Gasgoo Munich- SiEngine announced on August 12 that its self-developed automotive-grade 7-nanometer AI acceleration chip, the "TianGong 100" (NNA100), has entered full mass production. The company has simultaneously begun batch deliveries to OEMs, Tier 1 suppliers, and other ecosystem partners, according to Gasgoo.
Image Source: SiEngine (same below)
Designed as a plug-and-play, independent automotive-grade accelerator for large models, the TianGong 100 targets edge-side AI applications. It aims to provide a complete domestic computing foundation for the rapid deployment of local large models, addressing the current bottleneck in expanding edge-side computing power.
The chip's innovation lies in its independent, add-on automotive AI accelerator solution. It is reportedly the only specialized AI accelerator in China that requires no restructuring of the vehicle's electronic/electrical (E/E) architecture and is compatible with cockpit main controllers across all brands. This capability allows both new and existing models to upgrade their AI computing power with minimal deployment—no modification to original infotainment hardware or software is required—significantly shortening delivery cycles.
On the performance and safety front, the TianGong 100 delivers 96 TOPS of INT8 computing power and comes standard with 102GB/s of dedicated LPDDR5 memory bandwidth. This ensures AI inference resources are isolated, avoiding strain on the vehicle's basic computing power. The chip smoothly supports 3B to 7B large models or multiple 4B models running simultaneously. It has also achieved AEC-Q100 Grade 3 and ISO 26262 ASIL-B automotive certifications, featuring a built-in hardware security module to protect user privacy.
SiEngine emphasizes that by offering hardware costs in the thousand-yuan range and leveraging a highly open software and hardware ecosystem, the chip can cut total vehicle deployment costs by at least half compared to high-compute dual-control AI BOX solutions. Its hardware compatibility is robust, seamlessly integrating with mainstream domestic and international chips—including SiEngine's own "Longying 1" and "Xingchen 1." It covers passenger car platforms from entry-level to mid-to-high-end and can be extended to general edge-side AI scenarios like industrial control and robotics.
Dr. Wang Kai, founder and CEO of SiEngine, said the TianGong 100 aims to help automakers and Tier 1 partners bypass the high barriers to large model deployment. The goal is to accelerate the adoption of 7B-and-above models across more vehicle models and intelligent agents, driving the move toward autonomous and controllable AI computing power.
