Snowflake Releases
Snowflake blog and product updates. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 135 Snowflake releases · Updated
- Snowflake Blog blogobservabilitysnowflakegaengineer ·
Observe by Snowflake Boosts Incident Response with Customer Journey Observability and AI SRE
Observe by Snowflake now enables a customer journey-first approach to incident management, allowing SREs to start investigations from business impact and efficiently drill down to root cause. This methodology, demonstrated by Capital One, leverages a telemetry data lake and context graphs to provide clear impact assessment and streamlined troubleshooting. The platform introduces AI SRE, which operates over the same context graph to accelerate investigations, guiding engineers with evidence and interactive queries up to 10x faster. This integration aims to significantly reduce incident resolution times and improve collaboration across teams.
feature announcement - Snowflake Blog blogaisnowflakearchitect ·
Accelerating M&A Due Diligence and Integration with AI Data Platforms
This article explores how modern data and AI platforms are reshaping mergers and acquisitions (M&A). It explains how leveraging high-quality data and AI accelerates due diligence, improves risk assessment, and streamlines post-deal integration. This shift addresses challenges like siloed data, leading to faster execution and measurable ROI for organizations. The Snowflake AI Data Cloud, in partnership with Accenture, is highlighted as a critical enabler for this platform-led approach to M&A.
announcement - Snowflake Blog blogaisnowflakepreviewengineer ·
Snowflake Cortex AI adds Moonshot Kimi K3 large language model to private preview
Snowflake Cortex AI now offers Moonshot AI's Kimi K3, a 2.8-trillion-parameter open-weight Mixture-of-Experts (MoE) model, in private preview. This model is specifically designed for complex, long-running tasks that require multi-step reasoning and context retention across various inputs and turns. Teams can evaluate Kimi K3 using Cortex AI Functions and Cortex Inference, with future support planned for CoCo, CoWork, and Cortex Agents. It provides an advanced open-weight option for model matching, though users should start fresh sessions and note its proactive nature, as its overall experience still trails top proprietary models.
patch - Snowflake Blog blogaisnowflakepreviewengineer ·
Anthropic Claude Opus 5.5 AI Model Now in Public Preview on Snowflake Cortex AI
Anthropic's Claude Opus 5.5, its most capable Opus model, is now available in public preview on Snowflake Cortex AI. This integration provides Snowflake users with enhanced capabilities for complex coding and knowledge work, along with significant improvements in cost efficiency and safety compared to previous models. It enables teams to leverage advanced AI across Snowflake CoCo, Cortex Agents, AI Functions, and Cortex Inference for development, analysis, and custom applications. All model calls remain within Snowflake's robust security and governance perimeter, ensuring data protection for engineers and architects.
patch - Snowflake Blog blogobservabilitysnowflakepreviewengineer ·
Snowflake introduces AI Agent Observability for LLM application monitoring
Snowflake is introducing Agent Observability to its Observe product, designed to address the unique challenges of monitoring AI agents and LLM applications. This new capability will allow teams to trace interactions, connect agent behavior with cost and quality, and debug performance issues. It aims to cost-effectively retain and analyze high-fidelity agent telemetry at scale, helping identify what drives AI spend and improve business outcomes. The feature will soon be available in private preview.
feature announcement - Snowflake Blog blogdatasnowflakeengineer ·
Snowflake simplifies high-throughput data streaming to Apache Iceberg with Snowpipe Streaming
Snowflake showcases a streamlined method for high-throughput data streaming to Apache Iceberg tables, achieving over 1 million transactions per second (TPS) using Snowpipe Streaming and Snowpark Container Services. This approach simplifies large-scale streaming evaluations by eliminating the need for complex external infrastructure such as Kafka, EKS clusters, and IAM policy setups. Data engineers, SREs, and architects can now rapidly assess high-volume Iceberg streaming workloads within a fully Snowflake-managed environment. A companion GitHub repository provides a demo that allows setup and execution of the end-to-end streaming pipeline in half a day.
announcement - Snowflake Blog blogaisnowflakepreviewengineer ·
Snowflake Cortex AI Integrates Grok 4.6 for Advanced Agentic AI Capabilities
Snowflake Cortex AI now offers Grok 4.6 from SpaceXAI in public preview, enhancing its capabilities for complex, multi-step AI tasks. The frontier model is designed for long-running agents, supporting advanced interactive and visual workloads like topic research and code analysis. This integration enriches Snowflake CoCo for faster development and empowers CoWork to build advanced personal work agents. Developers can also leverage Grok 4.6 for high-accuracy analysis within Cortex AI Functions and to create secure AI applications via Cortex Inference.
