OpenLM Text2SQL Leaderboard: leaderboard
Aggregated text-to-SQL leaderboard tracking model performance on WikiSQL, Spider, BIRD, and SParC/CoSQL. Evaluates via exact match, execution accuracy, and efficiency metrics across single-domain to complex cross-domain databases.
Metric: Exact Match (%). Source: web.archive.org. Source: openlm.ai. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | SeaD+Execution-Guided Decoding | 93 | |
| 2 | SDSQL+Execution-Guided Decoding | 92.7 | |
| 3 | IE-SQL+Execution-Guided Decoding | 92.5 | |
| 4 | HydraNet+Execution-Guided Decoding | 92.2 | |
| 5 | BRIDGE+Execution-Guided Decoding | 91.9 | |
| 6 | X-SQL+Execution-Guided Decoding | 91.8 | |
| 7 | SDSQL | 91.4 | |
| 8 | BRIDGE | 91.1 | |
| 9 | Text2SQLGen + EG | 91 | |
| 10 | SeqGenSQL+EG | 90.5 |
Interactive version: theaggregate.ai/benchmark?slug=openlm-text2sql-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.