How do I monitor AI and Cortex spend in SELECT?

SELECT provides comprehensive monitoring for Snowflake AI and Cortex costs, including Cortex Functions (LLMs and embedding models), Cortex Search, Cortex Analyst, Document AI, and more. Here's how to track and optimize your AI spend.

View AI spend in Cost Explorer

Navigate to Cost Explorer and group by Service. AI-related costs will appear as a distinct service category, allowing you to see how much of your total Snowflake spend is attributable to AI services.

You can further break down AI spend by grouping by Resource Type to see costs for individual services such as:

- Cortex Functions — LLM inference and embedding calls (e.g. Complete, Embed)

- Cortex Search — text embedding and search indexing

- Cortex Search Serving — serving search queries from Cortex Search indexes

- Cortex Analyst — AI-powered SQL generation, attributed by user

- Cortex Agent — agentic workflows

- Cortex Code — AI code assistance (Snowsight and CLI), attributed by user

- Cortex Fine Tuning — custom model training, attributed by model

- Cortex Provisioned Throughput — reserved capacity for models, attributed by model

- Document AI — document processing

- Snowflake Intelligence — native Snowflake intelligence features

Track Cortex Code Costs

You can track the costs of Cortex Code in the cost explorer with the following filters applied:

1. Service = 'AI Services'

2. Resource Type = 'Cortex Code'

Each 'Resource' entry in the table represents a user in your account who is incurring Cortex Code costs.

Note that Cortex Code was in a free-to-use period until April 1st 2026.

Track Cortex Function costs by workload

1. Go to Workloads → All Workloads.

2. Click the blue + Add Column button in the table and add Cortex Function Spend.

3. Sort the column in descending order to identify your highest-spending workloads.

4. In the filter bar, apply a filter for Cortex Function Spend is not null to show only workloads with Cortex spend.

5. Optionally, save this as a Saved View to quickly return to this analysis.

You can also add Cortex Tokens as a column to understand token usage alongside cost.

Monitor AI spend by user

On the All Workloads page, click the three dots on the chart and Group by → User to see which users are driving the most AI spend. You can also add User as a column in the table and group by it for a tabular breakdown.

Set up spend alerts

You can create a Spend Monitor to receive alerts when AI costs spike:

1. From Cost Explorer, filter by the AI Services service type to isolate AI spend.

2. Click Create Monitor to set up an alert based on your filtered view.

3. Configure your threshold and notification preferences.

This ensures you're notified proactively if AI spend increases unexpectedly.

AI and Cortex columns

On the All Workloads page and other query-based workload pages, use the blue + Add Column button to add any of the following columns for AI cost visibility:

- Cortex Function Spend — total spend from Cortex Function calls (LLM inference, embeddings)

- Cortex Tokens — total tokens consumed by Cortex Function calls

- Cortex Function Credits — credits consumed by Cortex Functions

- Document AI Spend — spend from Document AI processing

- Document AI Credits — credits consumed by Document AI

These columns can be sorted, filtered, and used to create Saved Views for ongoing monitoring.