Super Micro Computer (SMCI) Stock: Drops 10% as Company Unveils New AI Solution With IBM and NVIDIA

TLDR

• SMCI stock plunges 10.03% to $26.63 despite a major AI partnership announcement.
• Supermicro joins IBM and NVIDIA to launch a new enterprise AI storage platform.
• KV cache testing delivers up to 56 times faster first-token response speeds.
• New shared storage technology aims to boost GPU efficiency and AI throughput.
• Supermicro says its deployment services can cut installation from weeks to days.

Super Micro Computer (SMCI) shares plunged 10.03% to $26.63, losing $2.97 amid heavy selling pressure. Meanwhile, Supermicro announced a new AI storage solution through its collaboration with IBM and NVIDIA. The technology aims to accelerate AI inference, improve GPU efficiency, and support expanding enterprise workloads.

Super Micro Computer, Inc., SMCI

Supermicro Partners With IBM and NVIDIA on AI Storage Solution

Supermicro introduced a new KV cache storage platform that combines technologies from three major infrastructure providers. The system integrates Supermicro Petascale all-flash servers, IBM Storage Scale software, and NVIDIA Dynamo. The architecture uses AMD EPYC 9005 processors and NVIDIA Spectrum-X Ethernet networking.

The collaboration targets growing demand for faster AI inference across enterprise applications and cloud infrastructure. Modern AI systems process larger datasets and longer conversations, increasing pressure on existing memory resources. Companies require storage systems that handle complex workloads without reducing processing efficiency.

Supermicro designed the platform to extend KV cache capacity beyond local GPU memory. This approach allows AI systems to reuse previously processed information instead of repeating calculations. Moreover, shared storage helps organizations expand inference capacity while maintaining consistent application performance.

New KV Cache Technology Delivers Faster AI Inference

Supermicro and NVIDIA tested the technology using a large language model containing 120 billion parameters. Their testing recorded up to 56 times faster time-to-first-token performance with 130,000-token prompts. The results demonstrated substantial improvements when the system reused cached information.

During testing, input lengths increased from 30,000 tokens to 130,000 tokens. However, response preparation time increased only 0.377 seconds with caching, compared with 29.55 seconds without caching. These findings highlighted the technology’s potential benefits for applications processing extensive information.

The platform also supports higher GPU utilization by reducing unnecessary computation and processing delays. Furthermore, its shared architecture allows multiple GPU clusters to access stored context efficiently. The companies validated the system under heavy workloads and competing network traffic conditions.

Supermicro Targets Enterprise AI Infrastructure Growth

The announcement expands Supermicro’s infrastructure portfolio as businesses increase spending on AI deployment and inference services. Enterprise applications increasingly use coding assistants, automated workflows, and systems that process lengthy documents. These applications require reliable storage capacity alongside powerful computing resources.

Supermicro also offers factory rack integration and deployment services for the new platform. According to the company, these services can shorten deployment periods from weeks to days. Meanwhile, IBM contributes storage management capabilities, while NVIDIA supplies inference software and networking technology.

The partnership strengthens Supermicro’s position in enterprise AI infrastructure despite the stock’s sharp decline. Its latest solution addresses computing efficiency, storage scalability, and system reliability across demanding applications. However, the announcement provides no financial forecasts or confirmed revenue contributions from the new platform.

The post Super Micro Computer (SMCI) Stock: Drops 10% as Company Unveils New AI Solution With IBM and NVIDIA appeared first on Blockonomi.