redis-ai-resources

Getting Started

New to Redis for AI applications? Here's how to get started:

1. First time with Redis? Start with our Redis Intro notebook
2. Want to try vector search? Check our Vector Search with RedisVL recipe
3. Building a RAG application? Begin with RAG from Scratch
4. Ready to see it in action? Play with the Redis RAG Workbench demo

Demos

No faster way to get started than by diving in and playing around with a demo.

Demo	Description
Redis RAG Workbench	Interactive demo to build a RAG-based chatbot over a user-uploaded PDF. Toggle different settings and configurations to improve chatbot performance and quality. Utilizes RedisVL, LangChain, RAGAs, and more.
Redis VSS - Simple Streamlit Demo	Streamlit demo of Redis Vector Search
ArXiv Search	Full stack implementation of Redis with React FE
Product Search	Vector search with Redis Stack and Redis Enterprise
ArxivChatGuru	Streamlit demo of RAG over Arxiv documents with Redis & OpenAI
Redis Movies Searcher	Demo of hybrid search using Java, Spring Boot, and Redis OM
Memory-Aware Alexa Assistant	Complete example of an Alexa skill that can recall previously stored conversations and memories to provide contextual responses to users. Utilizes Redis Agent Memory Server, LangChain4J, Terraform, and AWS. It showcases how to implement context engineering to dynamically leverage RAG, tools, short-term and long-term memories.
From Postgres to Redis Cloud with RDI	This demo shows how to deploy RDI on Kubernetes running locally so it can be used to build a moving/transformation data pipeline to enable data on Redis Cloud. The source database lives on-premises, and the data is movied and transformed to Redis Cloud.
Virtual Banking Assistant with Semantic Router	A Banking Agent demo that uses semantic routing to route user queries to right banking tool. (Eg: Credit Card Recommendation, Loan EMI calculation, Banking FAQ etc)
Shopping AI Agent	An Agentic AI grocery-shopping platform demo that combines Redis's speed with LangGraph's workflow orchestration. Get personalized recipe recommendations, smart product suggestions, and lightning-fast responses through semantic caching.
Restaurant Discovery AI Agent	An Agentic AI restaurant discovery platform that combines Redis's speed with LangGraph's intelligent workflow orchestration. Get personalized restaurant recommendations, make reservations, and get lightning-fast responses through semantic caching.
Podcast Chatbot with Agent Memory	A chatbot that discusses and recommends podcasts. Affectionately called Podbot, it demonstrates how to use Redis Agent Memory Server to build context-aware agents with Azure Managed Redis.
Car Dealership Assistant with Agent Memory	A Car Dealership chatbot with Redis Agent Memory Server, showcasing short-term and long-term memory.
Multi-Agent AI Text Adventure Game with Agent Memory and LangGraph.js	A text-adventure game that demonstrates how to build a multi-agent AI workflows with LangGraph.js, Azure Managed Redis, and Agent Memory Server.

Recipes

Need quickstarts to begin your Redis AI journey?

Getting started with Redis & Vector Search

Recipe	GitHub	Google Colab
🏁 Redis Intro - The place to start if brand new to Redis
🔍 Vector Search with RedisPy - Vector search with Redis python client
📚 Vector Search with RedisVL - Vector search with Redis Vector Library
🔄 Hybrid Search - Hybrid search techniques with Redis (BM25 + Vector)
🔢 Data Type Support - Shows how to convert a float32 index to float16 or integer dataypes
📊 Benchmarking Basics - Overview of search benchmarking basics with RedisVL and Python multiprocessing
📊 Multi Vector Search - Overview of multi vector queries with RedisVL
🗜️ HNSW to SVS-VAMANA Migration - Showcase how to migrate HNSW indices to SVS-VAMANA
🗜️ FLAT to SVS-VAMANA Migration - Showcase how to migrate FLAT indices to SVS-VAMANA

Retrieval Augmented Generation (RAG)

Retrieval Augmented Generation (aka RAG) is a technique to enhance the ability of an LLM to respond to user queries. The retrieval part of RAG is supported by a vector database, which can return semantically relevant results to a user's query, serving as contextual information to augment the generative capabilities of an LLM.

