| Benchmark | Model | Rank | Results |
|---|---|---|---|
| code-generation-on-mbpp | LFM-7B | – | Accuracy: 51.6 |
| long-context-understanding-on-longbench-v2 | LFM-7B | – | Accuracy: 36.1 |
| multi-task-language-understanding-on-mmlu | LFM-7B (5-shot) | – | Average (%): 69.34 |
| multi-task-language-understanding-on-mmlu-pro | LFM-7B (5-shot) | – | Accuracy: 42.42 |
| multiple-choice-question-answering-mcqa-on-arc-c | LFM-7B (25-shot) | – | Accuracy: 70.56 |
| sentence-completion-on-hellaswag | LFM-7B (10-shot) | – | Accuracy: 83.07 |
| zero-shot-question-answering-on-truthfulqa-mc2-0-shot | LFM-7B (0-shot) | – | MC2Acc: 63.89 |