Practical overview
When it fits
Use Spice when an application or agent needs a serving layer that combines federated data access with local acceleration, search, and LLM inference. It is also relevant when operational database data needs a separate analytics replica maintained through CDC.
Who it is for
- Developers building data-grounded AI applications.
- Teams creating AI agents that need searchable, queryable data.
- Infrastructure teams adding analytics beside operational databases.
- Projects needing a self-hosted data-federation and serving layer.
Documented uses
- Connect applications to databases, warehouses, lakes, and APIs through one data service.
- Accelerate selected datasets with local in-memory or on-disk copies.
- Query data with SQL over several standard interfaces.
- Provide vector, full-text, or hybrid search over connected data.
- Serve LLM inference and agent data access over the same data layer.
Requirements & limits
The repository is public, active, and licensed under Apache-2.0.
- Whether the required sources are covered by Spice’s documented connectors, including PostgreSQL, MySQL, Snowflake, Databricks, S3, and Iceberg.
- Whether local acceleration, source-side query push-down, or both fit the workload.
- Whether applications can use the available SQL, search, and inference interfaces.
- Whether the deployment should be a sidecar, standalone service, or distributed cluster.
- Whether native CDC replication is suitable for separating analytics from PostgreSQL, MySQL, or MongoDB production workloads.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Declare and connect datasets with Spice
Set up a Spice data service by declaring datasets from connected sources in a Spice configuration file.
Before you start
- A Spice installation from the public repository.
- Access to one or more supported data sources.
- A spicepod.yaml file.
Steps
Declare the datasets in spicepod.yaml.
Choose which datasets to accelerate locally.
Configure scheduled refreshes or change data capture where applicable.
Expose the data through Spice’s documented SQL, search, or inference interfaces.
Expected outcome
A Spice service configured to serve connected data to applications or agents.
Guide limitations
- The evidence documents the configuration stages but does not provide complete installation commands or a full execution example.
- Acceleration and refresh behavior depend on the selected dataset and source.
Compare options for Model inference and serving
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Spice | Connect applications to databases, warehouses, lakes, and APIs through one data service. | Self-hosted, Server | The repository is public, active, and licensed under Apache-2.0. |
| Morphingdb | Import a libtorch model and create predictions from values in a PostgreSQL table. | Docker, Linux, Self-hosted, Server | The README documents a Docker image and a build-from-repository option, followed by connecting to PostgreSQL and enabling the pgdl extension. |
| Gateway — pyadav | Send chat-completion requests to a provider through a common server endpoint. | Server | The repository is public and identifies the project as Apache-2.0 licensed. README installation instructions include binaries, platform package managers and Docker. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 30 September 2026
- Original website title
- spiceai/spiceai · Official homepage title · retained in its source language