| Llama-3-8B-MolInstClass=Foundational LLM for Chemistry2026.03 | 46.9 | 87.3 | |
| Llama-3.1Parameters=70B2026.03 | 45.6 | 100 | |
| ChemLLM-7BClass=Foundational LLM for Chemistry2026.03 | 25 | 100 | |
| LlaSMol-Mistral-7BClass=Foundational LLM for Chemistry2026.03 | 23.9 | 100 |