SCALE SQL Optimization — leaderboard
Metric: Ability Score (%). Source: sql-llm-leaderboard.com. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.5 Flash | 69.2 |
| 2 | Gemini 3.1 Pro (Preview) | 67.6 |
| 3 | Claude Opus 4.8 | 67.1 |
| 4 | Gemini 3 Pro | 65.7 |
| 5 | GPT-5 | 65.5 |
| 6 | Qwen 3.7 Max | 65.5 |
| 7 | Claude Opus 4.7 | 64.1 |
| 8 | MiniMax-M2.5 | 64 |
| 9 | GPT-5.5 | 63.5 |
| 10 | O4 Mini (High) | 63.1 |
| 11 | Claude Opus 4.5 | 63 |
| 12 | Claude Sonnet 4.5 | 62.8 |
| 13 | Qwen 3.5 397B A17B | 61.1 |
| 14 | Claude Sonnet 4.6 | 60.8 |
| 15 | DeepSeek V4 Pro | 60.4 |
Interactive version: aibenchmarks.dev/benchmark?slug=scale-sql-optimization · How the rankings work · Data refreshed daily, snapshot 2026-07-19.