Cupertino, CA, June 28, 2024 – Aizip, Inc., a technology and market leader in artificial intelligence (AI) for edge applications, today announced ultra-efficient small language models (SLMs) and compact AI agents for a broad range of applications in edge markets.
In recent years, large language models (LLMs) have transformed the AI landscape, endowing AI systems with logic and reasoning capabilities that extend beyond sensing and perception. Retrieval-augmented generation (RAG) enhances LLMs by integrating an external knowledge base and user data. By leveraging LLMs and RAG, we can now build AI agents capable of utilizing tools and performing tasks that follow complex or even ambiguous human commands. However, these advanced AI systems typically require substantial effort to train and significant resources to deploy.
Considerable research has therefore been directed towards developing compact models for on-device edge applications. Such on-device models offer many advantages, including privacy, resilient operation, and cost savings. Several companies have successfully reduced the size of language models for mobile phones, but further minimizing model sizes would allow the use for many applications on a wider array of low-cost edge devices.
Aizip’s mission is to enable pervasive intelligence by building ultra-efficient, robust, and scalable AI models that can be deployed anywhere, anytime. To achieve this, Aizip has pushed the boundaries of efficient AI to uncover key insights such as data-centric efficiency and AI-design automation. Leveraging its expertise in developing efficient and robust edge models, Aizip has created Gizmo, a series of ultra-efficient AI agents based on small language models (SLMs). These models support diverse platforms and applications, including smartphones, automotive systems, wearables, and smart appliances.
Remarkably efficient, Aizip’s SLMs can operate on readily available low-cost MPUs and application processors, such as the Renesas RZ/G2L and RZ/G3S, the MediaTek Genio 350, the Rockchip RK3566, and the Raspberry Pi. The smallest models can support IC devices priced at $10 or less. Despite their compactness, these SLMs and SLM-based AI agents achieve robustness and accuracy comparable to cloud-based large models for domain-specific applications. Utilizing an AI-design automation pipeline, these SLMs and SLM-based AI agents can be rapidly generated for new applications.
Since its founding in 2020, Aizip has embraced the mission “to make everything think,” establishing itself as a leader in tinyML with offerings such as Aizip Intelligent Audio (AIA), Aizip Intelligent Vision (AIV), and Aizip Intelligent Time-series (AIT). The newly developed Aizip Intelligent Language (AIL) reflects Aizip’s commitment to providing low-cost, high-performance solutions that benefit society at large. For additional information, please contact info@aizip.ai.
* * *
About Aizip, Inc.
Aizip, Inc., develops AI models for edge applications. Based in Silicon Valley, Aizip provides models with superior performance, quick turnaround time, and excellent ROI. These models can be used in a wide range of applications for an intelligent, automated, and connected world. For more information, visit www.aizip.ai.