All release notes

The dynamic tables feature formerly known as the immutability constraint is now called the frozen region of a dynamic table. As part of this rename, the IMMUTABLE WHERE clause is also renamed to FROZEN WHERE.

...

Read more

Snowflake Cortex AI_COMPLETE now supports Gemini 3.5 Flash for multimodal video and audio analysis, enabling organizations to extract brand, sentiment, behavioral, and content insights from rich media directly in Snowflake. With frontier multimodal intelligence and Flash-level speed, teams can analyze video and audio content for marketing, content intelligence, compliance, and media analytics workflows without moving data outside of Snowflake.

...

Read more

You can now generate a Sensitive Data Access report to view which users accessed sensitive data in your account during a specified lookback period. The report creates a view that lists user-table-role combinations identified in query and access history for classified tables.

...

Read more

With this preview release, you can use the Governance & security » Data protection policies area in Snowsight to create data protection policies from the Dashboard, list and manage policies on the Policies tab, and view protected objects on the Objects with policies tab.

...

Read more

The following new features and enhancements are now available in Snowflake Data Clean Rooms:

• DCR Agent (public preview): You can now use a conversational Snowflake Intelligence interface to join collaborations and run approved analyses including audience overlap and activation via standard_audience_overlap_v0 and standard_audience_overlap_activation_v0 templates. Learn more.
• Additional Cortex Code Skills: Users can now use cortex code with data clean room skills that support dropping/leaving a collaboration, linking/unlinking data offerings and requesting new templates to be added with approval flows.

Dynamic tables now support a new refresh mode: REFRESH_MODE = ADAPTIVE.

...

Read more

When standard refresh modes don't support your transformation pattern, CUSTOM_INCREMENTAL refresh mode lets you write MERGE or INSERT logic that Snowflake executes on each refresh. You define exactly how changes reach the dynamic table while Snowflake handles scheduling, retries, and transactional guarantees. This mode supports patterns such as stream-static joins, soft-deletes, stateful aggregation, and direct migration from streams and tasks.

...

Read more

With this release, write support for Snowflake-managed Apache Iceberg™ tables by using an external query engine through Snowflake Horizon Catalog is now generally available. You can read and write to Snowflake-managed Iceberg v2 and v3 tables from any external query engine that supports the open Iceberg REST protocol, such as Apache Spark™, by using a single Horizon Catalog endpoint and your existing Snowflake users, roles, policies, and authentication.

...

Read more

• Virtual columns (General availability)

Extraction scores for the Snowflake Cortex AI_EXTRACT function are now generally available. Key capabilities include:

• Confidence scores for extracted values: Pass scores => TRUE in an AI_EXTRACT call to get back a scoring object alongside the standard response object, with a score between 0 and 1 indicating the model's certainty about each extracted value.
• Support for entities, lists, and tables: Per-field scores are returned for entity extraction, and aggregate scores are returned for list and table extraction.
• Support for fine-tuned arctic-extract models: Extraction scores work with fine-tuned arctic-extract models in addition to the default model.

...

Read more

The following updates are now available in Snowflake Data Clean Rooms:

• General performance improvements and bug fixes.
• Updates to private preview features.

The dynamic tables documentation now includes 30 new and updated pages covering the full lifecycle of building, optimizing, monitoring, and troubleshooting dynamic table pipelines.

...

Read more

Attaching a storage lifecycle policy to a dynamic table is now generally available. You can attach a storage lifecycle policy directly to a dynamic table to delete or archive rows that match a predicate, on a schedule that runs independently of dynamic table refresh.

...

Read more

Cortex AI Function Studio is now available in public preview, enabling organizations to easily create, evaluate, and optimize production-ready Cortex AI Functions for unstructured and multimodal data workflows. AI Function Studio automates prompt engineering, model selection, benchmarking, and optimization, helping teams accelerate AI application development while improving quality and reducing experimentation overhead.

...

Read more

With this preview release, you can build, validate, and register your own custom container images for use with Snowflake Notebooks and ML Jobs on Container Runtime. Custom runtime images let you include specific packages, meet compliance and security requirements, integrate with existing CI/CD pipelines, and ensure reproducibility across environments.

...

Read more

The following improvements are now available for dbt Projects on Snowflake:

• Enhanced dbt DAG with column-level lineage: The dbt DAG visualization is now significantly more capable across all three surfaces (Workspaces, dbt project object details, and Query History). Each model node displays its columns, powered by the Snowflake Horizon Catalog as the single source of truth. Selecting a column highlights all upstream and downstream models in the DAG that use that column. Additional improvements include an in-DAG search bar to find models by name, expandable upstream and downstream depth controls, a unified side panel showing row count, column count, compiled SQL, and runtime information, and an increased model display limit (up from 100 to 300). For more information, see dbt projects on snowflake manage.