feature announcement - Snowflake Blog blogaisnowflakepreviewengineer ·
Snowflake Cortex AI Now Supports Anthropic's Claude Fable 5.1
Snowflake Cortex AI now offers Anthropic's powerful Claude Fable 5.1 in private preview, making it available within Snowflake's secure boundaries from its launch day. As Anthropic's launch partner, Snowflake integrates this model to enhance agentic coding, knowledge work, and scientific discovery capabilities. This integration benefits developers, data engineers, and knowledge workers by accelerating development, enabling advanced personal work agents, and improving data analysis. Users can leverage Fable 5.1 with Snowflake CoCo, CoWork, Cortex AI functions, and Cortex Inference for various enterprise use cases.
feature announcement - Snowflake Blog blogaisnowflakearchitect ·
Snowflake on Why Business Context is Critical for Effective AI Strategies
Snowflake discusses the critical role of "contextual advantage" in developing effective AI growth strategies, asserting that readily available high-performance models are insufficient without deep business understanding. The company highlights how AI decisions lacking crucial context can lead to suboptimal outcomes and erode customer trust, even with accurate factual data. This unique business context, encompassing customer history, operational policies, and past learnings, is presented as an enduring competitive differentiator. Snowflake positions its AI Data Cloud and Horizon Context as foundational tools for building a governed enterprise context layer to enhance AI precision and reliability.
announcement - Snowflake Blog blogaisnowflakeengineer ·
Snowflake shares best practices and tools for its CoCo AI coding agent
Snowflake provides a comprehensive guide on leveraging its CoCo AI coding agent for data engineering. The article outlines best practices for using AI agents to achieve reproducible, high-quality results, distinguishing professional data engineers. It highlights CoCo's unique advantages for Snowflake users, including in-platform inference, native integration, and extensive built-in skills. Data engineers can access CoCo via CLI, a VS Code-based desktop IDE, or directly within Snowsight, and build consistent workflows using open-standard skills and plugins.
announcement - Snowflake Blog blogaisnowflakeengineer ·
Snowflake outlines a 5-stage maturity model for autonomous data engineering
Snowflake details a five-stage maturity model for "Autonomous Data Engineering," a strategic shift enabling AI agents to build, maintain, and optimize data pipelines. This approach addresses the growing complexity of data engineering driven by AI, allowing human engineers to focus on defining business challenges, guiding agents, and ensuring governance. The model progresses from manual pipeline management to fully autonomous operations, emphasizing the creation of data products over traditional infrastructure-first methods. Organizations can leverage platforms like Snowflake to incrementally adopt agent-driven workflows and improve efficiency.
announcement - Snowflake Blog blogaisnowflakegaarchitect ·
Snowflake Announces GA of User-Level Quotas for AI and Compute Spend
Snowflake has made user-level quotas generally available, providing automated, individual spending guardrails for compute and AI services. This enables organizations to grant broad access for AI innovation while proactively managing potential runaway costs from individual processes. The feature offers granular daily and monthly limits, reducing financial risk and management burden for FinOps and platform teams. It covers warehouse compute, Cortex AI functions, agents, and Snowflake CoCo, with management options via SQL, Snowsight UI, or Snowflake CoCo.
feature announcement - Snowflake Blog blogcompliancesnowflakegaarchitect ·
Snowflake Integrates with STACKIT to Boost EU Data Sovereignty
Snowflake has integrated with STACKIT, Schwarz Group's cloud, to enhance data sovereignty options for customers in the EU. This collaboration combines Snowflake's data and AI capabilities with STACKIT's EU-based hosting, encryption, and open formats like Apache Iceberg. The initiative addresses strategic priorities for regulated industries by ensuring transparent data access, operational independence, and comprehensive compliance coverage. This integration offers EU enterprises and public institutions greater control over their data, aligning with strict regulatory requirements for digital sovereignty.
feature - Snowflake Blog blogdatasnowflakegaengineer ·
Snowpipe Streaming Elastic Channels Now Generally Available on Snowflake
Snowflake has announced the general availability of Snowpipe Streaming's Elastic Channels, simplifying direct data ingestion from applications, devices, and services into Snowflake or Snowflake-managed Iceberg tables. This feature reduces operational complexity, shortens latency, and lowers total cost of ownership by eliminating the need for separate ingestion systems. It supports high-throughput workloads, handling up to 20GB/s per table with data queryable in as little as 5 seconds. Elastic Channels are available across all AWS, GCP, and Azure commercial regions without additional cost, though standard Snowpipe Streaming ingestion fees apply.
