Releasesby GlyphData lineage for Cortex Agents
You can now use data lineage to see which data a Cortex Agent can reach through the tools declared in its specification. A semantic view that a Cortex Analyst tool references, and a Cortex Search service that a Cortex Search tool references, both appear upstream of the agent.
Also shipped Sep 2, 2026Snowflake in September 2026 →
Git Folder Serverless is in Beta
You can now use Git Folder Serverless to share serverless compute and a pyproject.toml -managed environment across notebooks and files in a Git folder. See Git Folder Serverless .
AlloyDB real-time BigQuery access adds limit pushdown and runtime projects
The preview feature pushes some LIMIT and OFFSET clauses down to BigQuery to cut network transfer, and optional runtime projects let queries run and bill compute separately from the owning project.
OpenJDK 11 will no longer be supported as of January 14, 2027
OpenJDK 11 will no longer be supported as of January 14, 2027 . Customer-hosted instances must upgrade to OpenJDK 21 .
Gemini Enterprise: Gemini 3.8 Flash available in Global, US, and EU regions
Gemini 3.8 Flash is generally available (GA) in the global , us , and eu regions. For more information, see: Manage features on the web app Data residency for Gemini Enterprise Standard and Plus Editions and Gemini Notebook Enterprise
Gemini 3.8 Flash is generally available
Gemini 3.8 Flash is now generally available (GA) and available for production use. For more information on 3.8 Flash, see the model page .
Deferred tier for autonomous agent scheduling (Preview)
The deferred tier is available in Preview. The deferred tier automatically queues non-latency-sensitive background agent workloads and schedules them to execute during off-peak hours. Key capabilities and benefits include: 50% token discount : Receive a 50% discount on consumed tokens for deferred workloads.
More from Snowflake Cortex AIEvery Snowflake Cortex AI release →
Restricted Session Scope for agents (General availability)
With this release, Restricted Session Scope (RSS) is generally available. An RSS is a privilege ceiling that limits what an agent can do on behalf of a user. It intersects with RBAC and never grants privileges the user doesn't already have through their roles.
Snowflake Native Apps: Code execution tools for Cortex Agents
Providers can now add the code_execution and code_toolset_all sandbox tools to Cortex Agents created by a Snowflake Native App.
Cortex Agents asynchronous API (General availability)
The Cortex Agents asynchronous API is now generally available on AWS and Microsoft Azure.
Snowflake recommends transitioning from Cortex Analyst to Cortex Agents
Snowflake recommends transitioning to Cortex Agents , which supports every Cortex Analyst capability with higher answer quality. Snowflake adds new natural language analytics capabilities to Cortex Agents.
Cortex Extension references in agent skills (General availability)
Referencing a Cortex Extension from an agent skill is now generally available. Set the skill source type to CORTEX_EXTENSION and path to the extension's fully qualified name, and Snowflake expands the reference into the extension's member skills at request time, so you don't list each skill individually.
Cortex Agents Coding Agent (General availability)
The Cortex Agents Coding Agent is now generally available. Add the code_toolset_all tool type to a Cortex Agent, and Snowflake provisions and manages a sandbox backed by the same runtime that powers Snowflake CoCo: bash, file read/write/edit, grep, glob, web search, SQL execution, and skills.
Multimodal AI_SUMMARIZE for automatic theme summarization (Public Preview)
Snowflake Cortex AI_SUMMARIZE is now available in public preview, enabling customers to summarize text and multimodal content, including images and documents, directly in Snowflake.
Version targeting for Cortex Agent and Cortex Analyst evaluations (General availability)
Cortex Agent and Cortex Analyst evaluations now let you control the versioned configuration used for scoring: Version targeting pins the configuration under evaluation, not the agent behavior that gets evaluated: each run invokes the agent again and scores a newly generated trace.