LiveSQLBench

coding

Multilingual: no

LiveSQLBench evaluates models on generating correct SQL queries against live PostgreSQL databases. The LiveSQLBench-Base-Full v1 dataset contains 600 questions across 22 PostgreSQL databases, testing real-world database reasoning and query generation.

Methodology

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