Portuguese Language Evaluation on FRMT
70.5AccuracyAMALIA-9B-SFT
Evaluation Results
| Method | Links | |
|---|---|---|
| AMALIA-9B-SFTModel Category=Fully open models, Training=SFT, Instruction-tuned=true2026.03 | 70.5 | |
| EuroLLM-9BModel Category=Fully open models, Instruction-tuned=true2026.03 | 70.4 | |
| AMALIA-9B-DPOModel Category=Fully open models, Training=DPO, Instruction-tuned=true2026.03 | 70.2 | |
| Gemma 3-12BModel Category=Open weight models, Instruction-tuned=true2026.03 | 69.5 | |
| Apertus-8BModel Category=Fully open models, Instruction-tuned=true2026.03 | 67.2 | |
| Gemma 2-9BModel Category=Open weight models, Instruction-tuned=true2026.03 | 66.5 | |
| Llama 3.1-8BModel Category=Open weight models, Instruction-tuned=true2026.03 | 66.3 | |
| Gervasio-8BModel Category=Open weight models, Instruction-tuned=true2026.03 | 66.3 | |
| Ministral-8BModel Category=Open weight models, Instruction-tuned=true2026.03 | 66.1 | |
| Salamandra-7BModel Category=Fully open models, Instruction-tuned=true2026.03 | 65.6 | |
| Qwen 3-8BModel Category=Open weight models, Instruction-tuned=true2026.03 | 65.1 | |
| Qwen 2.5-7BModel Category=Open weight models, Instruction-tuned=true2026.03 | 63.9 | |
| Mistral-7BModel Category=Open weight models, Instruction-tuned=true2026.03 | 63.3 | |
| OLMo 2-7BModel Category=Fully open models, Instruction-tuned=true2026.03 | 61 |