NVIDIA launches multilingual AI service to power custom chatbots

NVIDIA’s new NeMo Retriever microservices represent a significant advancement in multilingual AI capabilities, offering enterprises tools to process and retrieve information across multiple languages while dramatically improving efficiency and scalability.

Core technology introduction: NVIDIA has launched NeMo Retriever embedding and reranking microservices on their API catalog, designed to enhance multilingual information retrieval for enterprise AI systems.

• The technology enables AI systems to understand and process information across multiple languages and document formats
• New capabilities include 35x improved data storage efficiency through long context support and dynamic embedding sizing
• The system connects generative AI to enterprise data across major global languages

Key implementation showcase: Wikimedia’s implementation of NeMo Retriever, in partnership with DataStax, demonstrates the technology’s transformative potential.

• Vector embedding of over 10 million Wikidata entries was completed in under three days, compared to the previous 30-day timeline
• The system now enables real-time updates for hundreds of thousands of daily entries
• Near-zero latency and exceptional scalability support the dynamic demands of Wikipedia’s global user base

Industry partnerships and integration: Major technology companies are rapidly adopting NeMo Retriever to enhance their existing AI capabilities.

• Cloudera is expanding its AI Inference Service with NeMo Retriever for improved multilingual insights
• Cohesity has introduced an AI-powered conversational search assistant using the technology
• SAP is implementing the system to enhance its Joule copilot Q&A feature
• VAST Data and WEKA are integrating the technology for enhanced real-time analysis and multimodal AI solutions

Technical capabilities and applications: The system addresses critical enterprise needs through advanced features and optimizations.

• Long-context support enables processing of extensive documents while maintaining accuracy
• The technology supports various enterprise use cases including search, question-answering, and recommendation systems
• Storage optimization allows large knowledge bases to operate on a single server

Future implications: While NeMo Retriever’s immediate impact on multilingual AI processing is significant, its broader influence on global business communications and data accessibility suggests a fundamental shift in how enterprises will approach international operations and information management.