SQL-native LLM functions, Cortex Search, Analyst, and Agents inside Snowflake
Snowflake documents Cortex as a suite of managed AI features in docs.snowflake.com: Cortex AI Functions (AI_COMPLETE, AI_EXTRACT, AI_SENTIMENT, AI_TRANSLATE, and related SQL/Python surfaces) run unstructured analytics on text and images with models from OpenAI, Anthropic, Meta, Mistral AI, and DeepSeek subject to CORTEX_USER privileges. Cortex Search provides hybrid retrieval for RAG, Cortex Analyst turns natural-language questions into SQL over semantic models, and Cortex Agents orchestrate Analyst, Search, custom UDFs, and optional web search via agent APIs. Cortex Code adds CLI and Snowsight assistance for data engineering workflows, and Snowflake documents MCP support within the Cortex Code tooling hierarchy.
Use cases
- Run governed LLM tasks directly on warehouse tables without exporting data
- Expose semantic-model-backed chat analytics to business users through Analyst APIs
- Build agentic apps that ground answers in Snowflake Search indexes and SQL tools
- Estimate token spend with AI_COUNT_TOKENS before promoting prompts to production
- Pair Cortex features with Snowflake-managed MCP servers for IDE assistants
Key features
- Cortex AI Functions documented for completions, extraction, sentiment, translation, and PII redaction via SQL
- Cortex Search hybrid engine for low-latency RAG over unstructured enterprise data
- Cortex Analyst REST API converting natural language to SQL using semantic model YAML
- Cortex Agents combining Analyst, Search, stored procedures, and LLM orchestration
- Model lifecycle stages (Preview, GA, Legacy, EOL) and role-gated model access per Snowflake AI docs
Who Is It For?
- Data engineers shipping AI features without separate vector stacks
- Analytics teams needing natural-language access to curated semantic models
- Enterprises requiring RBAC-aligned AI inside existing Snowflake governance
Frequently Asked Questions
- Which role grants Cortex access?
- Snowflake docs reference SNOWFLAKE.CORTEX_USER and feature-specific roles such as CORTEX_AGENT_USER for Agents.
- Can I call Cortex from SQL only?
- Docs list SQL functions and Python interfaces; Analyst and Agents also expose REST APIs documented separately.
- How are models selected?
- Snowflake documents multiple vendor models with regional availability and model-level RBAC; some functions support cross-region inference when a model is unavailable locally.
Related
Related
3 Indexed items
Fireworks AI
Fireworks AI documents a REST platform at docs.fireworks.ai where developers call language, image, and embedding models with Bearer API keys from the dashboard or `firectl api-key create`. Models use globally unique IDs such as `accounts/<account>/models/<model-id>` and can be served via serverless inference for popular open weights (for example Llama 3.1 70B listed on fireworks.ai/models) or private dedicated GPU deployments for custom base models and LoRA addons. Official docs distinguish serverless per-token billing with best-effort uptime from dedicated deployments billed per GPU-second with private capacity, and state that prompts and generated outputs are not logged except for documented exceptions such as the FireFunction model or opt-in advanced features.
Langfuse
Langfuse is an open-source product for LLM application observability: it ingests traces and spans from your stack, supports datasets and prompt/version workflows, and offers optional Langfuse Cloud or self-hosted deployment. It integrates with popular Python/JS SDKs and frameworks that emit OpenTelemetry-compatible telemetry, so teams can debug agent loops, compare prompt iterations, and monitor production quality metrics without building a custom analytics pipeline from scratch.
LangSmith
LangSmith is LangChain's hosted and self-hostable platform for tracing, monitoring, and improving LLM applications. Official documentation at docs.langchain.com describes instrumenting apps via environment variables, framework integrations (OpenAI, Anthropic, CrewAI, Vercel AI SDK, Pydantic AI, and others listed on the integrations page), or the LangSmith SDK so teams can inspect multi-step runs, compare prompt versions, build datasets, run offline and online evaluations, configure automations, and collect feedback queues—without assembling bespoke analytics for agent loops.