Skip to content

Benchmark

MBPP EvalPlus (base)

Coding benchmarks compare models on programming tasks. Scores are percentages.

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

Score distribution

72.8%
Mean
72.8%
Median
0%
Std dev
72.8%
10th percentile
72.8%
90th percentile
72.8%
Min

Leaderboard

#ModelScore
1Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.8%

Data credits: Data by LLM Stats