Funding
Chenghengwei Raises Nearly RMB 200M Series A to Scale Edge AI SoC Output

Key PointsChenghengwei's Series A funding is earmarked to convert early CH37 deliveries into…
View more
- Chenghengwei's Series A funding is earmarked to convert early CH37 deliveries into scaled production and market delivery.
- The CH37 integrates CPU, GPU, NPU and image processors to cut power and system complexity versus multi-chip designs.
- State and private backers are expected to supply customer introductions and ecosystem resources beyond the capital itself.
Chenghengwei announced on November 19, 2026, that it had recently completed a Series A of nearly RMB 200 million. CICC Capital, CRRC Capital, Addor Capital, Xichuang Xingwu and Qi’an Investment participated in the financing for the edge AI chip developer.
The proceeds are designated mainly for daily operations and business expansion, specifically supporting scaled mass production and market delivery of its edge AI system-on-chip (SoC) products. The flagship CH37 series has entered small-batch delivery and is intended to supply compute for robots, drones, edge-computing terminals and smart interactive terminals.
State and Private Investors Back Production Ramp
Chenghengwei describes CICC Capital as a large state-owned private-equity investor and CRRC Capital as a large state-owned industrial investor. Xichuang Xingwu represents Wuxi local capital, Qi’an Investment is private industrial capital, and Addor Capital is a market-oriented investment institution active in Jiangsu’s semiconductor sector.
The company expects the mix of backers to provide resources for production delivery, customer introductions and ecosystem development. Alongside the new capital, that support is intended to connect the production ramp with customer adoption and the development of a partner base around CH37.
CH37 Puts AI Compute and Vision on One Chip
The CH37 series combines an eight-core CPU, 2D and 3D GPUs, a six-core neural processing unit (NPU), a GPGPU for general-purpose computing and image signal processors on a single chip. Chenghengwei says this integration is intended to replace conventional multi-chip designs, reducing power consumption and system complexity while increasing data throughput.
The CH37 product page lists peak AI compute of 64 trillion operations per second (TOPS) using INT8, or 8-bit integer, calculations, with support for FP16, FP32 and FP64 mixed-precision computation. Its six NPU cores are separately rated at up to 48 TOPS, a subsystem figure rather than a second rating for the whole series. The GPGPU provides up to 1 TFLOPS at FP32 and supports CUDA general-purpose computing.
The product page also lists two independent image-signal-processor channels for visible-light and infrared inputs, plus 4K video encoding and decoding at 120 frames per second. Its AI software supports ONNX models, PyTorch, TensorFlow, ONNX Runtime and an end-to-end NPU deployment toolchain, with CNN, DNN, RNN and Transformer model deployment.
From First Silicon to Partner Hardware
Chenghengwei was founded in June 2023 and is headquartered in Wuxi, with branches, subsidiaries or R&D centers in Beijing, Shanghai, Changsha and Shenzhen. Its official development timeline records first CH37 silicon being returned and powered on in December 2025, followed by small-batch production in 2026.
The chip was formally launched in Wuxi on August 13, 2026. At that event, Chenghengwei signed strategic cooperation agreements with Inspur Electronic Information, Guangzhou Gaoneng Computer Technology, Zhejiang Liangchi Construction Technology and Shenzhen Xinchuang Technology covering AI edge computing, smart sensing and embodied intelligence. Guangzhou Gaoneng also presented three CH37-based hardware products, including an industrial-control motherboard and an edge-computing box, with demonstrations spanning security, road monitoring, industrial, commercial and home settings.
Those partner products show how CH37 is being packaged for deployable systems, but the funding announcement shifts the emphasis to commercial execution. The next test is converting early deliveries and cooperation agreements into sustained deployments, with scaled production and market delivery supported by the new capital and the investors’ expected customer and ecosystem resources.












