# Spice AI

> Spice.ai is a data and AI platform that combines federated SQL query, hybrid search, and LLM inference in a portable, open-source runtime

Spice AI provides a unified data and AI infrastructure platform. Our open-source runtime enables federated SQL query across multiple data sources, hybrid vector and full-text search, and seamless LLM inference integration.

## Documentation

- [Spice AI Documentation](https://docs.spice.ai): Complete documentation for Spice AI
- [GitHub Repository](https://github.com/spiceai/spiceai): Open-source codebase and examples

## Local Content

- [About Us](https://spice.ai/about-us): Learn about Spice AI's mission, team, and vision for empowering developers to build intelligent apps with unified data and AI infrastructure.
- [2025 Spice AI Year in Review](https://spice.ai/blog/2025-spice-ai-year-in-review): From day one, Spice was designed to simplify building modern, intelligent applications. In 2025 that vision turned into reality.
- [A Developer's Guide to Understanding Spice.ai](https://spice.ai/blog/a-developers-guide-to-understanding-spice-ai): Learn what Spice.ai is, when to use it, and how it solves enterprise data challenges. A developer-focused guide to federation, acceleration, search, and AI.
- [A New Class of Applications That Learn and Adapt](https://spice.ai/blog/a-new-class-of-applications-that-learn-and-adapt): Explore the history of decision engines and how modern machine learning enables applications that learn, adapt, and make better decisions over time with Spice.ai.
- [Adding Spice - The Next Generation of Spice.ai OSS](https://spice.ai/blog/adding-spice-the-next-generation-of-spice-ai-oss): Learn how Spice.ai OSS was rebuilt in Rust to deliver fast, local SQL queries across databases, warehouses, and data lakes.
- [AI needs AI-ready data](https://spice.ai/blog/ai-needs-ai-ready-data): An introduction to AI-ready data and how Spice.ai handles normalization, encoding, and real-time data preparation for ML applications.
- [Spice.ai Now Supports Amazon S3 Vectors For Vector Search at Petabyte Scale!](https://spice.ai/blog/amazon-s3-vectors): Spice AI has partnered with AWS to integrate Amazon S3 Vectors into the Spice.ai Open Source data and AI compute engine.
- [Announcing Spice.ai Open Source 1.0-stable: A Portable Compute Engine for Data-Grounded AI - Now Ready for Production](https://spice.ai/blog/announcing-spice-ai-open-source-1-0-stable): Learn how Spice.ai OSS grounds AI in real data with federated query, fast retrieval, and portable deployment anywhere.
- [Spice Cloud v1.7.0: DataFusion v49, Full-Text Search Updates & More](https://spice.ai/blog/announcing-spice-cloud-v1-7-0): Spice Cloud v1.7.0 includes DataFusion v49, EmbeddingGemma support, and real-time indexing for full-text search
- [Apache Ballista at Spice AI: Distributed Query Execution Without the Operational Tax](https://spice.ai/blog/apache-ballista-at-spice-ai): A technical deep-dive into how Spice AI integrates Apache Ballista for distributed query execution with multi-active schedulers, fault-tolerant shuffle, and distributed acceleration.
- [Apache Iceberg at Spice AI: How we Query, Accelerate, and Write to Open Table Formats](https://spice.ai/blog/apache-iceberg-at-spice-ai): A technical deep-dive into how Spice AI integrates Apache Iceberg for federated queries, sub-second acceleration, and ACID-compliant writes to open table formats.
- [AWS Workshop: Federated Queries and Hybrid Search with Spice.ai](https://spice.ai/blog/aws-workshop-federated-queries-and-hybrid-search-with-spice): A new AWS catalog workshop that walks through deploying Spice as a data and AI substrate across AWS infrastructure for federated queries, hybrid search, and LLM inference.
- [Barracuda Networks Gains 100x Faster Query Responses and 50% Reduction in Operational Costs with Spice.ai OSS](https://spice.ai/blog/barracuda-networks-100x-faster-query-responses-with-spice): Barracuda Networks used Spice.ai OSS to accelerate archival and audit data access, improving query performance by up to 100x while reducing operational costs by 50%.
- [Basis Set Ventures Deploys Spice.ai to Power Natural Language Queries and Mitigate Hallucinations](https://spice.ai/blog/basis-set-ventures-deploys-spice-ai): Basis Set Ventures uses Spice.ai Enterprise to power natural language searches directly against real-time datasets.
