Bird-SQL (dev)

Multilingual: no

BIRD (BIg Bench for LaRge-scale Database Grounded Text-to-SQLs) is a comprehensive text-to-SQL benchmark containing 12,751 question-SQL pairs across 95 databases (33.4 GB total) spanning 37+ professional domains. It evaluates large language models' ability to convert natural language to executable SQL queries in real-world scenarios with complex database schemas and dirty data.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: reasoning. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 7 of 7 results

Gemini 2.0 Flash-Lite self-reported llm-stats

57.4%
Gemini 2.0 Flash self-reported llm-stats

56.9%
Gemma 3 27B self-reported llm-stats

54.4%
Gemma 3 12B self-reported llm-stats

47.9%
Nemotron 3 Super (120B A12B) self-reported llm-stats

41.8%
Gemma 3 4B self-reported llm-stats

36.3%
Gemma 3 1B self-reported llm-stats

6.4%