Skip to content

Benchmark

Arena ELO - Coding

Coding benchmarks compare models on programming tasks. Higher scores are better.

Models measured
361
Scores recorded
418
Sources
1
IQ weight
1.28

Score distribution

1,325
Mean
1,374.5
Median
162.1
Std dev
1,081.6
10th percentile
1,488
90th percentile
682.8
Min
682.81,564.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.28 for chat models, measured over 361 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Showing 401–418 of 418 results.

Data credits: LMArena, CC BY 4.0 (licence)