Feyntec Intelligent Technology (Shanghai) Co., Ltd. has recently finalized its Pre-A round of financing, amassing a total investment in the hundreds of millions of yuan. This round was co-led by Zhidao Capital and Yunqi Capital, with contributions from several other institutions. The cornerstone of Feyntec's competitive edge is its advanced physical engine technology. Its founder, Zhang Lihua, played a pivotal role in developing NVIDIA's PhysX physical engine, and the team boasts world-class R&D capabilities. This March, Feyntec unveiled Fysics, China's inaugural fully independently developed differentiable physical simulation engine, alongside the comprehensive simulation training platform MoziSim, the full-modality physical AI foundation model OmniFysics, and a bidirectional standardized evaluation benchmark system. This suite of products lays a full-stack technological groundwork spanning multiple development phases. Presently, the physical AI sector has emerged as a pivotal avenue for AI's transition from virtual to real-world applications, with overseas tech behemoths significantly increasing their investments. Nevertheless, China's robotics industry faces a notable gap in the autonomy of foundational tools. The widespread adoption of the industry is hindered by numerous challenges, such as physical hallucinations in data-driven models, inadequate stability in reasoning for complex scenarios, diminished practicality when transitioning from simulation to reality, exorbitant costs of acquiring real interaction data, and the difficulty of adapting traditional physical engines to AI training. Feyntec's Fysics engine aims to surmount these obstacles by emphasizing multi-physics unified solving, high-precision simulation, and full-stack collaboration, thereby bridging technological gaps in the industrial chain and propelling China's domestic physical AI technology to the forefront of the global stage.