Machine Translation on FLORES-200 XX ⇔ XX
88.74XCOMET-XXL ScoreGemini 3.1 Pro
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 3.1 ProThinking Mode=true2026.05 | 88.74 | 77.6 | 90.97 | |
| GPT-5.5Thinking Mode=true2026.05 | 88.44 | 77.36 | 90.93 | |
| GPT-5.5Thinking Mode=false2026.05 | 87.92 | 77.2 | 90.37 | |
| Gemma4-31BThinking Mode=true2026.05 | 87.77 | 77.28 | 90.11 | |
| Hy-MT2-30B-A3B2026.05 | 87.47 | 76.34 | 88.79 | |
| Hy-MT2-7B2026.05 | 86.89 | 76.03 | 87.23 | |
| Gemma4-31BThinking Mode=false2026.05 | 86.84 | 76.16 | 88.67 | |
| DeepSeek-V4-ProThinking Mode=true2026.05 | 86.81 | 76.66 | 89.86 | |
| Gemma4-26B-A4BThinking Mode=true2026.05 | 86.8 | 77.14 | 89.58 | |
| GLM5.1Thinking Mode=true2026.05 | 86.57 | 76.84 | 89.92 | |
| Qwen3.5-397B-A17BThinking Mode=false2026.05 | 86.29 | 76.39 | 88.87 | |
| Kimi K2.6Thinking Mode=true2026.05 | 86.17 | 76.72 | 89.44 | |
| Qwen3.5-397B-A17BThinking Mode=true2026.05 | 86.06 | 76.52 | 89.28 | |
| Gemma4-26B-A4BThinking Mode=false2026.05 | 85.6 | 76.58 | 87.97 | |
| Qwen3.6-35B-A3BThinking Mode=true2026.05 | 84.84 | 75.95 | 88.21 | |
| Kimi K2.6Thinking Mode=false2026.05 | 84.05 | 75.84 | 87.43 | |
| GLM5.1Thinking Mode=false2026.05 | 83.83 | 75.4 | 87.14 | |
| DeepSeek-V4-ProThinking Mode=false2026.05 | 83.29 | 67.64 | 75.46 | |
| Qwen3.6-35B-A3BThinking Mode=false2026.05 | 82.11 | 74.81 | 84.92 | |
| Gemma4-E4BThinking Mode=true2026.05 | 81.29 | 74.1 | 83.74 | |
| HY-MT1.5-7B2026.05 | 80.98 | 73.36 | 78.3 | |
| Microsoft-Translator2026.05 | 80.13 | 72.42 | 83.85 | |
| Gemma4-E4BThinking Mode=false2026.05 | 79.96 | 72.91 | 80.97 | |
| Hy-MT2-1.8B2026.05 | 79.77 | 73.41 | 78.64 | |
| Gemma4-E2BThinking Mode=true2026.05 | 79.69 | 73.78 | 82.54 | |
| HY-MT1.5-1.8B2026.05 | 78.4 | 71.82 | 75.12 | |
| Doubao-Translator2026.05 | 76.54 | 69.92 | 79.72 | |
| iFLYTEK-Translator2026.05 | 76.53 | 69.9 | — | |
| Gemma4-E2BThinking Mode=false2026.05 | 75.97 | 71 | 76.64 | |
| Tower-Plus-72B2026.05 | 70.02 | 65.53 | 67.85 |