GENIE Benchmark
by Stanford University · open-source · Last verified 2026-04-11
The GENIE Benchmark is a comprehensive dataset for evaluating the performance of text-to-SQL models. It includes a diverse set of SQL queries and corresponding natural language questions across multiple domains, designed to assess the generalization capabilities of these models.
https://github.com/stanfordnlp/genie ↗B+
B+—Good
Adoption: B+Quality: AFreshness: B+Citations: A+Engagement: B
Specifications
- License
- MIT License
- Pricing
- open-source
- Capabilities
- text-to-sql conversion, cross-domain generalization, semantic parsing
- Integrations
- Use Cases
- database query, virtual assistant, data analysis
- API Available
- No
- Tags
- text-to-sql, natural language processing, database, question answering
- Added
- 2026-04-11
- Completeness
- 100%
Index Score
79.2Adoption
75
Quality
85
Freshness
70
Citations
90
Engagement
65