Machine Translation on WMT De-En 2022
33.87BLEUGPT-4
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GPT-42025.12 | 33.87 | 85.62 | |
| ALMA-13Bparameters=13B2025.12 | 31.14 | 84.56 | |
| Qwen3-1.7B-SGRPOparameters=1.7B, training=SGRPO2025.12 | 30.18 | 86.17 | |
| ALMA-7Bparameters=7B2025.12 | 29.56 | 83.95 | |
| Qwen3-1.7B-SFTparameters=1.7B, training=SFT2025.12 | 28.55 | 85.79 | |
| TIM-7Bparameters=7B2025.12 | 27.91 | 82.8 | |
| Bayling-13Bparameters=13B2025.12 | 27.34 | 83.02 | |
| NLLB-54Bparameters=54B2025.12 | 26.89 | 78.94 | |
| BigTranslate-13Bparameters=13B2025.12 | 23.35 | 80.68 |