- [Localhost Latency at Scale: The Spice Cluster-Sidecar Architecture](https://spice.ai/blog/cluster-sidecar-architecture): How the Spice cluster-sidecar architecture gives applications, services, and AI agents a sandboxed, localhost-latency data and inference plane backed by a distributed Spice cluster, without exposing underlying data systems.
- [Spice AI Announces Contribution of TableProviders for PostgreSQL, MySQL, DuckDB, and SQLite to the Apache DataFusion Project](https://spice.ai/blog/contribution-of-tableproviders-to-datafusion): Spice AI has contributed new TableProviders for PostgreSQL, MySQL, DuckDB, and SQLite to the Apache DataFusion project.
- [Announcing Our Partnership with Databricks!](https://spice.ai/blog/databricks-partnership): Spice partners with Databricks to accelerate operational AI apps with fast SQL queries, Mosaic AI embeddings, and Unity Catalog governance.
- [Getting started with Amazon S3 Vectors and Spice](https://spice.ai/blog/getting-started-with-amazon-s3-vectors-and-spice): Learn how Spice AI integrates Amazon S3 Vectors for scalable, cost-effective vector search - combining semantic, full-text, and SQL queries in one runtime.
- [How we use Apache DataFusion at Spice AI](https://spice.ai/blog/how-we-use-apache-datafusion-at-spice-ai): A technical overview of how Spice extends Apache DataFusion with custom table providers, optimizer rules, and UDFs to power federated SQL, search, and AI inference.
- [Interviewing at Spice AI](https://spice.ai/blog/interviewing-at-spice-ai): A guide to the Spice AI interview process, covering what to expect at each stage, how we evaluate candidates, and tips for preparation.
- [Introducing Spice Cayenne: The Next-Generation Data Accelerator Built on Vortex for Performance and Scale](https://spice.ai/blog/introducing-spice-cayenne-data-accelerator): Spice Cayenne is the next-generation Spice.ai data accelerator built for high-scale and low latency data lake workloads.
- [Introducing Spice Skills for AI Agents](https://spice.ai/blog/introducing-spice-skills-for-ai-coding-agents): Spice Skills is a collection of packaged agent instructions for working with Spice.ai OSS, covering setup, data connections, acceleration, search, AI, and more.
- [Making Apps That Learn And Adapt](https://spice.ai/blog/making-apps-that-learn-and-adapt): Building intelligent applications is still too hard for most developers-not because ML is impossible, but because it's treated as something separate from the app.
- [Making Object Storage Operational for Real-Time and AI Workloads](https://spice.ai/blog/making-object-storage-operational): Transform object stores into real-time AI platforms. Spice adds federation, acceleration, hybrid search, and inference capabilities.
- [Migrating Off Dremio: A Phased Guide for Data Teams](https://spice.ai/blog/migrating-off-dremio-phased-guide): Learn how to migrate from Dremio to Spice in phases by running side by side, moving one dataset at a time, and validating each step before decommissioning.
- [Multi-Tenancy for AI Agents without the Pipelines](https://spice.ai/blog/multi-tenancy-for-ai-agents-without-pipelines): Learn how to serve multi-tenant AI agents from disparate enterprise data sources with tenant isolation, federated SQL, and no per-tenant pipelines.
- [On Writing](https://spice.ai/blog/on-writing): Writing is fundamental to formalizing thoughts, communicating effectively, and is the ultimate creation tool.
- [Building an Enterprise SRE Agent with OpenClaw and Spice](https://spice.ai/blog/openclaw-and-spice-governed-access-to-production-data-for-enterprise-agents): How to build an OpenClaw SRE with Spice for safe, unified, and observable access to production data, demonstrated with real-world incident workflows.
- [Operationalizing Amazon S3 for AI: From Data Lake to AI-Ready Platform in Minutes](https://spice.ai/blog/operationalizing-amazon-s3-for-ai): Transform Amazon S3 from passive storage to an AI-ready platform. Real-world example using Spice and S3 for hybrid search and LLM inference.
- [Real-Time Control Plane Acceleration with DynamoDB Streams ](https://spice.ai/blog/real-time-acceleration-with-dynamodb-streams): How to sync DynamoDB data to thousands of nodes with sub-second latency using a two-tier architecture with DynamoDB Streams and Spice acceleration.
- [Real-Time Hybrid Search Using RRF: A Hands-On Guide with Spice](https://spice.ai/blog/real-time-hybrid-search-using-rrf): Learn how to build hybrid search with Reciprocal Rank Fusion (RRF) directly in SQL using Spice - combining text, vector, and time-based relevance in one query for faster, more accurate results.
