| language-modelling-on-lambada | Falcon-180B (1-shot) | #4 | Accuracy: 84.4 |
| common-sense-reasoning-on-winogrande | Falcon-180B (5-shot) | #10 | Accuracy: 87.1 |
| question-answering-on-boolq | Falcon-180B | #13 | Accuracy: 87.8 |
| code-generation-on-humaneval | Falcon-180B | #115 | Pass@1: 35.4 |
| multi-task-language-understanding-on-mmlu | Falcon 40B | – | Average (%): 57.0 |
| multi-task-language-understanding-on-mmlu | Falcon 7B (5-shot) | – | Average (%): 28.0 |
| sentence-completion-on-hellaswag | Falcon-180B (0-shot) | – | Accuracy: 85.9 |
| sentence-completion-on-hellaswag | Falcon-40B (0-shot) | – | Accuracy: 82.7 |
| sentence-completion-on-hellaswag | Falcon-7B (0-shot) | – | Accuracy: 76.3 |