With the September 2026 release we are continuing to advance SAP HANA Cloud as an intelligent, AI-ready, and enterprise-scale data platform. This quarter introduces innovations that further enable agentic AI scenarios, bring AI closer to data modeling and development, strengthen business continuity, and simplify the management and optimization of enterprise workloads.
From SAP Agent Memory Service and enhanced data discovery and retrieval capabilities, to AI-assisted Calculation View modeling, Knowledge Graph Engine innovations, and new performance and administration enhancements, this release makes it easier to build intelligent applications and work with enterprise data.
At the same time, new capabilities for multi-region disaster recovery, multitenancy, scalability, and SAP HANA Deployment Infrastructure (HDI) continue to strengthen SAP HANA Cloud as a secure and resilient foundation for business-critical workloads.
Whether you’re building agentic AI applications, simplifying data modeling and development, or optimizing and protecting your enterprise data landscape, SAP HANA Cloud continues to provide the foundation for intelligent, scalable, and resilient data management.
Let’s take a look at some of the key innovations introduced with the September 2026 release.
Agentic AI, Multi-Model & Knowledge Graph Innovations
Agent Memory Service for SAP HANA Cloud
AI agents that start every session from scratch cannot accumulate knowledge, recall past decisions, or carry user context forward. In QRC3 2026, we address this with the Agent Memory Service powered by SAP HANA Cloud: a managed API service built on SAP CAP that gives agents durable long-term memory and conversation history, backed by the SAP HANA Cloud Vector Engine.
Memory entries are scoped by agentID and invokerID, keeping each agent-user pair isolated. The retrieval of past conversations and messages is semantic. Rather than flooding every prompt with an agent's full history, the /v1/memories/search endpoint uses SAP HANA Cloud's VECTOR_EMBEDDING and COSINE_SIMILARITY functions to surface only what is relevant to the current turn, keeping prompts lean and inference costs low. Conversation messages are grouped via /v1/messages for replay and audit. Users retain full oversight, as memory entries can be inspected, edited, or deleted at any time.
Agents can now build on accumulated context across sessions, deliver progressively better outcomes, and do so entirely within the SAP HANA Cloud ecosystem.
AI_CORE_REQUEST — Call any AI Core Model Directly from SQL
Integrating large language models into data workflows has traditionally required application-layer middleware, handling external API calls, separate credential management, and data movement out of and back into the database. In QRC3 2026, AI_CORE_REQUEST removes that overhead entirely.
The new SQL function calls any AI Foundation Model registered in SAP AI Core directly from a SELECT statement, accepting a model name, a messages array, and optional inference parameters, and returning the model response as a structured result. AI enrichment, classification, and summarization become inline SQL operations with no external calls from the application tier. Binding AI_CORE_REQUEST to a generated column takes this further. Every insert or update can automatically trigger model inference, keeping AI-derived attributes continuously synchronized with the underlying data with zero additional application code required.
The result is AI-powered data pipelines that are architecturally simpler, easier to secure, and fully governed within the database.
GeoSPARQL Support in SAP HANA Cloud Knowledge Graph Engine
Spatial data – locations, boundaries, distances, routes – is core to logistics, asset management, environmental compliance, and supply chain operations. Until now, combining spatial reasoning with semantic graph queries required two separate engines and architectuarl approaches as well as two separate query languages. In QRC3 2026, the SAP HANA Cloud Knowledge Graph Engine adds GeoSPARQL support, closing that gap.
Geometry objects like points, line strings, polygons, can now be stored natively in RDF graphs using standard WKT literals, fully aligned with the OGC GeoSPARQL specification. Spatial functions including intersects, within, contains, distance, and overlaps work directly on graph data and compose naturally with semantic graph traversal in a single SPARQL query.
The implementation draws on SAP HANA Cloud's proven spatial engine, so geospatial predicates benefit from the same optimized indexing already used for SQL-based spatial workloads.
Data Exploration Tools — Expanded AI Model Support and Calculation View Discovery
The Data Exploration Tools in SAP HANA Cloud Central received two meaningful updates in QRC3 2026, making natural language database querying more flexible and more useful for analytics workloads.
First, the AI_OBJECT_RETRIEVAL and AI_DATA_RETRIEVAL procedures now support multiple large language models. The MODEL_AND_VERSION parameter accepts both gpt-4.1 (the default) and anthropic–claude-4.6-sonnet, and model selection is exposed directly in the SAP HANA Cloud Central UI when configuring a tool pair. Users choose their preferred model at setup, and that choice is passed to the underlying retrieval procedure at runtime, with no changes required to the SQL procedures themselves.
