Updated

Which AI model writes the most type and compile errors?

As of Sep 27, 2026, Claude Sonnet 5 had the highest rate of type and compile errors, 116% above the all-model rate (2.16), statistically tied with Claude Fable 5.1, GPT-6 Astra, and Claude Opus 5. Claude Fable 5 had the lowest, 58% below average (0.42), statistically tied with Claude Opus 4.8 and GPT-5.6 Sol.

Models ranked by relative rate of type and compile errors, fewest first. Lower values are better.
RankModelRate
1= (statistically tied)Claude Fable 5
0.42
2= (statistically tied)Claude Opus 4.8
0.74
3= (statistically tied)GPT-5.6 Sol
0.79
4= (statistically tied)Claude Fable 5.1
1.15
5= (statistically tied)GPT-6 Astra
1.32
6= (statistically tied)Claude Opus 5
1.63
7= (statistically tied)Claude Sonnet 5
2.16

1.00 = all-model rate · lower is better · = statistically tied