...

Read more

Batch Cortex Search is now generally available. The CORTEX_SEARCH_BATCH table function lets you run millions of fuzzy-match queries against a Cortex Search Service in a single SQL statement, with no infrastructure to manage.

...

Read more

• New DEFAULT_PYTHON_ARTIFACT_REPOSITORY parameter
• New PROCEDURE SCOPED TEMP TABLE syntax for stored procedures

With this release, the GEN_ARM_G1 instance family is generally available for Snowpark Container Services on AWS.

...

Read more

Cortex AI Guardrails now provide run-time protection against prompt injection and jailbreak attacks for Snowflake Intelligence and Cortex Agents, in addition to Cortex Code.

...

Read more

The following updates are now available in Snowflake Data Clean Rooms:

• General performance improvements and bug fixes.
• Updates to private preview features.

The Snowflake Cortex AI_CLASSIFY function now accepts documents as input, enabling organizations to automatically categorize document files like PDF, DOCX, CSV, and more using simple SQL. This allows teams to easily streamline document intake and intelligently route content, such as contracts, invoices, and various reports, into the right downstream workflows.

...

Read more

You can now generate a Sensitive Data Entitlement report to view who can access sensitive data in your account. The report creates a view that lists users who have access control roles with privileges on tables containing sensitive data.

...

Read more

You can now grant per-function privileges for Snowflake Cortex AI Functions using USE AI FUNCTION . This gives account administrators finer-grained control over which AI functions individual roles can call, instead of granting blanket access to all AI functions with the existing USE AI FUNCTIONS privilege.

...

Read more

The ORGANIZATION_USAGE schema now includes two new premium views that are available in the organization account. These views provide visibility into secrets and Tri-Secret Secure customer-managed key history across all accounts in your organization.

...

Read more

The Snowflake Python connector now supports a JWT-based authentication method for AWS Workload Identity Federation (WIF). The new method uses the AWS GetWebIdentityToken API to produce a standards-based JWT instead of the existing GetCallerIdentity-based pre-signed request method. The existing GetCallerIdentity-based method remains supported, but Snowflake recommends upgrading to the JWT-based method.

...

Read more

You can now configure a Cortex Search Service to automatically suspend its serving compute after a period of query inactivity and resume when it receives a new query. This helps reduce serving costs on idle services without requiring manual intervention.

...

Read more

• Openflow BYOC: per-runtime cost attribution in Account Usage views

You can now specify tuple values for ingress private endpoint identifiers in network rules. This feature allows PrivateLink identifiers such as AWSVPCEID, AZURELINKID, and GCPPSCID to use tuple syntax that combines endpoint identifiers with IP addresses and/or CIDR ranges.

...

Read more

The following new features and enhancements are now available in Snowflake Data Clean Rooms:

• Activity history in collaborations: You can now use VIEW_ACTIVITY_HISTORY to audit analysis activity in a collaboration. Each collaborator sees only the activity relevant to their account and the resources they contribute. This feature is available only in collaborations created after this update. For details, see Monitor analysis activity in collaborations.
• Collaborators can clean up activation segment records: For collaborations created after this update, the CLEANROOM_PUBLIC_ROLE has permission to delete data from the ACTIVATION.SEGMENT_RECORDS table, allowing collaborators to clean up segment data after activation jobs complete.
• General performance improvements and bug fixes.

Support for version 3 (v3) of the Apache Iceberg™ table specification is now generally available.

...

Read more

Listing share restrictions are now generally available. Providers can enable share restrictions on private external listings to prevent consumers in lower editions or non-HIPAA accounts from accessing their data.

...

Read more

Organization administrators can now change the Snowflake edition of an account in the organization by using ALTER ACCOUNT SET EDITION = { 'STANDARD' | 'ENTERPRISE' | 'BUSINESS_CRITICAL' }. Run the command from an account other than the account whose edition you're changing.

...

Read more

The ORGANIZATION_USAGE schema now includes three new premium views that are available in the organization account. These views provide visibility into network policies and network rules across all accounts in your organization.

...

Read more

When defining a semantic view, you can now define a fact or dimension as a filter by specifying LABELS = (FILTER) in the create semantic view command. A filter is a fact or dimension that resolves to a BOOLEAN value and can be used as a condition in the WHERE clause of a query.

...

Read more

You can now assign tags to semantic views and the attributes within them, including logical tables, dimensions, facts, and metrics. You can use tags to track sensitive data, manage governance policies, and organize your semantic view objects.

...

