Skip to content

Benchmark

BIG-Bench

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

72.7%
Mean
74.9%
Median
3.2%
Std dev
68.2%
10th percentile
75%
90th percentile
68.2%
Min
68.2%75%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1gemini-1.0-pro· LLM Stats data75%
2Gemma 2 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.9%
3Gemma 2 9B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.2%

Data credits: Data by LLM Stats