parkervg/blendsql

Query language for blending SQL and local language models across structured + unstructured data, with type constraints.

48
/ 100
Emerging

BlendSQL helps data analysts and engineers combine the power of traditional SQL queries with large language models (LLMs) to extract insights from both structured tables and unstructured text. You can input your existing datasets and unstructured content, then use a familiar SQL-like syntax to ask complex questions that leverage an LLM, receiving a structured table or specific answer in return. This is designed for professionals who need to query and analyze diverse data types using a single, unified approach.

163 stars.

Use this if you need to perform complex data analysis that involves querying structured databases and simultaneously extracting information or making inferences from unstructured text using AI.

Not ideal if your data analysis needs are purely SQL-based without any requirement for integrating large language model capabilities, or if you prefer to keep your LLM interactions separate from your database queries.

data-analysis business-intelligence information-retrieval knowledge-discovery data-querying
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

163

Forks

12

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/parkervg/blendsql"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.