Second, Data Exploration Tools can now include Calculation Views alongside tables, views, and functions. Calculation Views are registered in the Knowledge Graph and RAG table, enabling the AI retrieval layer to discover them, understand their structure, and explain their purpose in response to natural language queries. This brings SAP HANA Cloud's analytic layer like cube models, dimension views, star joins into reach of the same conversational interface already available for relational objects. CATALOG READ privilege is recommended for the dedicated database user to improve query performance, though the tool operates without it.
AI-Assisted and Enhanced Calculation View Modeling
With Q3 2026, SAP HANA Cloud introduces a range of enhancements that make working with calculation views more intelligent and efficient. New AI-assisted capabilities help developers generate, explain, and optimize expressions using natural-language instructions, while calculation view metadata can now be published to a knowledge graph for AI-assisted discovery and retrieval. Additional modeling enhancements include Debug Mode to identify where specific records are lost, column impact analysis to understand how changes affect a model, dimension labels for clearer identification in analytics tools, and streamlined data preview and search experiences. Together, these updates simplify calculation view development, troubleshooting, and consumption across analytical and AI-driven scenarios.
To learn more please check this blogpost by Jan Zwickel.
Data Integration
SQL on Files: Apache Iceberg format version 3 support
Since QRC 3/2025, SAP HANA Cloud has enabled direct read-only access to Apache Iceberg tables in object storage, bringing real-time analytics to Iceberg lakehouses without moving data. That capability has continued to evolve, and with QRC 3/2026 it now extends to Apache Iceberg format version 3, keeping SAP HANA Cloud aligned with the latest open table format standard as the Iceberg ecosystem advances.
Format version 3 introduces binary deletion vectors, a more compact way to represent row-level deletes than the position delete files used in version 2. SAP HANA Cloud now reads format version 3 tables that adopt deletion vectors, preserving read correctness on these tables. Note that only the deletion vector capability of format version 3 is supported. Other version 3 features, including variant, geometry, geography, nanosecond-precision timestamps, column default values, and row-level lineage, are not supported. This selective adoption keeps behavior predictable while the broader ecosystem matures around the newer version 3 types.
Performance, Scalability & Administration
Analytics Advisor
Let’s start with an innovation designed to make performance optimization easier: the new Analytics Advisor.
For analytical workloads, identifying which queries are responsible for performance issues and determining the right optimization can require significant expertise and manual analysis.
The Analytics Advisor helps simplify this process by analyzing multidimensional query performance and identifying slow-running OLAP queries.
Partition Advisor
Staying with performance optimization, we’re also enhancing the Partition Advisor.
Until now, partitioning recommendations have primarily been based on the number of records within a partition. But row count alone doesn’t always provide the complete picture of how large a partition actually is.
With this release, the Partition Advisor can also take the actual memory size of partitions into account when generating recommendations.
This provides more accurate guidance and helps customers make better-informed partitioning decisions based on the real resource footprint of their data.
Support for additional 3 TB in the Memory Performance Class on AWS
A new intermediate memory performance instance class is now available on AWS, featuring 3 TB of memory and 185 vCPUs. This addition fills an important gap between the existing 1.8 TB—120 vCPU and 6 TB—440 vCPU instance sizes, providing a meaningful mid-tier option for customers whose workload requirements exceed the capabilities of smaller configurations but do not justify scaling to the largest available tier.
The new instance helps right-size infrastructure more effectively by avoiding the need to over-provision resources when additional memory capacity is required. The 3 TB/185 vCPU instance introduces a more balanced alternative, enabling better alignment between workload demands and infrastructure sizing.
With a 1:16 memory-to-vCPU ratio, the instance is particularly well suited for SAP HANA Cloud workloads that require a strong balance of compute and memory resources. It provides additional CPU capacity for compute-intensive operations while maintaining a mid-range memory footprint making it an attractive option where scalability and cost efficiency are key considerations.
More details about the available options can be found SAP HANA Database Size.
Business Continuity & Database Tenant Evolution
Supporting Multi-Region Disaster Recovery for Native Multitenancy
A regional outage shouldn't put your customers' data out of reach. With SAP HANA Cloud QRC3 we introduce a new Multi-Region Disaster Recovery capability for SAP HANA Cloud Multitenancy instances. It protects the database tenants of an SAP HANA database instance by replicating them, together with the database, to a secondary region.