feature - Snowflake Blog blogdatasnowflakegaarchitect ·
Snowflake Data Connectivity Proxy for Private Data Now Generally Available
Snowflake has introduced the Data Connectivity Proxy (DCP) for Openflow, a new feature designed to securely connect Snowflake to private data sources across hybrid environments. DCP uses a lightweight, outbound-only agent that establishes an encrypted tunnel to Snowflake on port 443, eliminating the need for complex inbound firewall configurations, VPNs, or PrivateLink for each topology. This simplifies data integration for enterprises with strict security requirements, such as those in finance, healthcare, and manufacturing. The Data Connectivity Proxy is now generally available for Standard, Enterprise, and Business-Critical Snowflake editions on AWS, Azure, and Google Cloud.
feature announcement - Snowflake Blog blogaisnowflakearchitect ·
Snowflake's Blueprint for Effective Enterprise AI Agent Management
Snowflake shares a strategic blueprint for enterprises to effectively deploy and manage AI agents, shifting the focus from prompt engineering to treating agents as organizational roles. This framework addresses key challenges like governance, human oversight, and data readiness, aiming to move AI agent initiatives beyond pilots into production. It outlines three deliberate steps: strategic hiring, comprehensive data "handbooks," and active management, leading to the emergence of specialized AI manager roles within organizations. This approach emphasizes human accountability and a structured organizational shift for successful agent adoption.
announcement - Snowflake Blog blogdatasnowflakegaarchitectmedia ·
Snowflake Snowpipe Streaming Elastic Channels Now Generally Available
Snowflake has announced the general availability of Elastic Channels for Snowpipe Streaming. This feature simplifies direct data ingestion from applications, devices, and services into Snowflake or Apache Iceberg tables, eliminating the need for complex intermediate message buses or staging pipelines. It significantly reduces operational overhead, latency, and TCO, supporting high-volume workloads up to 20 GB/s with data queryable in as little as five seconds. Designed for demanding use cases like IoT, industrial operations, and digital applications, it offers durable acknowledgments and at-least-once delivery through SDKs or a REST API.
security - Snowflake Blog blogcompliancesnowflakearchitect ·
Snowflake Integrates with STACKIT for EU Data Sovereignty
Snowflake has integrated its Data Cloud with Schwarz Digits’s STACKIT cloud to offer enhanced data sovereignty options for customers in Germany and the EU. This collaboration provides European encryption key systems, open data formats, and data hosting within European data centers, addressing critical regulatory and strategic needs. The integration enables regulated industries to leverage Snowflake’s data and AI capabilities while meeting stringent EU data sovereignty requirements, particularly through STACKIT's European Sovereign Stack (ES³) Standard certification.
announcement - Snowflake Blog blogsnowflakegaarchitectfinance ·
Snowflake's GA Per-User Quotas Enable Granular AI and Compute Spend Governance
Snowflake has made per-user quotas generally available, allowing enterprises to govern individual AI and compute spend. This feature eliminates the tension between fostering innovation and controlling costs, preventing runaway processes from draining budgets. It provides automated, proactive spend guardrails with block enforcement and granular daily or monthly limits. The quotas cover various services, including warehouse compute, Cortex AI Functions, Cortex Agents, and Snowflake CoCo, benefiting FinOps, platform teams, and developers.
feature - Snowflake Blog blogsecuritysnowflakegaarchitect ·
Snowflake Introduces Data Connectivity Proxy for Private Data Access
Snowflake has introduced Data Connectivity Proxy (DCP) for its Openflow data integration service, enabling secure and simplified access to private data sources. This lightweight agent model facilitates connecting to on-premises databases, private Kafka brokers, and cross-cloud environments without requiring inbound firewall rules or complex network configurations. Designed for deny-all-inbound networks, DCP establishes outbound TLS connections, eliminating the need for VPNs, PrivateLinks, or third-party agents. It is generally available on AWS, Azure, and GCP for various Snowflake editions, targeting organizations with hybrid data estates.
security
About Snowflake release tracking on ReleaseBytes
Snowflake's weekly releases, behaviour change bundles and feature announcements are easy to lose track of. ReleaseBytes follows the official Snowflake release channels and summarises each update, flagging behaviour changes and deprecations that need action before they are enforced.
Frequently asked questions
How often are Snowflake release notes updated on ReleaseBytes? ›
Continuously. ReleaseBytes monitors the official Snowflake release channels around the clock and publishes a plain-English summary of each announcement shortly after it lands.
What kinds of Snowflake changes does ReleaseBytes track? ›
New features, enhancements, bug fixes, security advisories, breaking changes, deprecations and end-of-life announcements. Every item is tagged by type so you can filter to just the changes that need action.
How can I get alerts for new Snowflake releases? ›
Set up a free email or Slack alert filtered to Snowflake, subscribe to the weekly digest, or follow the RSS feed. Teams can also install the ReleaseBytes GitHub App or connect via MCP.
Where does the Snowflake release data come from? ›
From the official sources: Snowflake blog and product updates. Every item links back to the original vendor announcement.