Vespa.ai– the company behind the leading platform to build and deploy large-scale, real-time AI applications powered by big data– joined Perplexity to announce the AI-powered answer engine’s shift bringing its search feature in-house. The move will significantly enhance the speed, accuracy, and relevance of search results at a scale only made possible on Vespa’s platform.
“The recipe: 1. Solve Search. 2. Use it to solve everything else,” said Aravind Srinivas, CEO of Perplexity.
Perplexity’s innovative approach of direct, sourced answers to search queries relies on a massive and scalable Retrieval-Augmented Generation (RAG) architecture that can efficiently retrieve and process vast amounts of information from the web, internal databases and users’ personal files. By building on Vespa’s platform, Perplexity delivers accurate, near-real-time responses to more than 15 million monthly users, handling more than 100 million queries each week.
Perplexity has used Vespa.ai’s managed platform to efficiently scale its RAG architecture, ensuring low-latency, real-time retrieval of relevant information from massive datasets. Vespa.ai provides Perplexity with the flexibility, speed, and reliability needed to deliver best-in-class conversational experiences to millions of users worldwide.
“We’re continually expanding Vespa’s capabilities to provide the flexibility, speed, and reliability necessary to deliver best-in-class conversational experiences to millions of users worldwide, and the project with Perplexity has allowed us to show just what the platform is capable of,” said Jon Bratseth, CEO of Vespa.ai. “We’re thrilled to partner with Perplexity to power their search capabilities.”
The new project, which developers from both companies have been working on, is centered on Vespa’s capabilities to provide:
- High-Performance Vector and Text Search: Vespa’s optimized vector search combined with advanced text search and relevance capabilities enable Perplexity to retrieve relevant information from large datasets with exceptional speed and accuracy.
- Scalability and Reliability: Vespa.ai’s distributed architecture ensures seamless scalability and high availability, allowing Perplexity to handle increasing user demand without compromising performance.
- Machine-learned ranking: Vespa.ai’s native distributed machine-learned ranking inference allows Perplexity to combine many signals to deliver state of the art relevance at scale.
- Cost Efficiency: Vespa’s efficient resource utilization helps Perplexity optimize its infrastructure costs while maintaining high performance.
The new project highlights Vespa.ai ability to power complex AI workflows, enabling Perplexity to achieve state of the art relevance at large scale with unparalleled cost efficiency.
This partnership underscores the growing importance of RAG in AI-powered applications and the critical role of high-performance hybrid text and vector search engines in enabling these applications. Vespa.ai’s robust and scalable platform empowers innovative companies like Perplexity to deliver exceptional search experiences and redefine how users access information.
For more such updates, follow us on Google News Martech News