# Spice Cloud v1.9.0: Introducing the Spice Cayenne Data Accelerator

Wyatt Wenzel

DevRel & Ops Leader at Spice AINovember 20, 2025

Spice Cloud & Spice.ai Enterprise 1.9.0 are live!

Our mission at Spice is to make building data-intensive applications and AI systems easier, faster, and more secure. With **v1.9.0**, we're taking a big step forward.

This release introduces **Spice Cayenne**, our new premier data accelerator based on [**Vortex**](https://github.com/vortex-data/vortex), upgrades to **DataFusion v50** and **DuckDB v1.4.2**, new **HTTP Data Connector** support for querying API endpoints as tables, and many more improvements across performance, scalability, and developer experience.

Spice Cloud customers will automatically upgrade to v1.9.0 on [deployment](/content/login?from=landing/index.html), while Spice.ai Enterprise customers can consume the Enterprise v1.9.0 image from the [Spice AWS Marketplace listing](https://aws.amazon.com/marketplace/pp/prodview-jmf6jskjvnq7i).

#### What's New in v1.9.0

##### Cayenne Data Accelerator (Beta)

_Cayenne_ is the new premier [data accelerator](/content/use-case/datalake-accelerator/index.html) for high-volume, multi-file workloads. Built on the [Vortex columnar format](https://github.com/vortex-data/vortex) from the Linux Foundation, Cayenne offers better ingestion and query performance than DuckDB without single-file scaling limits. Spice Cayenne supports high concurrency, retention policies, and SQL-based lifecycle management.

.png)Figure 1: TPC-H SF-100 benchmark.png)Figure 2: ClickBench benchmark

Learn more about getting started with Cayenne [in the docs](https://spiceai.org/docs/components/data-accelerators/cayenne).

#### Apache DataFusion v50 Upgrade

DataFusion v50 features faster filter pushdown, new SQL functions, and more reliable execution plans. Learn more in the Apache DataFusion [blog](https://datafusion.apache.org/blog/2025/09/29/datafusion-50.0.0/).

Figure 3: Apache DataFusion performance improvements

#### HTTP Data Connector: Query Endpoints as Tables

Query HTTP endpoints as tables in SQL queries with dynamic filters, with full support for results-caching including new stale-while-revalidate (SWR) support. Learn more [here](https://spiceai.org/docs/components/data-connectors/https).

.png)Figure 4: HTTP data connector query

#### Full-Text and Vector Search on Views

You can now add full-text indexes or embeddings to accelerated views for advanced [search](/content/platform/hybrid-sql-search/index.html) across pre-aggregated or transformed data. Vector engines on views are now also supported. Visit the [search](https://spiceai.org/docs/features/search) and [embeddings](https://spiceai.org/docs/features/embeddings) docs for more information.

.png)Figure 5: Full-text and vector search on views example

#### AWS Authentication Improvements

AWS SDK credential initialization now includes more robust retry logic and better handling, supporting transient and extended network and AWS outages without manual intervention. Learn more.

##### Additional Updates

- **DuckDB v1.4.2**: Composite ART indexes for accelerated table scans, intermediate materialization, and better refresh performance
- **Git Data Connector**: Query data directly from Git repositories
- **DynamoDB Connector Improvements**: Improved filter handling, parallel scan support, and better handling for misconfigured queries
- **Spice Java SDK v0.4.0** with configurable memory limits
- **CLI Improvements:** version pinning, persistent query history, and tab completion
- **Dedicated Query Thread Pool** is now enabled by default

For more details on v1.9.0, visit the [release notes](https://spiceai.org/releases/v1.9.0).