- [Spice 2.0: Real-Time Analytical Query on Operational Data, Without ETL](https://spice.ai/blog/spice-2-0-is-now-available): Spice 2.0 is now available: add real-time analytical query and search to operational data without ETL via high-throughput CDC replication, petabyte-scale distributed compute built on Apache Ballista, and enterprise-grade controls.
- [Spice AI achieves SOC 2 Type II compliance](https://spice.ai/blog/spice-ai-achieves-soc-2-type-ii-compliance): Spice AI completes SOC 2 Type II audit, demonstrating enterprise-grade security and compliance for its data and AI infrastructure platform.
- [The Spice.ai for GitHub Copilot Extension is now available!](https://spice.ai/blog/spice-ai-for-github-copilot-extension-now-available): With the Spice.ai Extension, developers can interact with data, like product requirements documents (PRDs), tickets, and tabular data, from any external data source directly within GitHub Copilot. Save hours copying and pasting across various platforms, relevant data and answers are now surfaced in Copilot Chat, right when you need it.
- [Spice.ai is now generally available!](https://spice.ai/blog/spice-ai-is-now-generally-available): Spice.ai is now available for everyone, including a new community-centric developer hub and Community Edition complimentary for developers.
- [Faster, Simpler Dashboards with Spice and Power BI](https://spice.ai/blog/spice-and-power-bi): Spice AI built a Microsoft Power BI Connector on top of the Flight SQL ADBC driver that makes it easy for Power BI users to query across operational databases, analytical warehouses, and object stores.
- [Spice Cloud v1.10: Caching Acceleration Mode, DynamoDB Streams Support, & More!](https://spice.ai/blog/spice-cloud-v1-10): Spice v1.10 includes a new caching acceleration mode, a new DynamoDB Streams data connector in preview, Amazon S3 location-based pruning, S3 Tables write support, and several performance and security improvements.
- [Spice Cloud v1.11: Spice Cayenne Reaches Beta, Apache DataFusion v51, DynamoDB Streams Improvements, & More](https://spice.ai/blog/spice-cloud-v1-11): v1.11 brings Spice Cayenne to Beta, DataFusion v51 and Apache Arrow v57.2, improved DynamoDB Streams, and more.
- [Spice Cloud v1.8.0: Iceberg Write Support, Acceleration Snapshots & More](https://spice.ai/blog/spice-cloud-v1-8-0-iceberg-writes): Announcing Spice Cloud v1.8.0 - now with Iceberg write support, acceleration snapshots, partitioned S3 Vectors indexes & a new AI SQL function
- [Spice Cloud v1.9.0: Introducing the Spice Cayenne Data Accelerator](https://spice.ai/blog/spice-cloud-v1-9-0-cayenne-data-accelerator): Spice Cloud v1.9.0 adds the Cayenne Data Accelerator, Apache DataFusion v50, HTTP data connector support for querying endpoints as tables, and much more.
- [Spice Cloud v2.0-rc.2: Cayenne RC, ADBC BigQuery, and Catalog Connectors](https://spice.ai/blog/spice-cloud-v2-0-rc-2): Spice Cloud v2.0-rc.2 introduces Spice Cayenne release candidate status, ADBC with BigQuery support, new catalog connectors, and major developer experience upgrades.
- [Spice Firecache | Cloud-Scale DuckDB](https://spice.ai/blog/spice-firecache): Cloud-Scale DuckDB
- [Spice OSS, rebuilt in Rust](https://spice.ai/blog/spice-oss-rebuilt-in-rust): Spice.ai OSS has been rebuilt from the ground up in Rust, delivering the performance, safety, and portability needed for production data infrastructure.
- [Getting Started with Spice.ai SQL Query Federation & Acceleration](https://spice.ai/blog/spice-sql-query-federation-acceleration): Learn how to use Spice.ai to federate and accelerate queries across operational and analytical systems with zero ETL.
- [Spice.ai's approach to Time-Series AI](https://spice.ai/blog/spiceais-approach-to-time-series-ai): Explore the challenges of time-series AI and why Spice.ai uses a data-driven reinforcement learning approach to help developers build adaptive, intelligent applications.
- [Spicepods: From Zero to Hero](https://spice.ai/blog/spicepods-from-zero-to-hero): A step-by-step guide to authoring a Spicepod from scratch and using it to build an application that learns and adapts over time.
- [Teaching Apps how to Learn with Spicepods](https://spice.ai/blog/teaching-apps-how-to-learn-with-spicepods): Learn how Spicepods define application goals, rewards, and learning behavior - making it easy for developers to build applications that learn and adapt over time.
