Build Data-Intensive AI Applications on AWS Infrastructure
Spice AI partners with Amazon Web Services to help organizations build real-time, data-driven applications and AI agents. Through native integrations with AWS data services like S3, DynamoDB, Redshift, and AI services like Bedrock, Spice AI enables customers to accelerate development, reduce costs, and deploy production-ready AI applications across AWS infrastructure.
With Spice AI + AWS, Enterprises can:
Build high-performance, secure, and scalable AI applications with unified access to your AWS data and AI services
Federate Data Across AWS Services
Execute fast, low-latency SQL queries across Amazon S3, DynamoDB, Redshift, Aurora, RDS, and Amazon MSK with Spice.ai's unified query engine - no data movement required.
Integrate Amazon Bedrock AI
Connect Amazon Nova, Titan, Claude, and other foundation models through Amazon Bedrock for LLM inference and embeddings, with Bedrock Guardrails support for safe AI applications.
Vector Search with S3 Vectors
Build semantic search and RAG applications using Amazon S3 Vectors - a new S3 bucket type for sub-second similarity queries on billions of vectors at up to 90% cost reduction.
Deploy via AWS Marketplace
Purchase and deploy Spice.ai Enterprise through AWS Marketplace with consolidated billing, Private Offers support, and fast onboarding using your existing AWS procurement workflows.
Native AWS Data & AI Integrations
Spice AI provides deep integrations across the AWS ecosystem for data federation, AI inference, vector search, and secret management:
Amazon S3 & S3 Tables
Query Parquet, CSV, and JSON files from S3 buckets. Access Iceberg tables in Amazon S3 Tables using AWS Glue catalog integration.
Amazon DynamoDB & Streams
Federated SQL queries on DynamoDB tables with automatic schema inference and real-time CDC via DynamoDB Streams.
Amazon Bedrock (Nova, Titan, Claude)
LLM inference with Amazon Nova and embeddings with Titan. Supports Bedrock Guardrails for content filtering.
Amazon S3 Vectors
Sub-second similarity queries on billions of vectors with up to 90% cost reduction compared to traditional vector databases.
AWS Glue Data Catalog
Discover and query tables from AWS Glue with glob pattern filtering. Supports Iceberg tables and tables stored as Parquet or CSV.
Amazon Redshift, Aurora & RDS
Connect to Redshift clusters, Aurora PostgreSQL/MySQL, and RDS instances with full SQL federation.
What Our Customers Say
Cloud Communications Platform
"Spice opened the door to take these critical control-plane datasets and move them next to our services in the runtime path."
AWS Marketplace Enterprise Solution
"Spice.ai Enterprise is available on AWS Marketplace, enabling teams to deploy with existing AWS billing and procurement workflows. With support for Private Offers, consolidated billing, and fast onboarding, enterprises can start building AI applications in minutes."
FAQs
Common questions about running Spice with AWS data and AI services.
Spice queries Amazon S3, S3 Tables, DynamoDB, Redshift, Aurora, and RDS through its data connectors. One SQL statement can join data from several of these services. Spice reads the data in place, so no ETL pipeline copies it first. See SQL federation and acceleration for how these queries run.
Yes. Spice calls Bedrock models from the Amazon Nova, Anthropic Claude, Meta Llama, and Mistral families through its Bedrock model provider. Spice also generates embeddings with Bedrock embedding models such as Amazon Titan. You can attach a Bedrock Guardrail to a model to filter unsafe content. See LLM inference in Spice.
The Glue catalog connector registers each Glue table that matches the glob patterns in its include field. It reads Apache Iceberg tables and tables stored as Parquet or CSV. Spice skips tables in other formats during discovery. After registration, you query the tables with standard SQL like any other dataset.
Data acceleration keeps a copy of hot datasets in an embedded engine such as DuckDB or SQLite next to the application. Queries then read local data and do not wait for a network round trip to S3 or Redshift. Acceleration also reduces request load on the source services. For DynamoDB, DynamoDB Streams keeps the accelerated copy current with table changes.
About Spice AI
Spice AI helps enterprises build fast, accurate, and scalable AI applications and agents with its portable, open-source data and AI compute engine. It connects data from disparate sources including AWS services, simplifies application development, and supports workloads across cloud, edge, and on-premises systems.
Deploy on AWS Marketplace
Ready to deploy Spice AI on AWS? Get Spice.ai Enterprise from AWS Marketplace with consolidated billing, Private Offers support, and enterprise-grade SLAs.
Visit AWS Marketplace (opens in a new tab)