| Benchmark | Model | Rank | Results |
|---|---|---|---|
| common-sense-reasoning-on-arc-easy | Qwen3-0.6B (Averis FP4) | – | Accuracy: 50.72 |
| language-modelling-on-lambada | Qwen3-0.6B (Averis FP4) | – | Accuracy: 38.62 |
| question-answering-on-boolq | Qwen3-0.6B (Averis FP4) | – | Accuracy: 57.46 |
| question-answering-on-piqa | Qwen3-0.6B (Averis FP4) | – | Accuracy: 66.7 |
| reading-comprehension-on-race | Qwen3-0.6B (Averis FP4) | – | Accuracy: 50.3600 |
| sentence-completion-on-hellaswag | Qwen3-0.6B (Averis FP4) | – | Accuracy: 37.51 |