- [The Analytics Replica Pattern: The Shortest Path to Data-Grounded AI](https://spice.ai/blog/the-analytics-replica-pattern-shortening-the-path-to-data-based-ai): The analytics replica pattern keeps operational systems isolated while delivering real-time analytical queries for AI agents and dashboards, without ETL or operational risk.
- [True Hybrid Search: Vector, Full-Text, and SQL in One Runtime](https://spice.ai/blog/true-hybrid-search): Build hybrid search without managing multiple systems. Query vectors, run full-text search, and execute SQL in one unified runtime.
- [Vortex at Spice AI: The Columnar Format for Data-Intensive Workloads](https://spice.ai/blog/vortex-at-spice-ai-the-columnar-format-for-data-intensive-workloads): How Spice AI uses the Vortex columnar format in Cayenne to improve query latency, reduce memory overhead, and support high-concurrency data-intensive workloads.
- [What Data Informs AI-driven Decision Making?](https://spice.ai/blog/what-data-informs-ai-driven-decision-making): Learn the three classes of data required for intelligent decision-making and how Spice.ai simplifies runtime data engineering for AI-powered applications.
- [Write to Apache Iceberg Tables with SQL in Spice](https://spice.ai/blog/write-to-apache-iceberg-tables-with-sql): Spice v1.8 adds native Apache Iceberg write support with standard SQL INSERT INTO statements. Build complete data workflows without ETL - query, accelerate, and write from one runtime.
- [Careers](https://spice.ai/careers): Join Spice AI and help build the future of data and AI infrastructure.
- [Contact](https://spice.ai/contact): Get in touch with the Spice AI team. Whether you're exploring enterprise deployments, pricing, integrations, or technical questions, we're here to help.
- [Spice.ai Cookbook](https://spice.ai/cookbook): A collection of guides and samples to help you build data-grounded AI apps and agents with Spice.ai Open-Source. Find ready-to-use examples for data acceleration, AI agents, LLM memory, and more.
- [AI Model Serving](https://spice.ai/feature/ai-model-serving): Serve, evaluate, and ground AI models directly inside Spice. Call LLMs locally or connect to hosted providers from one secure, high-performance runtime.
- [Distributed Query](https://spice.ai/feature/distributed-query): Scale beyond single-node limits with petabyte-scale, multi-node, distributed queries.
- [Edge to Cloud Deployments](https://spice.ai/feature/edge-to-cloud-deployments): Deploy Spice anywhere, from lightweight sidecars to enterprise clusters. Choose the architecture that fits your performance, scale, and governance needs.
- [MCP Server & Gateway](https://spice.ai/feature/mcp-server-gateway): Deploy MCP servers locally or over SSE, route tools to models, and expose Spice securely as an MCP gateway with full observability.
- [Real-Time Change Data Capture](https://spice.ai/feature/real-time-change-data-capture): Sync accelerated datasets with real-time changes using Change Data Capture (CDC) and maintain low-latency analytics without full-table refreshes.
- [Secure AI Sandboxing](https://spice.ai/feature/secure-ai-sandboxing): Safely connect AI to enterprise data. Spice isolates access for agents and models, enforcing least privilege, observability, and compliance across every query.
- [Get a demo](https://spice.ai/get-a-demo): Get in touch with the Spice AI team. Whether you're exploring enterprise deployments, pricing, integrations, or technical questions, we're here to help.
- [Home](https://spice.ai/home): Deploy analytics replicas alongside operational databases to give apps and agents fast, sandboxed access to real-time data. Fully open source.
- [Cybersecurity](https://spice.ai/industry/cybersecurity): Build fast, reliable, and intelligent cybersecurity applications. Spice delivers unified data access, real-time performance, and embedded AI integration across any environment.
- [Financial Services](https://spice.ai/industry/financial-services): Unify, govern, and accelerate sensitive financial data. Spice delivers federation, hybrid search, and integrated AI for regulated workloads.
- [SaaS](https://spice.ai/industry/saas): Power SaaS with live, governed data. Federate across warehouses and DBs, accelerate to millisecond latency, and add AI-all on one portable runtime.
- [Integrations](https://spice.ai/integrations): Spice offers 40+ integrations with leading databases, warehouses, data lakes, streaming systems, and more.
- [What is Apache Arrow?](https://spice.ai/learn/apache-arrow): Apache Arrow is a cross-language development platform for in-memory columnar data. Learn how Arrow works, its columnar memory format, Arrow Flight, Arrow IPC, and how it compares to Parquet, Protocol Buffers, and CSV.
