Reasoning on ARC Indonesian 1.0 (test)
91AccuracyTrinity Large (MoE)
Evaluation Results
| Method | Links | |
|---|---|---|
| Trinity Large (MoE)Evaluation Setting=Zero-shot2026.02 | 91 | |
| Qwen3 8BEvaluation Setting=Zero-shot2026.02 | 87 | |
| Qwen3 4BEvaluation Setting=Zero-shot2026.02 | 83 | |
| Datology 8BEvaluation Setting=Zero-shot2026.02 | 68 | |
| Llama-3.1 8BEvaluation Setting=Zero-shot2026.02 | 60 | |
| Datology 3BEvaluation Setting=Zero-shot2026.02 | 59 | |
| Granite-4.0 MicroEvaluation Setting=Zero-shot2026.02 | 57 | |
| Llama-3.2 3BEvaluation Setting=Zero-shot2026.02 | 50 | |
| SmolLM3 3BEvaluation Setting=Zero-shot2026.02 | 48 | |
| LFM2.5 1.2BEvaluation Setting=Zero-shot2026.02 | 30 | |
| Llama-3.2 1BEvaluation Setting=Zero-shot2026.02 | 29 |