Cortex Analyst: managed text-to-SQL over Snowflake structure via REST
Cortex Analyst
Snowflake docs on Cortex Analyst—an LLM feature for natural-language questions on structured data, exposed as a REST API.
Based on
Cortex Analyst | Snowflake Documentation
Cortex Analyst helps build applications that answer business questions from structured data in Snowflake. Users ask in natural language and get answers without writing SQL. Available as a REST API, it can plug into Streamlit, Slack, Teams, or custom chat UIs. The docs frame production conversational analytics as hard to balance for accuracy, latency, and cost; Analyst is a managed agentic system that generates text-to-SQL and runs queries on Snowflake’s engine. Key features include self-serve natural-language analytics and an API-first integration model, powered by large language models that run inside Snowflake. Snowflake recommends moving to Cortex Agents for the same capabilities with higher answer quality.
Data and product teams can expose governed warehouse data to non-SQL users without standing up their own text-to-SQL stack, RAG patterns, or GPU planning.
Reliable answers still depend on structured schemas and business meaning in Snowflake; the API is a consumer of that shared model, not a replacement for it.
Put this to work on CoreModels
Related connectors and recipes
Take the next step
Try CoreModels, talk with our team, or explore more resources.