Read more

With this release, four new instance families are generally available for Snowpark Container Services:

• GEN_X64_G2 (AWS, Azure): Second-generation general-purpose compute, replacing CPU_X64 as the recommended instance family for all new workloads.
• MEM_X64_G2 (AWS, Azure): Second-generation high-memory compute, replacing HIGHMEM_X64 as the recommended instance family for memory-intensive workloads.
• GPU_L40S (AWS): NVIDIA L40S GPU (Ada Lovelace) with 48 GB VRAM per GPU, scaling up to 8 GPUs (384 GB total).

...

Read more

• Apply data quality checks by group (General availability)
• New columns in DMF results and references views

The SAP® BDC Zerocopy Connector is now generally available, enabling zero-copy data sharing between SAP® Business Data Cloud and Snowflake without moving or duplicating data.

...

Read more

Snowflake Cortex AI_COMPLETE now enables advanced multimodal analysis of video and audio content in public preview, allowing organizations to extract brand, sentiment, and behavioral insights from rich media directly in Snowflake. This unlocks powerful new capabilities for marketing, content intelligence, and compliance workflows, without moving data outside of Snowflake.

...

Read more

The Snowflake Cortex AI_PARSE_DOCUMENT function has launched significant quality improvements for OCR mode, building on the strong OCR performance already available in AI_PARSE_DOCUMENT and delivering even higher accuracy and faster processing for enterprise-scale document workflows that require efficient, high-quality text extraction.

...

Read more

You can now use CREATE OR ALTER SEMANTIC VIEW to create a semantic view if it doesn't exist, or alter an existing semantic view to match the definition in the statement. This provides a declarative way to manage semantic view definitions that's safe to re-run without having to replace the view and re-grant privileges.

...

Read more

With this release, autoscaling policies for Snowpark Container Services are now generally available. Autoscaling policies extend the existing CPU-based autoscaling by letting you define metric-based rules that control how Snowflake scales service instances up, scales them down, or suspends them automatically. Policies support Snowflake-provided platform metrics and custom metrics that your service emits, giving you fine-grained control over scaling behavior based on CPU utilization, memory usage, ingress connection rate, and more.

...

Read more

The ORGANIZATION_USAGE schema now includes a new premium view: outbound privatelink endpoints. The view is available in the organization account. You can use this view to monitor outbound private connectivity endpoints across all accounts in your organization.

...

Read more

You can now authenticate SCIM API requests with the following types of tokens:

• Programmatic access tokens (PATs)
• OAuth tokens obtained from a third-party identity provider (External OAuth)

...

Read more

The Snowflake Cortex AI_PARSE_DOCUMENT function now supports processing documents of up to 2,000 pages in both LAYOUT and OCR modes, enabling organizations to efficiently handle large, multi-page documents such as healthcare records, insurance claims, regulatory filings, and technical manuals in a single request with reduced pre-processing and orchestration.

...

Read more

High-fidelity image extraction with the Snowflake Cortex AI_PARSE_DOCUMENT function using LAYOUT mode is now generally available, enabling organizations to unlock visual data from documents and build richer, multimodal document processing workflows. By extracting images alongside text and layout, teams can incorporate visual context from documents such as healthcare records, financial reports, and regulatory filings, without additional tooling or pipelines.

...

Read more

When you work with identifiers inside a catalog-linked database that uses CASE_INSENSITIVE mode (the default for case-insensitive catalogs like AWS Glue and Unity Catalog), you no longer need to surround schema, table, and column names in double quotes for DDL commands. Identifiers outside the catalog-linked database continue to follow standard Snowflake identifier resolution rules.

...

Read more

The following new features and enhancements are now available in Snowflake Data Clean Rooms:

• Creation timestamps for registry objects: VIEW_REGISTERED_DATA_OFFERINGS, VIEW_REGISTERED_TEMPLATES, and VIEW_REGISTERED_CODE_SPECS now also return the CREATED_ON timestamp, so you can see when the object was first created.
• Data Clean Rooms UI in Snowsight (public preview). You can now manage Data Clean Rooms collaborations directly in Snowsight. The UI lets you browse, create, join, edit collaborations, and run analysis. The UI also integrates with Cortex Code for natural-language assistance. Learn more.

With this preview release, you can create tags in Snowsight by opening Governance & security » Tags & policies, selecting the Tags tab, and selecting + Create tag.

...

Read more

The following Trust Center capabilities are now generally available:

• Detection findings: A new finding type to surface anomalous and potentially suspicious events. For more information, see trust center detections.
• Event-driven scanners: scanners that generate detection findings in response to relevant events, such as changes to sensitive parameters or sign-ins from unusual IP addresses, in addition to the existing schedule-based scanners. For more information, see trust center scanners event driven.
• New scanners in the Threat Intelligence scanner package that report detection findings:

...

Read more