| ethics-on-ethics | RuGPT-3 Large | #1 | Accuracy: 68.6 |
| ethics-on-ethics | RuGPT-3 Meduim | #2 | Accuracy: 68.3 |
| ethics-on-ethics | RuGPT-3 Small | #3 | Accuracy: 55.5 |
| ethics-on-ethics | Human benchmark | #4 | Accuracy: 52.9 |
| ethics-on-ethics-2 | Human benchmark | #1 | Accuracy: 67.6 |
| ethics-on-ethics-2 | RuGPT-3 Small | #2 | Accuracy: 60.9 |
| ethics-on-ethics-2 | RuGPT-3 Large | #3 | Accuracy: 44.9 |
| ethics-on-ethics-2 | RuGPT-3 Medium | #4 | Accuracy: 44.1 |
| logical-reasoning-on-ruworldtree | Human benchmark | #1 | Accuracy: 83.7 |
| logical-reasoning-on-ruworldtree | RuGPT-3 Large | #2 | Accuracy: 40.7 |
| logical-reasoning-on-ruworldtree | RuGPT-3 Medium | #3 | Accuracy: 38.0 |
| logical-reasoning-on-ruworldtree | RuGPT-3 Small | #4 | Accuracy: 34.0 |
| logical-reasoning-on-winograd-automatic | Human benchmark | #1 | Accuracy: 87.0 |
| logical-reasoning-on-winograd-automatic | RuGPT-3 Small | #2 | Accuracy: 57.9 |
| logical-reasoning-on-winograd-automatic | RuGPT-3 Medium | #3 | Accuracy: 57.2 |
| logical-reasoning-on-winograd-automatic | RuGPT-3 Large | #4 | Accuracy: 55.5 |
| question-answering-on-chegeka | Human benchmark | #1 | Accuracy: 64.5 |
| question-answering-on-chegeka | RuGPT-3 Large | #2 | Accuracy: 00 |
| question-answering-on-chegeka | RuGPT-3 Medium | #3 | Accuracy: 00 |
| question-answering-on-chegeka | RuGPT-3 Small | #4 | Accuracy: 00 |
| question-answering-on-multiq | Human benchmark | #1 | Accuracy: 91.0 |
| question-answering-on-multiq | RuGPT-3 Large | #2 | Accuracy: 00 |
| question-answering-on-multiq | RuGPT-3 Medium | #3 | Accuracy: 00 |
| question-answering-on-multiq | RuGPT-3 Small | #4 | Accuracy: 00 |
| question-answering-on-ruopenbookqa | Human benchmark | #1 | Accuracy: 86.5 |
| question-answering-on-ruopenbookqa | RuGPT-3 Small | #2 | Accuracy: 57.9 |
| question-answering-on-ruopenbookqa | RuGPT-3 Medium | #3 | Accuracy: 57.2 |
| question-answering-on-ruopenbookqa | RuGPT-3 Large | #4 | Accuracy: 55.5 |