Text-to-SQL under RBAC - BIRD: leaderboard
Access-controlled F1 on the BIRD split of the role-based-access text-to-SQL evaluation. Measures answering database questions while respecting the user's access permissions. Spider and LiveSQLBench are separate boards.
Metric: Access-Controlled F1 (%, BIRD). Source: arxiv.org. Saturation forecast: Around February 2027. 13 models tracked. Updated 4 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 87.65 | #262 | |
| 2 | DeepSeek V3.2 | Thinking | 87.1 | #423 |
| 3 | GPT-5 | 86.32 | #180 | |
| 4 | GPT-5 Mini | 84.03 | #365 | |
| 5 | Gemini 2.5 Flash | 75.59 | #469 | |
| 6 | Gemma 3 27B | 65.67 | #1669 | |
| 7 | GPT-4o Mini | 63.18 | #1678 | |
| 8 | DeepSeek V3.2 Coder | 57.89 | ||
| 9 | Qwen 2.5 14B Instruct | 54.64 | #1785 | |
| 10 | Snowflake-R1-7B | 51.45 | ||
| 11 | Qwen 2.5 Coder 7B Instruct | 50.75 | #3987 | |
| 12 | Llama3-SQLCoder-8B | 50.49 | ||
| 13 | Gemma 3 4B | 50.22 | #5301 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=text-to-sql-under-rbac-bird · How It Works · Data refreshed daily, snapshot 2026-11-20.