Multitask Language Understanding on MMLU Japanese (test)
73AccuracyTrinity Large (MoE)
Evaluation results
| Method | Links | |
|---|---|---|
| Trinity Large (MoE)Zero-shot=true2026.02 | 73 | |
| Qwen3 8BZero-shot=true2026.02 | 63 | |
| Qwen3 4BZero-shot=true2026.02 | 59 | |
| Datology 8BZero-shot=true2026.02 | 52 | |
| LFM2.5 1.2BZero-shot=true2026.02 | 49 | |
| Llama-3.1 8BZero-shot=true2026.02 | 48 | |
| Granite-4.0 MicroZero-shot=true2026.02 | 48 | |
| SmolLM3 3BZero-shot=true2026.02 | 46 | |
| Datology 3BZero-shot=true2026.02 | 44 | |
| Llama-3.2 3BZero-shot=true2026.02 | 40 | |
| Llama-3.2 1BZero-shot=true2026.02 | 26 |