Neuchips Partners with Vecow and GSH for Enhanced Data Processing
In a significant announcement ahead of Embedded World 2025, Neuchips, known for its cutting-edge application-specific integrated circuits (ASICs) tailored for AI applications, has joined forces with Vecow and Golden Smart Home (GSH) Technology Corp. This collaboration aims to revolutionize SQL data processing through a secure and efficient AI solution that offers real-time insights from proprietary databases, using natural language queries.
The partnership highlights Neuchips’ commitment to driving innovation in the field of edge AI technology. By utilizing their Viper series cards, which are specifically designed for energy efficiency and offline large language models (LLMs), they are providing businesses with enhanced privacy and security for their data. According to Ken Lau, CEO of Neuchips, “The collaboration with Vecow and GSH represents the future of industrial AI deployment.” He adds that attendees at Embedded World 2025 will see how Neuchips’ Viper AI accelerator cards can substantially enhance Vecow’s robust industrial edge AI computing systems and GSH’s ShareGuru SLM solutions.
As the demand for generating AI applications on-site rises, so too does the need for multimodal LLMs. Joseph Huang, Executive Vice President at Vecow, notes the surge in demand for RAG-based LLM solutions that allow users to access the latest data without the need to retrain models, leading to more accurate, relevant, and high-quality outcomes. This partnership promises to deliver a cost-effective, compact, and energy-efficient AI workstation, outperforming traditional cloud-based GPU solutions.
Addressing Industry Challenges
The collaboration comes at a time when the complexity of databases is increasing, yet SQL expertise remains scarce within organizations. This often leads to significant delays in extracting crucial insights from vast amounts of data. The paired solution leverages the Vecow ECX-3100 RAG Edge AI Inference Workstation, a platform equipped with RAG-activated LLM computing capabilities. Meanwhile, GSH’s ShareGuru QA 2.0 solution, based on the ShareGuru SLM platform, works in tandem with Neuchips’ Viper series AI cards.
This powerful combination revolutionizes the way SQL queries are generated, enabling users to employ natural language for query creation. Not only does this enhance accessibility, but it also reduces the need for SQL specialists, lowering operational costs in the process. The solution’s key advantages include:
- - Maximum Data Protection and Security: Neuchips’ offline cards run the ShareGuru solution locally, providing enhanced security for proprietary data.
- - High Accuracy: Enabled by AI Power-Query validation processes.
- - Superior Performance Efficiency: The Viper series achieves a remarkable energy efficiency of 45 watts, even with models incorporating 12 billion parameters.
Viper Series AI Accelerator Card
Unveiled at COMPUTEX 2024, the Viper series alleviates over 90% of the computational load typically required for generative AI, thereby unlocking the full potential of large language models. Noteworthy features include:
- - An additional 32GB of memory capacity.
- - Native support for BF16 Structured Language Models.
- - The Raptor Gen AI accelerator chip from Neuchips, also revealed at CES 2024.
Looking forward, Neuchips plans to concentrate its efforts on developing energy-efficient, multimodal ASICs to enhance performance capabilities by 2026.
About Neuchips
As a leader in AI ASIC solutions, Neuchips is at the forefront of developing specialized hardware for Deep Learning Recommendation Models (DLRMs) and Large Language Models (LLMs). With a dedicated team of experts, a commitment to innovation, and a strong presence in industry associations, Neuchips is poised to shape the future of AI hardware, ushering in a new era of efficiency and performance. For more information, visit
Neuchips Official.