Redis targets AI application development with the launch of new features, including RAG.

Redis introduces a series of new features to better support AI application development, not only reducing cache costs and improving performance but also enhancing data integration capabilities.

The in-memory database company Redis has launched a range of product updates to support enterprise AI application development. These updates include Redis for AI, Redis Flex, Redis Copilot, and Redis Data Integration. These new features not only offer more efficient AI development tools and flexible infrastructure but also strengthen data integration capabilities to support the rapid and scalable development of applications.

As many databases are adding specialized features to meet the demands of AI application development, Redis is no exception. Redis for AI offers several features for AI application development, starting with support for Retrieval-Augmented Generation (RAG). Combined with the fast Redis in-memory database, it provides a real-time architecture that speeds up the response times of generative AI applications. The new semantic caching feature can quickly retrieve stored answers, reducing the cost of calling large language models.

Redis for AI also includes large language model memory capabilities, enabling personalized user interactions. The proxy memory feature accelerates the complex reasoning capabilities of agents, providing more accurate and faster responses. Redis for AI also incorporates a feature store that can deliver high-speed predictions in production environments with sub-millisecond latency, significantly enhancing model performance.

Following Redis's recent acquisition of the key-value storage engine Speedb, Redis Flex, supported by this engine, is also part of this update. Redis Flex is a service that significantly reduces cache costs. According to the company, developers can get five times the cache capacity for the same price. Previously, Redis's auto-tiering feature was only available for large caches, while small caches often suffered from cache misses due to cost, strategy, and application limitations. Redis Flex, designed for DRAM and SSD storage, further optimizes cache performance, offering a more cost-effective cache service compared to traditional memory solutions.

Redis also offers a free virtual assistant called Redis Copilot to help developers quickly retrieve text messages, generate code snippets and commands, and perform data queries using natural language. Redis Copilot provides a user experience similar to other virtual assistants but is more closely aligned with Redis's functions and needs, allowing developers to get more precise responses.

Additionally, Redis has introduced new data integration features aimed at accelerating application operations. Data integration automates data pipelines, synchronizing data from external databases to Redis. Developers can connect using APIs, which not only simplifies the development process but also improves data reliability.