- [What is Apache Ballista?](https://spice.ai/learn/apache-ballista): Apache Ballista is a distributed SQL query engine that extends Apache DataFusion across multiple nodes. Learn how Ballista works, its architecture, and how it compares to Spark and Trino.
- [Apache DataFusion vs DuckDB: How to Choose](https://spice.ai/learn/apache-datafusion-vs-duckdb): Apache DataFusion and DuckDB are both fast analytical query engines built for in-process use. Learn the key differences in architecture, extensibility, and when to use each.
- [What is Apache DataFusion?](https://spice.ai/learn/apache-datafusion): Apache DataFusion is an open-source, extensible SQL query engine written in Rust. Learn how DataFusion works, its architecture, how it compares to Trino and DuckDB, and how teams extend it for production use cases.
- [Apache Iceberg vs Delta Lake: How to Choose](https://spice.ai/learn/apache-iceberg-vs-delta-lake): Apache Iceberg and Delta Lake are both open table formats that bring transactions, schema evolution, and time travel to data lakes. Learn the key architectural differences and when to use each.
- [What is Apache Iceberg?](https://spice.ai/learn/apache-iceberg): Apache Iceberg is an open table format for large analytic datasets. Learn how Iceberg works, its architecture, key features like schema evolution and time travel, and how it compares to Delta Lake and Hive tables.
- [Best Alternatives to ETL Pipelines for AI Agents](https://spice.ai/learn/best-alternatives-to-etl-pipelines-for-ai-agents): Objective guide to alternatives to ETL pipelines for AI agents, including federation, CDC, event streaming, and hybrid data architecture patterns.
- [What is BM25 Full-Text Search?](https://spice.ai/learn/bm25-full-text-search): BM25 (Best Match 25) is the standard ranking function for full-text search. Learn how BM25 scores documents using term frequency, inverse document frequency, and document length normalization, and how it compares to TF-IDF.
- [Cache Invalidation at Scale: Why Manual Strategies Break](https://spice.ai/learn/cache-invalidation-at-scale): Learn why TTLs, purge-on-write, and event-driven cache invalidation break down as systems scale, and how declarative refresh policies replace hand-written invalidation logic.
- [Caching vs Data Acceleration: How to Choose](https://spice.ai/learn/caching-vs-data-acceleration): Compare caching and data acceleration: two approaches to making data access faster. Learn the key differences in invalidation, freshness, query flexibility, and operational cost, and when each approach fits.
- [How to Implement Change Data Capture (CDC)](https://spice.ai/learn/change-data-capture): Change data capture (CDC) tracks row-level changes in databases and streams them in real time. Learn how to implement CDC with log-based, trigger-based, and polling patterns for real-time data pipelines.
- [Comparing Data Federation Tools for AI Agents: How to Choose](https://spice.ai/learn/comparing-data-federation-tools-for-ai-agents): Compare data federation tool categories for AI agents across latency, freshness, governance, and operational overhead. Learn which approach fits your architecture.
- [What is Data Acceleration?](https://spice.ai/learn/data-acceleration): Data acceleration caches frequently accessed data locally for sub-second query performance without moving data permanently. Learn how it works, acceleration strategies, and when to use it.
- [Data Lakehouse vs Data Warehouse: How to Choose](https://spice.ai/learn/data-lakehouse-vs-data-warehouse): Compare data lakehouses and data warehouses: two architectures for analytical data. Learn the key differences in storage formats, cost, workload fit, and application-serving latency, and when each is appropriate.
- [What is a Data Substrate?](https://spice.ai/learn/data-substrate): A data substrate is a co-located data and AI layer that provides applications with sub-millisecond access to any data source. Learn how it differs from a data warehouse, lake, or lakehouse, and why it is an architectural primitive rather than a destination.
- [Data Virtualization vs Data Replication: How to Choose](https://spice.ai/learn/data-virtualization-vs-replication): Compare data virtualization and data replication: two foundational approaches to data integration. Learn the key differences, tradeoffs, and when each approach is appropriate for your workloads.
- [What is Data Virtualization?](https://spice.ai/learn/data-virtualization): Data virtualization provides a unified view of data across multiple sources without physical replication. Learn how it works, how it compares to ETL, and when to use it.
- [Why Query Latency Gets Worse as Your Application Scales](https://spice.ai/learn/database-query-latency-at-scale): Learn why database query latency degrades as applications scale: rising concurrency, growing data volume, and the distance between where data lives and where applications run.
- [What is Delta Lake?](https://spice.ai/learn/delta-lake): Delta Lake is an open-source storage layer that brings ACID transactions, schema enforcement, and time travel to data lakes. Learn how Delta Lake works, its architecture, how it compares to Apache Iceberg and Hudi, and how Spice connects to Delta tables for federated queries.
