Benchgen

Bird-SQL (dev) — Results

RankModelScore
1sensenova-loop-9b-preview71.38
2gemini-2-0-flash-lite57.4
3gemini-2-0-flash56.9
4gemma-3-27b54.4
5gemma-3-12b47.9
6granite-4-2-30b41.85
7nemotron-3-super-120b-a12b41.8
8granite-4-2-8b41.07
9gemma-3-4b36.3
10gemma-3-1b6.4
B

Bird-SQL (dev) — Phases

1 phaseActive

Phase 1·

Bird-SQL (dev)

Jul 24, 2026
Current