AI Model Release Tracker
Supabase coding

Supabase Evalsno skills

The same Supabase scenarios, but with none of Supabase's skills loaded — so it measures what the model already knows about building on Supabase, rather than how well it follows Supabase's supplied instructions. Higher is better.

Scores from Supabase Evals, which runs the benchmark and publishes the full field.

with skillsno skills

Rankings

Higher is better
#1
Claude Opus 5
89.9%
2
Claude Sonnet 5
79.7%
3
Kimi K3
78.3%
4
GPT-5.6 Sol
75.4%
5
GPT-5.6 Luna
60.9%
6
GPT-5.4 mini
55.1%

Supabase Evals — frequently asked questions

The same Supabase scenarios, but with none of Supabase's skills loaded — so it measures what the model already knows about building on Supabase, rather than how well it follows Supabase's supplied instructions. Higher is better.