Scientel Announces Gensonix AI LLM For AMD Radeon Series GPUs

Got story updates? Submit your updates here. ›

Scientel, a U.S.-based systems technology company, has announced the launch of its Gensonix AI LLM (Large Language Model) solution that is designed to run on the smaller members of the AMD Radeon series GPUs. The company claims that the combination of Gensonix AI DB's efficiency and the power of Meta's Llama 3B model, along with AMD's Radeon GPU architecture, makes LLMs practical on small footprint systems, addressing the power constraints faced by data centers hosting these systems.

Why it matters

As Language Learning Models (LLMs) become more popular, there is a growing need for efficient and small-footprint systems that can run on local networks or in the cloud, rather than relying on power-hungry data centers. Scientel's Gensonix AI LLM solution aims to address this need by leveraging the company's own Gensonix NewSQL AI DB and the Llama 3B model from Meta, along with AMD's Radeon GPU architecture, to provide a more efficient and practical LLM solution.

The details

Scientel's LLM systems are designed to operate in conjunction with its Gensonix NewSQL AI DB, which is capable of storing and handling various data types, including Relational, Document, Text, and Vector data, in their native form. This allows for high efficiency and eliminates the need for data conversion. The Llama 3.2 3B model, developed by Meta, is a high-performance LLM designed for efficient, low-latency deployment on small devices like servers and laptops, offering strong capabilities in tasks like summarization, retrieval, and multilingual dialogue. The AMD Radeon RX6800 GPU, one of the smaller members of the AMD Radeon GPU family, is capable of comfortably running the Gensonix AI LLM system.

• Scientel announced the Gensonix AI LLM solution on February 25, 2026.

What they’re saying