The setup pairs two subaccounts in a BTP region pair, each with a database instance and a hana-multitenancy service instance. While the primary region is active, the secondary holds a passive replica that becomes operational only on failover. While replication is active, tenant data and metadata are continuously replicated asynchronously, meaning that the secondary handles no workload.
On failover, the secondary becomes the primay source. All tenants fail over together, and tenant metadata — IDs, access information, lifecycle changes, container references, and labels — is reconstructed so you can manage tenants via the Tenant Management Service REST API.
More information can be found in the SAP HANA Cloud online help.
Migrating Schemas and HDI Containers to SAP HANA Cloud Database Tenants
Already running your multi-tenant application on schemas or HDI containers? You can now transform them in place into full-fledged SAP HANA Cloud database tenants — and unlock tenant-level encryption, recovery, tenant move or copy, and workload isolation.
The migration is driven through the Tenant Management Service API and transforms an existing schema or HDI container into a database tenant within the same database instance. It runs asynchronously and is non-destructive: your original source container stays intact throughout.
Because it's non-destructive, migration is also low-risk. You can monitor progress until the tenant is ready, and if anything goes wrong, a rollback restores the original state. Once migration succeeds, you simply bind your application to the new database tenant and run it.
You can find more information in the SAP HANA Cloud online help.
Developer & SQL Experience
SQL Console
Query Data Visualization
The SQL Console now allows users to visualize query results as charts directly within the console, eliminating the need to export data or switch to a separate tool. After executing a query, users can seamlessly switch between the standard tabular view and a chart view. This enables quick visual analysis without leaving their workflow.
Supported chart types include Bar, Line, Pie, Donut, and Scatter charts. Users can configure chart axes using the columns returned by their query, providing flexibility in how data is represented and analyzed. This makes it easier to spot trends, compare values, and derive insights from query results at a glance.
Visual Indicator for Statement Results
When running multiple queries in the SQL Console, it can be easy to lose track of which result belongs to which statement. A new SQL toggle button in the results table header helps address this by displaying the executed SQL statement above its associated results tab. This makes the connection between the query and its output immediately clear.
Additionally, when enabled through user settings, navigating between results tabs automatically highlights the corresponding SQL statement in the editor, making it even easier to trace results back to their source without manual searching.
Fixed-Width Font Option
Users can now switch the SQL results view to a fixed-width (monospace) font, providing improved alignment and readability for query output. This is particularly useful when working with formatted text output or structured data, where consistent character spacing helps preserve structure and visual clarity.
Fixed-Width Font Enabled:
Fixed-Width Font Disabled:
The font option can be toggled to suit individual preferences, allowing users to work in the format that best fits their workflow. By improving the presentation of query results, this enhancement makes it easier to review output, compare values, and interpret formatted data directly within the SQL Console.
Expanded Availability on SAP Sovereign Cloud
As part of our continued investment in SAP Sovereign Cloud, we’re expanding the availability of SAP HANA Cloud to meet growing customer demand for digital sovereignty.
In the Q3 2026 timeframe, we delivered two key updates:
First, SAP HANA Cloud is now available in two new SAP Sovereign Cloud regions: India, deployed on AWS, and Europe/Germany, deployed on SAP Cloud Infrastructure. The Europe/Germany availability is particularly significant for regulated European customers. It is an SAP Sovereign Cloud deployment where SAP delivers sovereignty across the full stack on SAP-owned hardware located in Germany.
Second, SAP HANA Cloud on SAP NS2 has received FedRAMP certification. This makes the offering ready for adoption by U.S. federal agencies for their regulated workloads.
Together, these updates allow more highly regulated organizations to adopt SAP HANA Cloud while meeting stringent compliance and regulatory requirements. Learn more in this
Thanks for taking the time to explore the What’s New in SAP HANA Cloud on Q3 2026 innovations!
If you’d like to go deeper into the technical details of the September 2026 release, we recommend visiting the What’s New Viewer in the technical documentation. It provides a comprehensive overview of the full release scope. To stay informed about upcoming innovations, announcements, and best practices, follow the SAP HANA Cloud tag.
You can also find our latest blog posts by searching for the #whatsnewinsaphanacloud hashtag.
If you missed earlier What’s New webinars, you’ll find recordings of past sessions and upcoming events in our YouTube playlist. Our What’s New webinar covering the Q3 2026 innovations will be posted on this playlist very soon.
Do you have questions about SAP HANA Cloud or want to share your thoughts on the new features?
Join the conversation in the SAP HANA Cloud Community Q&A or leave a comment below.
We’d love to hear from you!
Cheers!
