I'm LongbridgeAI, I can summarize articles.On October 28, Hangzhou MicroNano Core Electronic Technology Co., Ltd. (hereinafter referred to as "MicroNano Core") announced the completion of its C2 financing round of over RMB 1 billion. The round was led by Jian tou Investment, with participation from Hangzhou Capital, Kaitai Capital, Xiaochi Capital, Yantai Finance, SDIC Taikang, and other institutions, while several existing shareholders continued to increase their stakes.
MicroNano Core stated that, combined with the RMB 1 billion raised in its C1 round in September, the company secured over RMB 2 billion in financing within two months, reaching a cumulative total of RMB 3 billion in the past six months. The valuation for this round was not disclosed. The funds will be primarily used for product development iterations and supply chain preparations for mass production.
As financing continues to advance, MicroNano Core's first-generation trio of chips is currently in the transition phase from tape-out to customer verification.
Founded in 2021 and incubated by the Peking University Institute of Information Science and Technology in Zhejiang Province, MicroNano Core mainly develops processing-in-memory (PIM) chips designed for artificial intelligence inference.
Traditional computing architectures require frequent data movement between processors and memory, resulting in bandwidth and energy consumption overheads. PIM technology attempts to perform some computations within or near the memory to reduce data movement.
MicroNano Core employs its self-developed 3D-CIM architecture, which integrates DRAM-based 3D near-memory computing, SRAM-based processing-in-memory, and RISC-V heterogeneous computing. Its products target edge devices such as AI smartphones and AI PCs, as well as cloud-side inference scenarios like intelligent computing centers.
The company claims that its SRAM-based PIM architecture can improve computing density by 4 to 6 times and energy efficiency by 5 to 10 times compared to traditional von Neumann architectures in previous tape-out tests. However, public materials do not provide complete specifications for the reference chips or testing conditions.
Regarding product progress, MicroNano Core disclosed that all three chips in its first-generation LP-CIM and PCIe-CIM series have been taped out. They are expected to undergo formal customer testing and initiate mass shipments in early 2027.
Prior to this, the company has collaborated with multiple mobile platform providers, ODM manufacturers, and embodied AI enterprises on algorithm adaptation and solution development, completing multiple rounds of emulator verification. Some projects have entered the stage of in-depth joint debugging. On the cloud side, the company is working with cloud service providers and AI infrastructure vendors to evaluate computing power solutions and conduct scenario-based proof-of-concept (POC) tests.
These efforts lay the foundation for subsequent customer adoption, but emulator verification, solution testing, and formal testing with physical chips remain distinct stages. Currently, the company has not disclosed the specific customers for testing, bulk purchase orders, or revenue scale for the aforementioned three chips.
In the PIM sector where MicroNano Core operates, some companies have already entered the mass production or large-scale shipment phase.
Among them, Houmo Intelligence has laid out related products using SRAM-based PIM technology. In mid-October, it publicly disclosed that its M30 and M50 chips had achieved mass production, with the M50 securing over 150 design wins. Its first 3D-CIM chip, the A20, has also entered production. Zhicun Technology, on the other hand, entered the low-power terminal market with Flash-based in-memory computing technology, disclosing on its official website that cumulative chip shipments have exceeded ten million units.
In the cloud sector, Yizhu Technology is developing PIM chips based on 3D DRAM technology, with related products having completed tape-out. On October 23, Yizhu Technology announced the completion of financing exceeding RMB 1 billion.
While the storage media, product positioning, and commercialization stages vary among these companies, the competition in PIM chips has extended from architectural design and laboratory performance metrics to customer adoption, mass production delivery, and actual shipments.
MicroNano Core's financing lineup includes several companies from the industry chain.
In September, MicroNano Core disclosed details of its C1 round, in which industrial investors such as Huaqin Technology and Meige Intelligent participated. In the official announcement of MicroNano Core's financing, both companies expressed their intention to collaborate around scenarios such as smart terminals, inference servers, AI smartphones, and AI PCs.
Previously, MicroNano Core also disclosed receiving investment support from storage industry chain companies such as GigaDevice and Biwin Storage.
The participation of industrial capital helps the company connect with the storage supply chain and terminal application ecosystems. However, currently available public information is insufficient to confirm whether these companies have placed chip orders with MicroNano Core.
For MicroNano Core, the RMB 3 billion raised in six months provides financial support for chip R&D, supply chain preparation, and customer adoption. Assessing commercialization progress still depends on product delivery.
According to the company's latest disclosed plan, early 2027 will be a key milestone. Whether this financing can support the company's transition from technical R&D to scaled sales depends on the actual pace of formal customer testing and the initiation of mass shipments for the three chips, as well as the actual orders the company ultimately secures.