This repository provides a curated collection of resources, demos, and recipes for leveraging Redis within the AI ecosystem, targeting developers and researchers building AI applications. It aims to simplify the integration of Redis for tasks like Retrieval Augmented Generation (RAG), semantic caching, and vector search.

How It Works

The project showcases Redis's capabilities as a vector database and a general-purpose data store for AI workloads. It highlights integrations with popular AI frameworks like LangChain and LlamaIndex, demonstrating how to implement advanced AI patterns such as RAG, semantic caching, and semantic routing using Redis's vector search capabilities and specialized libraries like RedisVL.

Quick Start & Requirements

• Install/Run: Primarily through Jupyter notebooks (.ipynb files) requiring a Python environment.
• Prerequisites: Python, Redis Stack (or Redis Enterprise), and various Python libraries (e.g., redisvl, langchain, llamaindex, streamlit, openai). Specific demos may require API keys (e.g., OpenAI).
• Resources: Notebooks can be run locally or in cloud environments. Demos may have varying resource requirements.
• Links: Demos, Recipes, Tutorials, Integrations, Docs.

Highlighted Details

• Comprehensive demos for RAG, vector search, and full-stack implementations.
• Recipes cover core AI patterns: RAG (from scratch, LangChain, LlamaIndex), LLM session management, semantic caching, semantic routing, computer vision, and recommendation systems.
• Integrations with key AI libraries and platforms including LangChain, LlamaIndex, LiteLLM, and AWS Bedrock.
• Includes benchmarks and talks on Redis for AI-powered search.

Maintenance & Community

• Actively maintained by the redis-developer community.
• Links to official Redis documentation and resources are provided.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive license suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

The repository is a collection of examples and recipes; it does not provide a standalone application or library. Users need to set up their own Redis instance and manage dependencies. Some demos may rely on external services or API keys.