- [What is DuckDB?](https://spice.ai/learn/duckdb): DuckDB is an open-source, in-process analytical database designed for fast OLAP queries. Learn how DuckDB works, its columnar architecture, and how Spice uses DuckDB as a data accelerator engine.
- [What are Embeddings?](https://spice.ai/learn/embeddings): Embeddings are dense vector representations of text, images, or code that capture semantic meaning. Learn how embedding models work, what dimensions represent, and how embeddings enable semantic search, RAG, and classification.
- [Full-Text Search vs Vector Search: How to Choose](https://spice.ai/learn/full-text-search-vs-vector-search): Full-text search matches exact keywords using BM25 scoring, while vector search finds semantically similar content using embeddings. Learn the key differences, when to use each approach, and how hybrid search combines both for optimal results.
- [How to Connect AI Agents to Live Operational Data Without ETL](https://spice.ai/learn/how-to-connect-ai-agents-to-live-operational-data-without-etl): Practical guide to connecting AI agents to live operational data using federation, acceleration, and policy controls instead of batch ETL pipelines.
- [How to Give Each AI Agent Its Own Isolated Data Environment](https://spice.ai/learn/how-to-give-each-ai-agent-its-own-isolated-data-environment): Practical architecture guide for isolating AI agent data environments using scoped credentials, runtime boundaries, and policy enforcement patterns.
- [How to Reduce Data Lakehouse Costs for Agentic Workloads](https://spice.ai/learn/how-to-reduce-data-lakehouse-costs-for-agentic-workloads): Practical framework for reducing data lakehouse costs in agentic workloads by separating serving paths, minimizing expensive query patterns, and optimizing retrieval architecture.
- [How to Sandbox Data Access for AI Agents](https://spice.ai/learn/how-to-sandbox-data-access-for-ai-agents): Step-by-step guide to sandboxing AI agent data access with least-privilege policies, query guardrails, redaction, and runtime controls for production safety.
- [What is a Hybrid Data Architecture?](https://spice.ai/learn/hybrid-data-architecture): A hybrid data architecture combines application sidecars for sub-millisecond reads with a centralized cluster for data ingestion, acceleration, and distributed compute. Learn how the sidecar-cluster pattern works and when to use it.
- [What is Hybrid Search?](https://spice.ai/learn/hybrid-search): Hybrid search combines vector similarity search with keyword matching to deliver more accurate results than either method alone. Learn how hybrid search works, ranking algorithms like RRF, and when to use it.
- [Learn Data & AI](https://spice.ai/learn/index): Learn about the core technologies behind Spice.ai: SQL federation, data virtualization, RAG, hybrid search, change data capture, the Model Context Protocol, and more.
- [What is LLM Inference?](https://spice.ai/learn/llm-inference): LLM inference is the process of generating text by running input through a trained large language model. Learn how inference works, key performance metrics, and optimization techniques like KV caching, quantization, and speculative decoding.
- [What is LLM Tool Calling?](https://spice.ai/learn/llm-tool-calling): LLM tool calling is a capability where a model outputs structured function calls instead of plain text, enabling AI agents to query databases, call APIs, and take actions. Learn how tool calling works, security considerations, and how MCP standardizes tool use.
- [Managed Apache DataFusion: Federated SQL at Scale](https://spice.ai/learn/managed-apache-datafusion): Managed Apache DataFusion provides federated SQL across heterogeneous systems with centralized operations, query optimization, and production controls. Learn how it works and when to use it.
- [What is the Model Context Protocol (MCP)?](https://spice.ai/learn/model-context-protocol): The Model Context Protocol (MCP) is an open standard for connecting AI models to external data and tools. Learn how MCP works, its architecture, and how MCP servers enable agentic AI.
- [RAG vs Fine-Tuning: How to Choose](https://spice.ai/learn/rag-vs-fine-tuning): RAG retrieves external data at inference time while fine-tuning embeds knowledge into model weights. Learn the key differences, tradeoffs, and when to use each approach for production AI applications.
- [What is Reciprocal Rank Fusion (RRF)?](https://spice.ai/learn/reciprocal-rank-fusion): Reciprocal Rank Fusion (RRF) is a ranking algorithm that merges result lists from multiple search methods by position rather than score. Learn how RRF works, its parameters, and why it is the standard fusion method in hybrid search.
- [What is Retrieval Augmented Generation (RAG)?](https://spice.ai/learn/retrieval-augmented-generation): Retrieval augmented generation (RAG) grounds LLM responses in real data by retrieving relevant context at inference time. Learn how RAG works, its architecture, and production best practices.
