Lingsi Raised ₽35 Billion for On-Device AI Chips

Lingsi Tech, a Chinese startup specializing in microchips for local AI inference, raised funding in a Series B round totaling approximately 500 million yuan (roughly ₽35 billion). Investment came from state investment funds from Anhui Province and Hefei City, as well as from venture capital funds: Deep Report Capital, Sky Intelligence Ventures, Scientia Venture, Yingke Capital, Dongrui Capital, and Yongxin Ark. Taihé Capital was appointed as the financial advisor to the company.

Where the Investments Will Go

Lingsi will direct the attracted funds toward developing the next generation of chips for edge AI inference — running neural network models on peripheral devices. Specifically, the company plans to evolve from its current specialization in chips for perception models (image recognition, sensor data processing) to supporting large language models and reasoning systems that execute directly on the user's device. This enables users to run powerful AI systems locally, without relying on cloud connectivity.

• Series B round: approximately 500 million yuan (≈₽35 billion)
• Primary investors: state funds from Anhui and Hefei
• Current direction: edge inference for LLM and cognitive-level models
• Product evolution: from perception models to cognition-oriented chips

Why This Matters for the AI Market

Edge computing chips are winning market share as cloud computing faces challenges with latency, privacy, and cost. Users and corporations demand that critical data and models run locally. Lingsi competes with major players like Qualcomm, Apple (Neural Engine), and NVIDIA, but targets the Asian market and specialized applications. The attraction of state investment demonstrates that edge AI is viewed as a strategic area even at the government level.

What It Means

Growth in edge AI inference investment signals the establishment of local AI inference as a separate segment of the microchip market. Instead of centralized cloud — distributed processing. For developers, this means new optimization libraries and APIs for running models on devices. For end users — faster and more private AI.