Mathematical Reasoning on GSM8K Numerical +8 (test)
67.8AccuracyCLAUDE-3-OPUS
Evaluation Results
| Method | Links | |
|---|---|---|
| CLAUDE-3-OPUSPrompting=CoT, Framework=DARG2024.06 | 67.8 | |
| GPT-4-OPrompting=CoT, Framework=DARG2024.06 | 42 | |
| GPT-4-TURBOPrompting=CoT, Framework=DARG2024.06 | 39.2 | |
| GEMINI-1.5-PROPrompting=CoT, Framework=DARG2024.06 | 33 | |
| COMMAND R+104BPrompting=CoT, Framework=DARG2024.06 | 32.8 | |
| GEMINI-1.5-FLASHPrompting=CoT, Framework=DARG2024.06 | 26.4 | |
| MIXTRAL 8x22BPrompting=CoT, Framework=DARG2024.06 | 26.2 | |
| GPT-3.5-TURBOPrompting=CoT, Framework=DARG2024.06 | 26.2 | |
| LLAMA-3-70BPrompting=CoT, Framework=DARG2024.06 | 21.4 | |
| DEEPSEEKMATH-7BPrompting=CoT, Framework=DARG2024.06 | 20.4 | |
| WIZARDLM-2 8x22BPrompting=CoT, Framework=DARG2024.06 | 18 | |
| MIXTRAL 8x7BPrompting=CoT, Framework=DARG2024.06 | 14.4 | |
| PHI-3-MINI-3.8BPrompting=CoT, Framework=DARG2024.06 | 13.8 | |
| LLAMA-3-8BPrompting=CoT, Framework=DARG2024.06 | 12.8 | |
| MISTRAL-7BPrompting=CoT, Framework=DARG2024.06 | 5.4 |