LLM Stats (Bird-SQL (dev)): leaderboard
BIg Bench for LaRge-scale Database Grounded Text-to-SQL: 12,751 text-to-SQL pairs across 95 real-world databases testing complex query generation.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.0 Flash Lite | 57.4 | #450 |
| 2 | Gemini 2.0 Flash | 56.9 | #342 |
| 3 | Gemma 3 27B | 54.4 | #607 |
| 4 | Gemma 3 12B | 47.9 | #671 |
| 5 | Granite 4.2 30B | 41.85 | #428 |
| 6 | Nemotron 3 Super | 41.8 | #335 |
| 7 | Granite 4.2 8B | 41.07 | #635 |
| 8 | Gemma 3 4B | 36.3 | #1089 |
| 9 | Gemma 3 1B | 6.4 | #1528 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-bird-sql-dev · How It Works · Data refreshed daily, snapshot 2026-10-20.