UNVERIFIED · Qwen2 Technical Report · arXiv:2407.10671 · Tables 2, 4
Multilingual MGSM aggregate score.
Results
10 runsResult trend
| literature/qwen2_7b | self reported | 57.50 | 2026-09-04 19:49:26 |
| literature/qwen1.5_7b | self reported | 37.30 | 2026-09-04 19:49:26 |
| literature/llama_3_8b | self reported | 36.30 | 2026-09-04 19:49:26 |
| literature/gemma_7b | self reported | 39.10 | 2026-09-04 19:49:26 |
| literature/mistral_7b_v0.2 | self reported | 26.30 | 2026-09-04 19:49:25 |
| literature/qwen2_72b | self reported | 76.00 | 2026-09-04 19:49:25 |
| literature/qwen1.5_110b | self reported | 64.40 | 2026-09-04 19:49:25 |
| literature/qwen1.5_72b | self reported | 61.70 | 2026-09-04 19:49:25 |
| literature/llama_3_70b | self reported | 67.10 | 2026-09-04 19:49:25 |
| literature/mixtral_8x22b | self reported | 62.90 | 2026-09-04 19:49:24 |