Machine Translation on FLORES
88.9ScoreEuroLLM-22B (old)
Evaluation Results
| Method | Links | |
|---|---|---|
| EuroLLM-22B (old)Openness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 88.9 | |
| Gemma-3-27BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 88.9 | |
| EuroLLM-9B (old)Openness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 88.8 | |
| EuroLLM-9B (new)Openness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 88.8 | |
| EuroLLM-22B (new)Openness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 88.8 | |
| Gemma-3-12BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 88.2 | |
| Mistral-3.2-24BOpenness=Open-weights, Region=European, Tuning=Instruction-tuned2026.02 | 87.9 | |
| Llama-3.3-70BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 87.9 | |
| Apertus-8BOpenness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 87.8 | |
| Qwen-3-30B-A3BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 86.8 | |
| Qwen-3-32BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 86.5 | |
| Qwen-3-14BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 86.3 | |
| Apertus-70BOpenness=Fully-open, Region=European, Tuning=Instruction-tuned2026.02 | 85 | |
| Llama-3.1-8BOpenness=Open-weights, Region=Non-European, Tuning=Instruction-tuned2026.02 | 84.3 | |
| OLMo-3.1-32BOpenness=Fully-open, Region=Non-European, Tuning=Instruction-tuned2026.02 | 81.7 | |
| OLMo-3-7BOpenness=Fully-open, Region=Non-European, Tuning=Instruction-tuned2026.02 | 70.9 | |
| GPT-3.5Parameter Scale=API, Evaluation Setup=0-shot2024.03 | 51.1 | |
| Qwen-14B-ChatParameter Scale=13~20B, Evaluation Setup=0-shot2024.03 | 17.3 | |
| Mixtral-8x7B-Instruct-v0.1Parameter Scale=13~20B, Evaluation Setup=0-shot2024.03 | 17.2 | |
| InternLM2-Chat-20B-SFTParameter Scale=13~20B, Evaluation Setup=0-shot, Training Protocol=SFT2024.03 | 17 | |
| InternLM2-Chat-20BParameter Scale=13~20B, Evaluation Setup=0-shot, Training Protocol=RLHF2024.03 | 16.9 | |
| InternLM2-Chat-7B-SFTParameter Scale=< 7B, Evaluation Setup=0-shot, Training Protocol=SFT2024.03 | 15.5 | |
| InternLM2-Chat-7BParameter Scale=< 7B, Evaluation Setup=0-shot, Training Protocol=RLHF2024.03 | 15.2 | |
| Qwen-7B-ChatParameter Scale=< 7B, Evaluation Setup=0-shot2024.03 | 12.9 | |
| Mistral-7B-Instruct-v0.2Parameter Scale=< 7B, Evaluation Setup=0-shot2024.03 | 12.8 | |
| Baichuan2-13B-ChatParameter Scale=13~20B, Evaluation Setup=0-shot2024.03 | 12.4 | |
| Baichuan2-7B-ChatParameter Scale=< 7B, Evaluation Setup=0-shot2024.03 | 11 | |
| ChatGLM3-6BParameter Scale=< 7B, Evaluation Setup=0-shot2024.03 | 9.7 | |
| Mixtral-8x7B-v0.1Model size=13 ~ 20B, shot=0-shot2024.03 | 8.8 | |
| Llama2-13BModel size=13 ~ 20B, shot=0-shot2024.03 | 7.2 | |
| Mistral-7B-v0.1Model size=< 7B, shot=0-shot2024.03 | 6.9 | |
| Llama2-13B-ChatParameter Scale=13~20B, Evaluation Setup=0-shot2024.03 | 6.6 | |
| InternLM2-20BModel size=13 ~ 20B, shot=0-shot2024.03 | 6.5 | |
| Baichuan2-13B-BaseModel size=13 ~ 20B, shot=0-shot2024.03 | 6.4 | |
| Llama2-7BModel size=< 7B, shot=0-shot2024.03 | 6 | |
| InternLM2-20B-BaseModel size=13 ~ 20B, shot=0-shot2024.03 | 6 | |
| InternLM2-7BModel size=< 7B, shot=0-shot2024.03 | 5.9 | |
| Qwen-14BModel size=13 ~ 20B, shot=0-shot2024.03 | 5.9 | |
| Baichuan2-7B-BaseModel size=< 7B, shot=0-shot2024.03 | 5.8 | |
| InternLM2-7B-BaseModel size=< 7B, shot=0-shot2024.03 | 5.5 | |
| Llama2-7B-ChatParameter Scale=< 7B, Evaluation Setup=0-shot2024.03 | 5.3 | |
| Qwen-7BModel size=< 7B, shot=0-shot2024.03 | 3.3 | |
| ChatGLM3-6B-BaseModel size=< 7B, shot=0-shot2024.03 | 0.9 |