DBLifeBench - Text2SQL: leaderboard
Text-to-SQL in the operation phase of DBLifeBench: 1,534 questions over BIRD and textbook databases, including UPDATE and DELETE statements, scored by execution accuracy in percent; higher is better.
Metric: Execution accuracy (%; 1,534 text-to-SQL examples built from BIRD and textbook databases, including UPDATE and DELETE statements; temperature 0.3, top-p 0.2, SQLite). Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 56.19 | #741 |
| 2 | GPT-4o Mini | 47.72 | #1623 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=dblifebench-text2sql · How It Works · Data refreshed daily, snapshot 2026-11-13.