Multiple-choice question answering on Multiple-choice benchmarks ES
69.98AccuracyQwen3 14B merge-ES
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3 14B merge-ESBackbone=Qwen3 14B, Variant=merge-ES2026.03 | 69.98 | |
| Qwen3 14B merge-multiBackbone=Qwen3 14B, Variant=merge-multi2026.03 | 69.47 | |
| Qwen3 14B merge-CABackbone=Qwen3 14B, Variant=merge-CA2026.03 | 69.39 | |
| Qwen3 14B merge-GLBackbone=Qwen3 14B, Variant=merge-GL2026.03 | 69.3 | |
| Qwen3 14B merge-EUBackbone=Qwen3 14B, Variant=merge-EU2026.03 | 68.88 | |
| Llama 3.1 8B merge-ESBackbone=Llama 3.1 8B, Variant=merge-ES2026.03 | 68.87 | |
| Llama 3.1 8B merge-multiBackbone=Llama 3.1 8B, Variant=merge-multi2026.03 | 68.46 | |
| Llama 3.1 8B merge-GLBackbone=Llama 3.1 8B, Variant=merge-GL2026.03 | 68.23 | |
| Llama 3.1 8B InstructBackbone=Llama 3.1 8B, Variant=Instruct2026.03 | 68.22 | |
| Llama 3.1 8B merge-EUBackbone=Llama 3.1 8B, Variant=merge-EU2026.03 | 67.87 | |
| Llama 3.1 8B merge-CABackbone=Llama 3.1 8B, Variant=merge-CA2026.03 | 67.87 | |
| Qwen3 14B InstructBackbone=Qwen3 14B, Variant=Instruct2026.03 | 67.87 | |
| Qwen3 8B merge-ESBackbone=Qwen3 8B, Variant=merge-ES2026.03 | 65.91 | |
| Qwen3 8B merge-CABackbone=Qwen3 8B, Variant=merge-CA2026.03 | 65.9 | |
| Qwen3 8B merge-EUBackbone=Qwen3 8B, Variant=merge-EU2026.03 | 65.8 | |
| Qwen3 8B merge-multiBackbone=Qwen3 8B, Variant=merge-multi2026.03 | 65.65 | |
| Qwen3 8B merge-GLBackbone=Qwen3 8B, Variant=merge-GL2026.03 | 65.22 | |
| Llama 3.1 8B joint-EUBackbone=Llama 3.1 8B, Variant=joint-EU2026.03 | 64.59 | |
| Qwen3 8B InstructBackbone=Qwen3 8B, Variant=Instruct2026.03 | 64.04 | |
| ALIA 40B InstructBackbone=ALIA 40B, Variant=Instruct2026.03 | 62.93 | |
| Salamandra 7B InstructBackbone=Salamandra 7B, Variant=Instruct2026.03 | 52.79 | |
| Salamandra 2B InstructBackbone=Salamandra 2B, Variant=Instruct2026.03 | 34.68 |