# Supabase Evals (no skills) — AI model rankings

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.

5 tracked models have a published Supabase Evals (no skills) score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Kimi K3 | Moonshot AI | 100% | Jul 16 2026 |
| 2 | GPT-5.6 Sol | OpenAI | 94.7% | Jun 26 2026 |
| 2 | Claude Opus 5 | Anthropic | 94.7% | Jul 24 2026 |
| 4 | Claude Sonnet 5 | Anthropic | 78.9% | Jun 30 2026 |
| 5 | GPT-5.4 mini | OpenAI | 73.7% | Mar 17 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/supabase-evals-no-skills
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.