SQL Capability Leaderboard: leaderboard
Monthly SQL capability leaderboard measuring SQL understanding, optimization, and dialect conversion ability scores for LLMs and SQL-focused applications.
Metric: Average Ability Score. Source: sql-llm-leaderboard.com. Source: sql-llm-leaderboard.com. Saturation forecast: Around January 2027. 37 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash | 76.4 | #60 |
| 2 | Gemini 3 Pro | 76.27 | #80 |
| 3 | Qwen 3.7 Max | 75.5 | #75 |
| 4 | GPT-5 | 74.23 | #98 |
| 5 | Claude Sonnet 4.6 | 73.73 | #90 |
| 6 | Claude Opus 4.7 | 73.43 | #49 |
| 7 | Gemini 3.1 Pro (Preview) | 73.17 | #58 |
| 8 | GPT-5.5 | 72.93 | #31 |
| 9 | O4 Mini (High) | 72.87 | #180 (O4 Mini) |
| 10 | Gemini 2.5 Pro | 72.83 | #152 |
| 11 | Claude Opus 4.8 | 72.73 | #37 |
| 12 | Claude Sonnet 4.5 | 72.1 | #142 |
| 13 | Gemini 2.5 Flash | 71.43 | #245 |
| 14 | DeepSeek R1 | 70.9 | #251 |
| 15 | MiniMax-M2.5 | 70.7 | #303 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=sql-capability-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.