Monday, December 23, 2024

Neuchips Introduces Plug-and-Play Enterprise GenAI Solution, Viper GenAI PCIe Card, at COMPUTEX 2024

Related stories

Doc.com Expands AI developments to Revolutionize Healthcare Access

Doc.com, a pioneering healthcare technology company, proudly announces the development...

Amesite Announces AI-Powered NurseMagic™ Growth in Marketing Reach to Key Markets

Amesite Inc., creator of the AI-powered NurseMagic™ app, announces...

Quantiphi Joins AWS Generative AI Partner Innovation Alliance

Quantiphi, an AI-first digital engineering company, has been named...
spot_imgspot_img

Neuchips, a leading Artificial Intelligence (AI) Application-Specific Integrated Circuits (ASIC) provider, will be launching its latest product, the Viper GenAI PCIe Card, at COMPUTEX 2024. The plug-and-play enterprise GenAI solution allows for a frictionless upgrade from traditional PCs to AI-enabled systems, empowering enterprises to unlock AI’s full potential.

At the core of the Viper GenAI PCIe Card lies the Viper Generative AI Processor, designed to unleash the full potential of large language models (LLM) by offloading more than 90% of the resources required for generative AI from the CPU for maximum LLM-focused performance. The card’s Embedded Engine also boosts accuracy and efficiency by more than ten times for vector similarity searches within the Retrieval-Augmented Generation (RAG) process, which maintains factual accuracy of LLMs and addresses other issues such as hallucinations.

One of the standout features of the Viper GenAI PCIe Card is its extra 48GB memory capacity, which serves as an on-premise vector database that enterprises can deploy in PCs as well as Network Attached Storage (NAS) systems. This capability supports extensive vector data storage and mitigates data leakage concerns by keeping sensitive information offline, and the combination with the card’s Embedded Engine makes Viper the perfect solution for running RAG-enhanced LLMs.

Furthermore, the card supports multiple open-source LLM models — including Llama2, Llama3, Mistral, Phi2, TAIDE, and Breeze — as well as both English and Traditional Chinese, ensuring its value to a wide spectrum of enterprises.

Also Read: ZeroPoint Technologies Closes Funding Round for Groundbreaking Hardware-Accelerated Memory Compression Technology

The Viper GenAI PCIe Card also provides substantial resource savings for enterprises by eliminating the need for extensive AI expertise and infrastructure investments. Low power consumption and minimal installation requirements ensure a seamless upgrade process without additional costs. The card’s offline usage capability is crucial for enterprises with stringent data security and privacy requirements, allowing them to leverage AI technology without exposing sensitive data to potential breaches. In addition, the ability to integrate with software partners for customized AI applications enhances productivity and efficiency, offering tailored solutions that meet the unique needs of each enterprise.

“The Viper GenAI PCIe Card unlocks new possibilities for enterprises looking to leverage the power of AI without substantial investments in new hardware or infrastructure,” said Ken Lau, CEO of Neuchips. “Its seamless integration and cutting-edge features demolish the barriers that have hindered businesses from fully embracing the potential of large language models, paving the way for unprecedented innovation and operational excellence.”

The Viper GenAI PCIe Card is ideal for enterprise users and government agencies looking to harness the power of AI technology securely and efficiently, and software service providers aiming to add value to their hardware offerings through advanced AI capabilities. When combined with Taiwan’s trusted generative AI model TAIDE, the card becomes an invaluable asset for local and central governments, enabling secure and efficient generation of official announcements, reports, and other critical documents.

Source: PRNewsWire

Subscribe

- Never miss a story with notifications


    Latest stories

    spot_img