SingleStore, the real-time data platform, announced a suite of powerful updates on June 17, 2025, during its inaugural Upsert event, aimed at enhancing enterprise AI deployment, developer experience, and data integration. The enhancements, centered around performance, AI capabilities, and seamless workflows, solidify SingleStore’s position as a leading foundation for intelligent applications.
SingleStore Flow enhances data ingestion from Snowflake, Postgres, and Oracle.
New Apache Iceberg speed layer boosts data lakehouse performance.
Aura Container Service now supports serverless Cloud Functions for AI agents.
Performance upgrades include JSON multi-value indexing and query optimization.
Developer tools feature GitHub integration, SQrL AI assistant, and SQL Editor autocomplete.
Upsert event series to showcase new features every few months.
SingleStore’s updates focus on enabling faster, scalable AI deployments. “When you’re creating with AI, it’s vital to choose the right data foundation to build upon,” said Raj Verma, CEO of SingleStore. “As the world’s fastest database, and the only one to natively offer single-shot retrieval of all types of data, SingleStore has always been the ultimate foundational layer for AI.” The platform’s ability to handle transactional, analytical, and vector data in real time supports 95%+ query response times under 20 milliseconds for AI workloads.
The SingleConnect ecosystem, powered by SingleStore Flow (via the BryteFlow acquisition), simplifies data ingestion from Snowflake, Postgres, SQL Server, Oracle, and MySQL. “In SQL, an upsert either updates an existing row or adds a new one,” said Dave Eyler, Vice President of Product Management at SingleStore. “It’s a great analogy for this event, which is to tell customers about new features and updates they can try.” A new speed layer on Apache Iceberg, combined with bi-directional integration, enhances data lakehouse performance, enabling faster analytics on external tables.
SingleStore’s Aura Container Service now hosts Cloud Functions, serverless lambda-style functions for building Data APIs, AI agents, and inference tools. Developer experience improvements include SQrL, an AI-powered query assistant, GitHub integration, and a multi-tab SQL Editor with autocomplete. Notebooks scheduling and pipeline monitoring further streamline workflows, reducing development cycles by up to 30% compared to legacy systems.
New features like multi-value indexing for JSON, automatic query re-optimization, and cross-workspace database branching boost scalability. Database branching, point-in-time recovery, and SmartDR are now available on GCP and AWS, ensuring enterprise-grade resilience. These enhancements deliver 10-100x better performance at one-third the cost of traditional architectures, as validated by customers like Adobe.
SingleStore’s innovations have earned a 2025 TrustRadius Buyer’s Choice Award and a spot in the Gartner Magic Quadrant. The recurring Upsert event series will showcase ongoing advancements, with SingleStore Now 2025 set for October 1 at Nasdaq MarketSite. These updates position SingleStore to meet the growing demand for real-time, AI-driven applications, empowering developers to “feel the flow” in their work.
SingleStore’s latest enhancements deliver a robust, developer-friendly platform for enterprise AI, driving innovation with unmatched speed and simplicity.
SingleStore empowers the world’s leading organizations to build and scale modern applications using the only data platform that allows you to transact, analyze and search data in real time. With streaming data ingestion, support for both transactions and analytics, horizontal scalability and vector and full text search capabilities, SingleStore helps deliver 10-100x better performance at 1/3 the costs compared to legacy architectures. Customers worldwide — including Fortune 500 companies and global data leaders — use SingleStore to power real-time applications and analytics.