- Home
- Data Virtualization Software
- Spice.ai OSS
Spice.ai OSS
What it is
Spice.ai OSS is a free, open-source SQL query and AI-inference engine built in Rust for developers. Its SQL API can query structured and unstructured data across databases, warehouses, and data lakes. Data can be materialized in DuckDB, SQLite, or PostgreSQL, in memory or on disk, with results caching included. The AI compute engine offers an OpenAI-compatible API for local and hosted inference, search, memory, evaluations, and observability. Search methods include vector, full-text BM25, and hybrid approaches exposed through SQL table-valued functions. Spice lists more than 40 data connectors, including PostgreSQL, MySQL, S3, Snowflake, Databricks, and DuckDB. Model options include hosted OpenAI, Anthropic, and xAI models, as well as local Llama and Phi models with CUDA and Metal acceleration. It supports clients such as DBeaver, Tableau, and Power BI, and monitoring options include Prometheus-compatible metrics, OpenTelemetry export, and Zipkin tracing. Spice can run as a binary, container, Kubernetes workload, or managed app on Spice Cloud. The Apache 2.0-licensed software is self-hostable; listed systems include Apple Silicon macOS, Linux, and Windows 10 or later with WSL.
For and against
For
- Queries structured and unstructured data through a SQL API.
- Includes more than 40 data connectors.
- Offers vector, BM25, and hybrid search methods.
- Can run as a binary, container, or Kubernetes workload.
Against
- Windows support requires WSL.
- Enterprise support is provided for up to three years from release.
The verdict
Spice.ai OSS combines federated SQL, data acceleration, and AI inference in a free, Apache 2.0-licensed package. Its range of deployment options and connectors gives developers several ways to fit it into a data stack.
Who it suits
Spice.ai OSS suits developers who need SQL access across multiple data sources alongside AI inference or search. It also fits teams seeking self-hosted or container and Kubernetes deployment options.
Plans and prices
Spice.ai OSS
Apache 2.0 licensed · self-hostable binary or Docker image
Read 30 Sep 2026 · spiceai.org
On the record
- BI and SQL clients
- Documented client integrations include DBeaver, JetBrains DataGrip, Apache Superset, Tableau, and Microsoft Power BI.spiceai.org · read 30 Sep 2026
- Data acceleration
- It can materialize data in DuckDB, SQLite, and PostgreSQL in memory or on disk, with results caching included.spiceai.org · read 30 Sep 2026
- Data integrations
- The runtime components page says Spice supports over 40 data connectors, including PostgreSQL, MySQL, S3, Snowflake, Databricks, and DuckDB.spiceai.org · read 30 Sep 2026
- Deployment
- Spice runs as a binary, container, Kubernetes workload, or managed app on the Spice Cloud Platform.spiceai.org · read 30 Sep 2026
- Federated SQL
- It provides a SQL API to query structured and unstructured data across databases, data warehouses, and data lakes.spiceai.org · read 30 Sep 2026
- LLM interface
- Its AI compute engine provides an OpenAI-compatible API for local and hosted inference, search, memory, evaluations, and observability.spiceai.org · read 30 Sep 2026
- Model providers
- The feature overview lists hosted OpenAI, Anthropic, and xAI models and local Llama and Phi models, with CUDA and Metal acceleration.spiceai.org · read 30 Sep 2026
- Monitoring
- Observability includes Prometheus-compatible metrics, OpenTelemetry metric export, and Zipkin distributed tracing, with integrations such as Datadog and Grafana.spiceai.org · read 30 Sep 2026
- Operating systems
- The installation page lists macOS with Apple Silicon, Linux on x86_64 or aarch64, and Windows 10 or later with WSL; it says the prebuilt binary needs no other dependencies.spiceai.org · read 30 Sep 2026
- Product
- Spice is an open-source SQL query and AI-inference engine built in Rust for developers.spiceai.org · read 30 Sep 2026
- Search
- Search supports vector, full-text BM25, and hybrid methods through SQL table-valued functions.spiceai.org · read 30 Sep 2026
- Security
- Spice supports optional API-key authentication for API endpoints, and its secret stores can use environment variables, files, or external managers such as AWS Secrets Manager and Azure Key Vault.spiceai.org · read 30 Sep 2026
- Support
- The getting-started documentation says Spice.ai Enterprise provides full support for up to three years from release date.spiceai.org · read 30 Sep 2026
