Mathematical Reasoning on GSM8K (test) (GSM8K Score)
57.1GSM8K ScoreExpertWeaver-Instruct
Evaluation results
| Method | Links | |
|---|---|---|
| ExpertWeaver-InstructInstruction-tuned=true2026.02 | 57.1 | |
| LLaMA-MoE-v22026.02 | 55 | |
| OLMoE-1B-7B2026.02 | 40.94 | |
| LLaMA-MoE-3B-7B2026.02 | 4.62 |