To get started with RAG, either from scratch or using a popular framework like Llamaindex or LangChain, go with these recipes:

Recipe	GitHub	Google Colab
🧩 RAG from Scratch - RAG from scratch with the Redis Vector Library
⛓️ LangChain RAG - RAG using Redis and LangChain
🦙 LlamaIndex RAG - RAG using Redis and LlamaIndex
🚀 Advanced RAG - Advanced RAG techniques
🖥️ NVIDIA RAG - RAG using Redis and Nvidia NIMs
📊 RAGAS Evaluation - Utilize the RAGAS framework to evaluate RAG performance
🔒 Role-Based RAG - Implement a simple RBAC policy with vector search using Redis

LLM Memory

LLMs are stateless. To maintain context within a conversation chat sessions must be stored and re-sent to the LLM. Redis manages the storage and retrieval of message histories to maintain context and conversational relevance.

Recipe	GitHub	Google Colab
💬 Message History - LLM message history with semantic similarity
👥 Multiple Sessions - Handle multiple simultaneous chats with one instance

Semantic Caching

An estimated 31% of LLM queries are potentially redundant (source). Redis enables semantic caching to help cut down on LLM costs quickly.

Recipe	GitHub	Google Colab
🧠 Gemini Semantic Cache - Build a semantic cache with Redis and Google Gemini
🦙 Llama3.1 Doc2Cache - Build a semantic cache using the Doc2Cache framework and Llama3.1
⚙️ Cache Optimization - Use CacheThresholdOptimizer from redis-retrieval-optimizer to setup best cache config
🎯 Context-Enabled Caching - Context-aware semantic caching with Redis for enhanced LLM performance

Semantic Routing

Routing is a simple and effective way of preventing misuse with your AI application or for creating branching logic between data sources etc.

Recipe	GitHub	Google Colab
🔀 Basic Routing - Simple examples of how to build an allow/block list router in addition to a multi-topic router
⚙️ Router Optimization - Use RouterThresholdOptimizer from redis-retrieval-optimizer to setup best router config
🛠️ Advanced Routing - Strategies for improving routing accuracy when deployed in production
📊 Routing Visualization - Visualize the semantic router and measure the quality of your route references.

AI Gateways

AI gateways manage LLM traffic through a centralized, managed layer that can implement routing, rate limiting, caching, and more.

Recipe	GitHub	Google Colab
🚪 LiteLLM Proxy - Getting started with LiteLLM proxy and Redis

Agents

Recipe	GitHub	Google Colab
🕸️ LangGraph Agents - Notebook to get started with lang-graph and agents
👥 CrewAI Agents - Notebook to get started with CrewAI and lang-graph
🧠 Memory Agent - Building an agent with short term and long term memory using Redis
🛠️ Full-Featured Agent - Notebook builds full tool calling agent with semantic cache and router
🥗 Autogen Agent - Builds a blog writing agent with Autogen and Redis memory
🧩 NeMo Agent Toolkit + Redis - NVIDIA NeMo Agent Toolkit agent with self-hosted Redis Agent Memory
☁️ NeMo Agent Toolkit + Redis Cloud - Same agent backed by managed Redis Cloud Agent Memory

Computer Vision

Recipe	GitHub	Google Colab
👤 Facial Recognition - Build a facial recognition system using the Facenet embedding model and RedisVL

Recommendation Systems

Recipe	GitHub	Google Colab
📋 Content Filtering - Intro content filtering example with redisvl
👥 Collaborative Filtering - Intro collaborative filtering example with redisvl
🏗️ Two Towers - Intro deep learning two tower example with redisvl
🎯 Personalized Content Filtering - Blend query embeddings with user interaction history using RedisVL

Fraud Detection

Recipe	GitHub	Google Colab
🛡️ Fraud Detection - Detect financial fraud with vector k-NN scoring, anomaly detection, and velocity rules using RedisVL

Feature Store

Recipe	GitHub	Google Colab
💳 Credit Scoring - Credit scoring system using Feast with Redis as the online store
🔍 Transaction Search - Real-time transaction feature search with Redis

Model Context Protocol (MCP)

Recipe	GitHub	Google Colab
🤖 RedisVL MCP (Google ADK) - Build an agent with Google ADK, paired with tools from RedisVL MCP Server.
✴️ RedisVL MCP (Claude Agent SDK) - Build an agent with Claude Agent SDK, paired with tools from RedisVL MCP Server.