- [What is the Sidecar Pattern?](https://spice.ai/learn/sidecar-pattern): The sidecar pattern deploys a helper process alongside the main application in the same host or pod, communicating over local loopback. Learn how sidecars work, when to use them, and how the pattern applies to data and AI runtimes.
- [Sidecar vs Microservice Architecture: How to Choose](https://spice.ai/learn/sidecar-vs-microservice-architecture): Sidecar and microservice are two deployment architectures for data and AI runtimes. Learn the key differences in latency, scaling, resource usage, and when to use each pattern.
- [SQL Federation vs ETL: How to Choose](https://spice.ai/learn/sql-federation-vs-etl): SQL federation and ETL are two approaches to accessing data across distributed systems. Learn the key differences, when to use each, and how modern platforms combine both for real-time performance.
- [How to Do SQL Query Federation](https://spice.ai/learn/sql-federation): SQL query federation lets you query multiple databases with a single SQL statement without moving data. Learn how to set up federated queries, optimize with predicate pushdown, and combine federation with acceleration.
- [What is Tantivy?](https://spice.ai/learn/tantivy): Tantivy is an open-source, full-text search engine library written in Rust, inspired by Apache Lucene. Learn how Tantivy works, its architecture, how it compares to Lucene and Elasticsearch, and how Spice uses it for search.
- [How to Use Text-to-SQL](https://spice.ai/learn/text-to-sql): Text-to-SQL uses large language models to translate natural language questions into SQL queries. Learn how to implement text-to-SQL with schema-aware generation, prompt engineering, and production safeguards.
- [What is Vector Search?](https://spice.ai/learn/vector-search): Vector search (semantic search) finds the most similar items by comparing vector embeddings using distance metrics like cosine similarity. Learn how ANN algorithms like HNSW work, vector index tradeoffs, and when to use vector search vs. keyword search.
- [What is Vortex?](https://spice.ai/learn/vortex): Vortex is an open-source compressed columnar file format designed for analytical queries. Learn how Vortex compares to Parquet, its adaptive encoding system, and when to use it over Parquet for analytical workloads.
- [What is Zero-ETL?](https://spice.ai/learn/zero-etl): Zero-ETL is a data architecture approach that eliminates traditional extract-transform-load pipelines by querying data in place or using CDC-backed acceleration. Learn what zero-ETL means in practice and when to use it.
- [Spice AI for AWS](https://spice.ai/partners/aws): Build Fast, Scalable AI Applications with Spice AI and Amazon Web Services
- [Spice AI for Databricks](https://spice.ai/partners/databricks): Build Fast, Accurate AI Applications with Spice AI and Databricks
- [Spice AI for NetApp](https://spice.ai/partners/netapp): Build Accelerated, Data-Grounded AI Applications with Spice AI and NetApp ONTAP
- [Partners](https://spice.ai/partners): Partner with Spice AI to deliver faster data, search, and AI solutions for your customers. Build integrations, co-market solutions, and power data-intensive applications and AI agents together.
- [Analytics Replica](https://spice.ai/platform/analytics): Add a real-time analytics replica to PostgreSQL, MySQL, and MongoDB in minutes. Spice replicates operational data with high-throughput CDC for sub-second analytics, without ETL and without load on production.
- [Hybrid SQL Search](https://spice.ai/platform/hybrid-sql-search): Combine vector similarity, full-text, and keyword search in one SQL query. Fast, scalable, and production-ready.
- [LLM Inference](https://spice.ai/platform/llm-inference): Call LLMs directly from SQL. Generate, summarize, and enrich data inline using the SQL AI function or natural language queries.
- [SQL Federation & Acceleration](https://spice.ai/platform/sql-federation-acceleration): Query any data source with sub-second speed. Spice combines SQL federation and acceleration in a single runtime with zero ETL.
- [Spice Cloud Plans](https://spice.ai/pricing/cloud): Flexible cloud pricing plans for teams of all sizes
- [Pricing](https://spice.ai/pricing): Start free and deploy anywhere: on your laptop, on-prem, at the edge, or in the cloud. Flexible pricing designed for teams building data-intensive applications & AI agents.
- [Privacy Policy](https://spice.ai/privacy-policy): Your privacy matters. Read Spice AI's policy on data collection, usage, protection, and your rights to control your personal information.
- [Security](https://spice.ai/security): Learn how Spice AI protects your data with SOC 2 Type II compliance, strong access controls, encryption, secure coding, and a principled, defense-in-depth approach.
