OpenLM Text2SQL - BIRD R-VES: leaderboard
Reward-based Valid Efficiency Score on BIRD, evaluating both correctness and execution efficiency of generated SQL queries against large-scale databases.
Metric: R-VES (%). Source: openlm.ai. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | OpenSearch-SQL, v2 + GPT-4o | 69.36 | |
| 2 | ExSL + granite-34b-code | 68.79 | |
| 3 | CHASE-SQL + Gemini | 68.44 | |
| 4 | Distillery + GPT-4o | 67.41 | |
| 5 | AskData + GPT-4o | 66.92 | |
| 6 | Insights AI | 66.39 | |
| 7 | ExSL + granite-20b-code | 66.25 | |
| 8 | RECAP + Gemini | 65.7 |
Interactive version: theaggregate.ai/benchmark?slug=openlm-text2sql-bird-r-ves · How It Works · Data refreshed daily, snapshot 2026-10-20.