Fixstars Adds Autonomous Optimization Feature for Edge AI Inference to its AI Acceleration Platform "AIBooster"
Enhances Autonomous Optimization for AI Training and Achieves Dramatic Improvement in AI Development Efficiency; Introduces SaaS Plan
IRVINE, Calif., Oct. 10, 2025 /PRNewswire/ -- Fixstars Corporation (TSE Prime: 3687, US Headquarters: Irvine, CA), a leading company in performance engineering technology, today announced the release of the latest version of its AI processing acceleration platform, "Fixstars AIBooster" (hereinafter "AIBooster"), with significantly expanded performance and features.
The latest version of AIBooster features two new functionalities: autonomous optimization via model compression for Edge AI inference, and autonomous hyperparameter optimization for AI training. Furthermore, a new plan offering the performance observation feature as a Software-as-a-Service (SaaS) solution has been introduced to significantly reduce the effort required for initial deployment and allow immediate use. Fixstars will continue to support customers in making their AI development cycles more efficient and faster, enabling AI utilization across various devices.
New Features of the Latest AIBooster
When performing AI inference on local devices such as in-vehicle systems or mobile devices (Edge AI inference), limited computing and memory resources make it difficult to run AI models as is. This previously required extensive trial and error to tune models for each device. AIBooster's new autonomous optimization feature solves this by automatically converting AI models for edge devices while maintaining performance.
The feature targets models built with the common PyTorch framework and converts them into formats optimized for each device using a model conversion backend. The initial release supports NVIDIA TensorRT, optimizing inference in NVIDIA GPU environments. By automating conversion and applying techniques such as quantization and kernel fusion, it dramatically boosts inference speed while reducing development time.
In AI model training, hyperparameters such as batch size and learning rate greatly affect model accuracy and efficiency, but their optimization has traditionally required extensive trial and error. The new autonomous hyperparameter optimization feature integrates tuning across the model, resources, and hardware to maximize overall training performance.
Key features include:
Integrated Optimization: Balances model accuracy and training efficiency while maximizing resource utilization.
Hardware Control: Automatically adjusts CPU core and GPU usage schedules.
Distributed Training Support: Operates efficiently in large-scale environments such as Slurm and Kubernetes.
The system outputs trial data, initial and optimized settings, and improvement rates for review. By visualizing and automating the optimization process, this feature significantly enhances AI training efficiency and knowledge acquisition.
The latest AIBooster introduces a SaaS-based performance observation feature that removes setup complexity and allows users to start immediately.
In the SaaS version, performance data from GPU servers is stored in the cloud and visualized in real time on a unified dashboard. It supports multi-cloud and distributed environments, enabling system-, node-, and job-level monitoring on one screen. Fixstars ensures SaaS security to protect customers' AI data, while on-premise installation remains available for flexible deployment.
About Fixstars Corporation
Fixstars accelerates AI inference and training through advanced software optimization. Its technologies drive innovation across industries including healthcare, manufacturing, finance, and mobility. For more information, visit https://www.fixstars.com/.
Media Contact:
Aki Asahara
(408) 400-3679
402507@email4pr.com
View original content to download multimedia:https://www.prnewswire.com/news-releases/fixstars-adds-autonomous-optimization-feature-for-edge-ai-inference-to-its-ai-acceleration-platform-aibooster-302580385.html
SOURCE Fixstars