PeakVision AI Targets SMBs with Locally Deployed Enterprise Agent Framework
Hangzhou-based PeakVision AI is developing a local computing stack designed to help small and medium-sized businesses deploy AI agents while addressing data security and cost concerns.
Founded in 2024, PeakVision AI has introduced FengKan AIOS, an operating system designed to execute enterprise tasks locally rather than through individual applications. The system focuses on intent-based workflows, where users input goals into a centralized chat interface, allowing AI agents to handle cross-environment execution. According to CTO Liangyu Nie, this model addresses the technical complexity and data security challenges that often deter smaller organizations from adopting enterprise AI.
To compete with established enterprise platforms, PeakVision AI utilizes cost-effective local hardware and partners with firms like AMD and ChangXin Memory Technologies to optimize compute performance and memory costs. The startup is currently transitioning from a research phase to commercial deployment, with plans to make its core AIOS platform open-source by the end of October.
While currently serving sectors such as manufacturing and financial analysis, the company is eyeing global expansion, citing interest from distributors in the U.S., the Middle East, and Southeast Asia. The firm aims to scale its client base to approximately 100 enterprises across 10 industries by 2027.