| Benchmark | Model | Rank | Results |
|---|---|---|---|
| common-sense-reasoning-on-arc-easy | Phi-4 (14B) with LoopUS | #14 | Accuracy: 83.5 |
| common-sense-reasoning-on-winogrande | Phi-4 (14B) with LoopUS | #42 | Accuracy: 77.5 |
| question-answering-on-piqa | Phi-4 (14B) with LoopUS | #43 | Accuracy: 81.8 |
| language-modelling-on-lambada | Phi-4 (14B) with LoopUS | #75 | Perplexity: 3.49 |
| sentence-completion-on-hellaswag | Phi-4 (14B) with LoopUS | #149 | Accuracy: 60.58 |