agents benchmark
LiveSQLBench
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.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
LiveSQLBench Ranking
Higher score ranks better on this benchmark.
| 01 | MI | 40.2% | 100.0% | 1 | C |
LiveSQLBench Highlights
The leading models and scores on this benchmark.
What is LiveSQLBench?
What LiveSQLBench measures and how its scores work.
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.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- LiveSQLBench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- livesqlbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about LiveSQLBench.
Which model scores highest on LiveSQLBench?
MiniMax M3 is currently ranked first with 40.2%.
What does LiveSQLBench measure?
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.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.