Mathematical Reasoning on GSM8K (Acc, pass@5)
80.64AccuracyLlama-3-8B-Instruct
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama-3-8B-InstructShot Setting=5-shot2026.06 | 80.64 | 88.23 | |
| Qwen2.5-7BShot Setting=5-shot2026.06 | 79.28 | 87.64 | |
| Med42-Llama3-8BShot Setting=5-shot2026.06 | 77.06 | 85.8 | |
| Llama-3-8B-InstructShot Setting=0-shot2026.06 | 75.82 | 83.47 | |
| Qwen2.5-7BShot Setting=0-shot2026.06 | 73.47 | 81.83 | |
| Med42-Llama3-8BShot Setting=0-shot2026.06 | 71.89 | 80.14 | |
| Qwen2.5-Med-7BShot Setting=5-shot2026.06 | 71.35 | 80.63 | |
| ReLARShot Setting=5-shot2026.06 | 71.28 | 84.2 | |
| ReLARShot Setting=0-shot2026.06 | 68.45 | 78.23 | |
| MedGemma-4BShot Setting=5-shot2026.06 | 66.71 | 75.29 | |
| Qwen2.5-Med-7BShot Setting=0-shot2026.06 | 64.41 | 73.85 | |
| Mistral-7B-InstructShot Setting=5-shot2026.06 | 63.82 | 82.47 | |
| MedGemma-4BShot Setting=0-shot2026.06 | 61.18 | 70.32 | |
| Mistral-7B-v0.3Shot Setting=5-shot2026.06 | 57.84 | 79.36 | |
| Gemma-7BShot Setting=5-shot2026.06 | 56.28 | 76.43 | |
| Med42-Mistral-7BShot Setting=5-shot2026.06 | 54.82 | 81.47 | |
| Mistral-7B-InstructShot Setting=0-shot2026.06 | 53.47 | 74.83 | |
| Gemma-7BShot Setting=0-shot2026.06 | 47.63 | 68.84 | |
| Med42-Mistral-7BShot Setting=0-shot2026.06 | 46.53 | 72.84 | |
| Mistral-7B-v0.3Shot Setting=0-shot2026.06 | 40.63 | 67.84 | |
| LLaMA-2-7BShot Setting=5-shot2026.06 | 30.56 | 48.73 | |
| Falcon-7BShot Setting=5-shot2026.06 | 25.63 | 36.82 | |
| LLaMA-2-7BShot Setting=0-shot2026.06 | 23.82 | 31.47 | |
| Falcon-7BShot Setting=0-shot2026.06 | 17.84 | 24.47 |