- [Analytics](https://spice.ai/use-case/analytics): Add a real-time analytics replica to PostgreSQL, MySQL, and MongoDB in minutes. Run sub-second analytics on operational data with high-throughput CDC replication, without ETL and without load on production.
- [Application Search](https://spice.ai/use-case/application-search): Add fast, relevant search to your app with hybrid SQL search. Governed, low-latency, and easy to ship anywhere.
- [Datalake Accelerator](https://spice.ai/use-case/datalake-accelerator): Accelerate query performance in your data lake with Spice. Run SQL locally on federated datasets for up to 100x faster performance.
- [Operational Data Lakehouse](https://spice.ai/use-case/operational-data-lakehouse): Federate, accelerate, and serve data-intensive apps and AI agents directly from object storage with millisecond performance.
- [Retrieval-Augmented Generation](https://spice.ai/use-case/retrieval-augmented-generation): Build more accurate and trustworthy RAG systems. Spice unifies SQL federation, vector search, and model inference for data-grounded AI responses.
- [Secure AI Agents](https://spice.ai/use-case/secure-ai-agents): Build and deploy AI agents that are secure by design. Federate-governed context, enforce policy inline, and route to any model with full auditability.

## Blog

- [2025 Spice AI Year in Review](https://spice.ai/blog/2025-spice-ai-year-in-review)
- [A Developer's Guide to Understanding Spice.ai](https://spice.ai/blog/a-developers-guide-to-understanding-spice-ai)
- [A New Class of Applications That Learn and Adapt](https://spice.ai/blog/a-new-class-of-applications-that-learn-and-adapt)
- [Adding Spice - The Next Generation of Spice.ai OSS](https://spice.ai/blog/adding-spice-the-next-generation-of-spice-ai-oss)
- [AI needs AI-ready data](https://spice.ai/blog/ai-needs-ai-ready-data)
- [Spice.ai Now Supports Amazon S3 Vectors For Vector Search at Petabyte Scale!](https://spice.ai/blog/amazon-s3-vectors)
- [Announcing Spice.ai Open Source 1.0-stable: A Portable Compute Engine for Data-Grounded AI - Now Ready for Production](https://spice.ai/blog/announcing-spice-ai-open-source-1-0-stable)
- [Spice Cloud v1.7.0: DataFusion v49, Full-Text Search Updates & More](https://spice.ai/blog/announcing-spice-cloud-v1-7-0)
- [Apache Ballista at Spice AI: Distributed Query Execution Without the Operational Tax](https://spice.ai/blog/apache-ballista-at-spice-ai)
- [Apache Iceberg at Spice AI: How we Query, Accelerate, and Write to Open Table Formats](https://spice.ai/blog/apache-iceberg-at-spice-ai)
- [AWS Workshop: Federated Queries and Hybrid Search with Spice.ai](https://spice.ai/blog/aws-workshop-federated-queries-and-hybrid-search-with-spice)
- [Barracuda Networks Gains 100x Faster Query Responses and 50% Reduction in Operational Costs with Spice.ai OSS](https://spice.ai/blog/barracuda-networks-100x-faster-query-responses-with-spice)
- [Basis Set Ventures Deploys Spice.ai to Power Natural Language Queries and Mitigate Hallucinations](https://spice.ai/blog/basis-set-ventures-deploys-spice-ai)
- [Localhost Latency at Scale: The Spice Cluster-Sidecar Architecture](https://spice.ai/blog/cluster-sidecar-architecture)
- [Spice AI Announces Contribution of TableProviders for PostgreSQL, MySQL, DuckDB, and SQLite to the Apache DataFusion Project](https://spice.ai/blog/contribution-of-tableproviders-to-datafusion)
- [Announcing Our Partnership with Databricks!](https://spice.ai/blog/databricks-partnership)
- [Getting started with Amazon S3 Vectors and Spice](https://spice.ai/blog/getting-started-with-amazon-s3-vectors-and-spice)
- [How we use Apache DataFusion at Spice AI](https://spice.ai/blog/how-we-use-apache-datafusion-at-spice-ai)
- [Interviewing at Spice AI](https://spice.ai/blog/interviewing-at-spice-ai)
- [Introducing Spice Cayenne: The Next-Generation Data Accelerator Built on Vortex for Performance and Scale](https://spice.ai/blog/introducing-spice-cayenne-data-accelerator)
- ... and 31 more blog posts

## Optional

- [Full LLMs Content](https://spice.ai/llms-full.txt): Complete content for deep context
- [Sitemap](https://spice.ai/sitemap.xml): Full sitemap for crawling