The TRUSTA AI Scaler Extended Memory Solution represents a significant step forward in addressing the limitations of GPU memory for AI workloads. By integrating software and hardware, this solution allows enterprises to extend model deployment beyond GPUs alone, utilizing system memory and high-speed SSDs to optimize resource use. This approach is designed to reduce the cost of AI model inference and fine-tuning by more than 50%, making it a compelling option for organizations looking to adopt AI within constrained budgets.
Large language models often require substantial GPU resources, creating barriers to entry and limiting scalability. The TRUSTA solution seeks to address these challenges by dynamically allocating computing resources across GPUs, DRAM, and SSDs. This flexibility not only improves efficiency but also enables more cost-effective AI deployment strategies. The solution is built around the AI Scaler Toolkit, a free and open-source platform that supports mainstream model families such as Llama, Qwen, Mistral, Mixtral, GPT-OSS, DeepSeek, Phi, and Gemma, with ongoing expansion for additional models.
The TRUSTA AI Scaler Extended Memory Solution has been recognized for its innovation, earning the COMPUTEX 2026 Best Choice Award in the AI category. This award underscores its potential to reshape enterprise AI infrastructure by providing a more economical and scalable memory configuration that spans GPU, DRAM, and SSD. The solution is now available for download, offering enterprises a practical way to integrate advanced AI capabilities without relying solely on high-end GPUs.
The TRUSTA TD7P51 ECO PCIe Gen 5 enterprise SSD, also showcased at COMPUTEX, complements this initiative with capacities of up to 15.36 TB and support for multiple form factors, including U.2, E1.S, and E3.S. This SSD incorporates advanced technologies like Flexible Data Placement (FDP), which enhances reliability and stability by intelligently managing data placement.
- AI Scaler Toolkit: A free, open-source platform that extends model deployment to GPU memory, DRAM, and SSDs, reducing costs by over 50%.
- Supported Models: Llama, Qwen, Mistral, Mixtral, GPT-OSS, DeepSeek, Phi, and Gemma.
- SSD Specifications: TRUSTA TD7P51 ECO PCIe Gen 5 with up to 15.36 TB capacity and FDP technology for enhanced reliability.
The solution is designed to support a wide range of AI agent applications, including OpenClaw, NemoClaw, and Hermes Agentic, helping users integrate AI Scaler into more comprehensive Agentic AI workflows. This forward-looking approach positions TRUSTA as a key player in the evolving landscape of enterprise-built and edge AI deployment.
As AI infrastructure continues to grow at a projected CAGR of 26% through 2034, solutions like the TRUSTA AI Scaler Extended Memory Solution will play a crucial role in shaping how enterprises approach AI adoption. By addressing data privacy, regulatory compliance, and cost controllability, this solution offers a practical path forward for organizations seeking to leverage AI without the traditional constraints of GPU-only architectures.
Looking ahead, the availability of such solutions could significantly impact the global memory market, which is projected to reach $1.28 trillion by 2027. The TRUSTA initiative underscores the importance of integrated software-hardware approaches in meeting the demands of modern AI workloads, providing a scalable and flexible foundation for enterprise AI infrastructure.
