Mathematical Reasoning on GSM8K Numerical +4 (test)
88AccuracyCLAUDE-3-OPUS
Evaluation results
| Method | Links | |
|---|---|---|
| CLAUDE-3-OPUSPrompting=CoT, Framework=DARG2024.06 | 88 | |
| GPT-4-OPrompting=CoT, Framework=DARG2024.06 | 80.4 | |
| GPT-4-TURBOPrompting=CoT, Framework=DARG2024.06 | 74.8 | |
| GEMINI-1.5-PROPrompting=CoT, Framework=DARG2024.06 | 68.6 | |
| CLAUDE-3-OPUSPrompting=CoT, Framework=DARG2024.06 | 68.5 | |
| GPT-4-TURBOPrompting=CoT, Framework=DARG2024.06 | 67.1 | |
| GPT-4-OPrompting=CoT, Framework=DARG2024.06 | 67 | |
| GEMINI-1.5-PROPrompting=CoT, Framework=DARG2024.06 | 63.5 | |
| LLAMA-3-70BPrompting=CoT, Framework=DARG2024.06 | 62.5 | |
| MIXTRAL 8x22BPrompting=CoT, Framework=DARG2024.06 | 62.5 | |
| GEMINI-1.5-FLASHPrompting=CoT, Framework=DARG2024.06 | 61.5 | |
| WIZARDLM-2 8x22BPrompting=CoT, Framework=DARG2024.06 | 59.5 | |
| COMMAND R+104BPrompting=CoT, Framework=DARG2024.06 | 57 | |
| GPT-3.5-TURBOPrompting=CoT, Framework=DARG2024.06 | 56.5 | |
| GPT-3.5-TURBOPrompting=CoT, Framework=DARG2024.06 | 55.8 | |
| DEEPSEEKMATH-7BPrompting=CoT, Framework=DARG2024.06 | 55.5 | |
| GEMINI-1.5-FLASHPrompting=CoT, Framework=DARG2024.06 | 55.4 | |
| LLAMA-3-70BPrompting=CoT, Framework=DARG2024.06 | 53.2 | |
| COMMAND R+104BPrompting=CoT, Framework=DARG2024.06 | 52.5 | |
| PHI-3-MINI-3.8BPrompting=CoT, Framework=DARG2024.06 | 51.5 | |
| MIXTRAL 8x22BPrompting=CoT, Framework=DARG2024.06 | 51.4 | |
| WIZARDLM-2 8x22BPrompting=CoT, Framework=DARG2024.06 | 46.6 | |
| LLAMA-3-8BPrompting=CoT, Framework=DARG2024.06 | 46 | |
| MIXTRAL 8x7BPrompting=CoT, Framework=DARG2024.06 | 46 | |
| PHI-3-MINI-3.8BPrompting=CoT, Framework=DARG2024.06 | 41.2 | |
| DEEPSEEKMATH-7BPrompting=CoT, Framework=DARG2024.06 | 41 | |
| LLAMA-3-8BPrompting=CoT, Framework=DARG2024.06 | 32 | |
| MISTRAL-7BPrompting=CoT, Framework=DARG2024.06 | 31.5 | |
| MIXTRAL 8x7BPrompting=CoT, Framework=DARG2024.06 | 30.8 | |
| MISTRAL-7BPrompting=CoT, Framework=DARG2024.06 | 11.4 |