Skip to content

Benchmark

Multilingual MGSM (CoT)

Math benchmarks measure how well models solve maths problems. Scores are percentages.

Models measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

82.5%
Mean
86.9%
Median
9.8%
Std dev
68.9%
10th percentile
91.6%
90th percentile
68.9%
Min
68.9%91.6%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Llama 3.1 405B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.6%
2Llama 3.1 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86.9%
3Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.9%

Data credits: Data by LLM Stats