Reading Comprehension on Belebele Japanese (test)
87AccuracyTrinity Large (MoE)
Evaluation Results
| Method | Links | |
|---|---|---|
| Trinity Large (MoE)Zero-shot=true2026.02 | 87 | |
| Qwen3 8BZero-shot=true2026.02 | 81 | |
| Qwen3 4BZero-shot=true2026.02 | 79 | |
| Granite-4.0 MicroZero-shot=true2026.02 | 71 | |
| Datology 8BZero-shot=true2026.02 | 66 | |
| Llama-3.1 8BZero-shot=true2026.02 | 65 | |
| Datology 3BZero-shot=true2026.02 | 55 | |
| SmolLM3 3BZero-shot=true2026.02 | 51 | |
| Llama-3.2 3BZero-shot=true2026.02 | 44 | |
| LFM2.5 1.2BZero-shot=true2026.02 | 44 | |
| Llama-3.2 1BZero-shot=true2026.02 | 25 |