Alluxio Reports Strong Q2 with Customer Growth, AI Data Latency Improvements

Alluxio reported results for the second quarter of its 2026 fiscal year, highlighting the launch of Alluxio Enterprise Artificial Intelligence (AI) 3.7, which provides sub-millisecond time to first byte (TTFB) latency for AI workloads accessing cloud storage. The company also noted more than 50% customer growth compared to the previous period, driven by its platform's capabilities in various industries.

The latest release enhances data retrieval performance, significantly reducing latency when accessing AI data stored in the cloud. Alluxio aims to eliminate bottlenecks in AI infrastructure by combining low latency with a robust distributed caching technology, thus improving model training and deployment.

CEO Haoyuan Li emphasized the strong performance and customer engagement during the quarter, stating, “We’ve eliminated one of the most stubborn bottlenecks in AI infrastructure, cloud storage performance.” This focus on performance is supported by Alluxio's achievements in the MLPerf Storage v2.0 benchmark, which reaffirmed the platform’s efficiency in maximizing Graphics Processing Unit (GPU) utilization across various workloads.

The company reported its distributed caching architecture yielded substantial performance gains, showcasing high training throughput for models such as ResNet50 and 3D-Unet, achieving near peak GPU utilization. Organizations in technology, finance, and media sectors have increasingly adopted the platform to improve training throughput, speed up feature store access, and enhance inference workflows.

Alluxio Enterprise AI 3.7 strengthens its role in the AI infrastructure ecosystem and is now accessible for organizations looking to optimize their data handling in hybrid and multi-cloud environments. The growth in market adoption and the platform’s performance results suggest that Alluxio is addressing the rising demand for efficient AI data infrastructure.