Updated

Best AI model for Go code

As of Sep 30, 2026, Claude Sonnet 5 had the lowest relative bug rate in Go code (0.18, 82% fewer counted bugs than the Go average), among 7 ranked models.

Models ranked by relative bug rate in Go code. Lower values are better.
RankModelRate
1Claude Sonnet 5
0.18
2= (statistically tied)Claude Opus 4.8
0.78
3= (statistically tied)GPT-5.6 Sol
0.84
4= (statistically tied)GPT-6 Astra
0.86
5Claude Opus 5
1.20
6= (statistically tied)Claude Fable 5.1
1.87
7= (statistically tied)Claude Opus 5.5
2.52

1.00 = Go average · lower is better · = statistically tied

Other languages

Questions and answers

Which AI model writes the most bugs in Go?
As of Sep 30, 2026, Claude Opus 5.5 had the highest relative bug rate in Go code (2.52, 152% more counted bugs than the Go average), statistically tied with Claude Fable 5.1.