Redis Agent Memory (RAM)

Recipe	GitHub	Google Colab
💡 Redis Agent Memory - Store and retrieve memories that can be used across conversation sessions.

☕️ Java AI Recipes

A set of Java recipes can be found under /java-recipes.

Notebooks

Notebook	Description
notebooks/RAG/spring_ai_redis_rag.ipynb	Demonstrates building a RAG-ba sed beer recommendation chatbot using Spring AI and Redis as the vector store
RAG/spring_ai_redis_rag.ipynb	Demonstrates building a RAG-based beer recommendation chatbot using Spring AI and Redis as the vector store
semantic-routing/1_semantic_classification.ipynb	Demonstrates how to perform text classification with vector search (RedisVL) instead of LLMs
semantic-routing/2_semantic_tool_calling.ipynb	Demonstrates how to perform tool selection with vector search (RedisVL) instead of LLMs
semantic-routing/3_semantic_guardrails.ipynb	Demonstrates how to implement guardrails with vector search (RedisVL)
semantic-caching/1_pre_generated_semantic_caching.ipynb	Demonstrates how to perform pre generated semantic caching with RedisVL
semantic-caching/2_semantic_caching_with_langcache.ipynb	Demonstrates how to perform pre generated semantic caching with LangCache

Applications

Application	Description
applications/agent-long-term-memory	Demonstrates how to implement long-term memory for AI agents using Spring AI Advisor abstraction with Redis Vector Search
applications/agent-short-term-memory	Demonstrates how to implement short-term memory for AI agents using Spring AI ChatHistory abstraction
applications/vector-similarity-search/redis-om-spring	Demonstrates building a vector similarity search application using Spring Boot and Redis OM Spring
applications/vector-similarity-search/spring-ai	Demonstrates building a vector similarity search application using Spring Boot and Spring AI

Tutorials

Need a deeper-dive through different use cases and topics?

Integrations

Redis integrates with many different players in the AI ecosystem. Here's a curated list below:

Integration	Description
RedisVL	A dedicated Python client lib for Redis as a Vector DB
AWS Bedrock	Streamlines GenAI deployment by offering foundational models as a unified API
LangChain Python	Popular Python client lib for building LLM applications powered by Redis
LangChain JS	Popular JS client lib for building LLM applications powered by Redis
LlamaIndex	LlamaIndex Integration for Redis as a vector Database (formerly GPT-index)
LiteLLM	Popular LLM proxy layer to help manage and streamline usage of multiple foundation models
Semantic Kernel	Popular lib by MSFT to integrate LLMs with plugins
RelevanceAI	Platform to tag, search and analyze unstructured data faster, built on Redis
DocArray	DocArray Integration of Redis as a VectorDB by Jina AI

Other Helpful Resources

• Vector Databases and Large Language Models - Talk given at LLMs in Production Part 1 by Sam Partee.
• Level-up RAG with RedisVL
• Improving RAG quality with RAGAs
• Vector Databases and AI-powered Search Talk - Video "Vector Databases and AI-powered Search" given by Sam Partee at SDSC 2023.
• NVIDIA RecSys with Redis
• Benchmarking results for vector databases - Benchmarking results for vector databases, including Redis and 7 other Vector Database players.
• Redis Vector Library Docs
• Redis Vector Search API Docs - Official Redis literature for Vector Similarity Search.
• Redis Retrieval Optimizer - Library for optimizing index, embedding, and search method usage within Redis.

Contributing

We welcome contributions to Redis AI Resources! Here's how you can help:

1. Add a new recipe: Create a Jupyter notebook demonstrating a Redis AI use case
2. Improve documentation: Enhance existing notebooks or README with clearer explanations
3. Fix bugs: Address issues in code samples or documentation
4. Suggest improvements: Open an issue with ideas for new content or enhancements

To contribute:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

Please follow the existing style and format of the repository when adding content.