The Institute of Microelectronics of the Chinese Academy of Sciences has made significant progress in non-volatile computing-in-memory chip research, addressing critical challenges in modern computing. As intelligent computing and big data technologies advance, there is a surging demand for data-intensive operations characterized by low computational intensity and high memory access frequency. Non-volatile computing-in-memory technology, which enables data storage and processing at the same location, has attracted widespread attention. However, a major hurdle remains the difficulty in synchronizing memory capacity with computational power to handle diverse tasks effectively. To solve this issue, the research team focused on two main areas: in-memory computing and near-memory computing. For in-memory computing, they designed a macro chip based on charge-trapping transistors within a hybrid domain. The team investigated high-density differential gain-based arrays, high-energy-efficiency analog prediction and digital computation mechanisms, and circuits for unified fixed-point and floating-point data processing with low hardware overhead. This technology was successfully verified through tape-out. It can efficiently support INT4/8 and FP4 matrix-vector calculations, achieving simultaneous improvements in both storage density and computational density for non-volatile computing-in-memory chips. In the realm of near-memory computing, the team developed a chip based on a new type of ferroelectric NAND flash. Key innovations include ferroelectric NAND flash arrays utilizing post-gate ferroelectric field-effect transistors, a charging and discharging alternating read scheme, and polarity self-converting sense amplifier circuits. Additionally, they created low-overhead, multi-functional phase-domain near-memory computing units. This approach was also verified through tape-out and can efficiently handle high-dimensional, high-parallelism, and multi-bit vector approximate search tasks. These advancements provide a new pathway for developing large-capacity NAND-type near-memory technologies. This research was supported by the National Natural Science Foundation of China and the Strategic Priority Program of the Chinese Academy of Sciences. The breakthroughs mark a pivotal step toward resolving the bottlenecks between storage and processing speeds, offering a practical solution for the next generation of data-intensive computing systems. By demonstrating that high-density storage and powerful computing can be integrated on non-volatile chips, the team has opened new possibilities for efficient data processing architectures that were previously difficult to realize.