GIGABYTE has expanded its AI TOP ATOM desktop AI computing platform, adding 64GB of unified memory to the already available 128GB of memory, thus giving more flexibility for developers, researchers, and enterprises. The 64GB memory option will be available starting from October 23. This new configuration is based on the same hardware configuration as the 128GB option but has been developed specifically for local AI development and includes model inference, prototyping, data analysis and application validation. By working with the models and data locally, AI TOP ATOM provides a better control over development resources and project data and limits dependency on cloud computing.
Also Read: Toshiba Introduces Low-Power Dual-Channel Standard Digital Isolators for Industrial Equipment
The platform allows using CUDA-accelerated AI software together with GIGABYTE AI TOP Utility that provides access to the model downloads, inference, and retrieval-augmented generation (RAG). Developers can use open-source models, AI assistants and documents in order to develop knowledge-based applications. GIGABYTE is exploring agentic AI applications through demonstrations integrating NVIDIA Nemotron open models and the NVIDIA NemoClaw open agent blueprint with scientific simulation workflows. The expanded lineup is intended to support AI development across different model sizes, workflows and application scales, from proof-of-concept projects to practical deployments.
