Optional filters now work with virtual replicas, enabling boosting or burying of specific results after sorting. Previously limited to standard replicas, this capability extends advanced relevance tuning to virtual replica configurations without requiring additional replica indices. Search teams can now apply promotion and demotion strategies across all sorting configurations, combining the efficiency of virtual replicas with fine-grained relevance control. Click here to learn more.

Agent Studio introduces dynamic search parameter passing for more flexible retrieval, along with new analytics for conversation depth and user feedback. It adds DeepSeek support, thread depth controls via API, new templates, and expanded feedback APIs and UI components. DocSearch now supports Agent Studio capabilities, helping improve customization, visibility, and control over AI experiences.

DocSearch v4.5 introduces a new side-panel Ask AI experience available in standalone mode or triggered from search in hybrid mode, enabling richer assistance without leaving the page. This release also adds Agent Studio support, allowing DocSearch to use the same agent architecture as other Algolia products, and expands framework support with VitePress compatibility and upcoming Docusaurus adoption. Ask AI thread depth controls let customers set a maximum conversation depth, improving response accuracy and cost predictability for longer side-panel or full-page conversations. Click here to learn more.

AI Synonyms now suggests only candidates that meet a higher quality threshold. As a result, customers will see fewer recommendations overall, with each suggestion more likely to be relevant and actionable. The updated scoring threshold reflects observed acceptance patterns, where most approved synonyms scored above 0.7 and most rejected ones fell below. Email notifications for AI Synonyms suggestions are now sent monthly instead of weekly, with the option to opt out. This update reduces noise in both the dashboard and inbox.

The latest DocSearch release brings a more powerful, flexible, and conversational experience for both developers and end users. A redesigned Playground Editor offers live previews and deeper configuration control, while DocSearch v4.3 introduces Suggested Questions, a composable API, and improved UI/UX stability. Field Mapping now supports custom data shapes, expanding compatibility beyond crawler-generated content. With enhanced performance and new Ask AI–powered search on Algolia Docs, DocSearch delivers faster setup, richer customization, and smarter discovery. Click here to learn more.

Algolia SiteSearch UI Components introduce a flexible, open-source registry for creating AI-powered search interfaces faster. Built with shadcn on top of InstantSearch, these components simplify implementation across multiple use cases, including Ask AI. Developers and agents can easily customize layouts and interactions, even using generative tools like v0 or Cursor. SiteSearch accelerates development with production-ready UI patterns that reduce engineering effort and enable rapid, tailored search experiences. Click here to learn more.

Ask AI now delivers a more engaging and cost-effective experience. New Suggested Questions help users start conversations faster by surfacing curated, clickable prompts that instantly generate responses. Analytics Enhancements add token-level insights for deeper understanding of conversation flow. Token Efficiency improvements reduce costs by up to 50% while maintaining speed and accuracy. Expanded provider support, including OpenRouter, Anthropic, OpenAI, and Vercel AI SDK v5, ensures compatibility with the latest LLMs. Click here to learn more.

Algolia's Search Analytics now provide data insights at the Composition level. It tracks key metrics such as popular search queries, search volumes, click and conversions rates, or revenue for specific Compositions, all presented in intuitive dashboards that are easy to filter and consume.

Algolia MCP Server enables LLMs and Agents to securely access Algolia’s Search and Recommend APIs through the open Model Context Protocol (MCP). This managed, fully hosted service provides a universal gateway between AI systems and Algolia, simplifying integration and enhancing data-driven AI experiences. Click here to learn more and get started with the Algolia MCP Server. 

Ask AI is now generally available, expanding beyond documentation search to support broader site experiences - across support portals, blogs, marketing pages, and more. With new flexible UI options and streamlined setup, teams can quickly deploy accurate, conversational search powered by Algolia and the LLM of their choice. Click here to learn more.