On 6, January 2026, ArcherMind joined hands with Lenovo to jointly launch Auto AI Box—a cockpit AI computing solution built on, NVIDIA DRIVE AGX Thor, NVIDIA’s latest in-vehicle computing platform. Auto AI Box integrates Lenovo’s profound hardware capabilities in in-vehicle computing and is equipped with ArcherMind’s FusionOS 4.0, the latest Agentic-based AIOS. It incorporates edge-side multimodal large models, supports natural language interaction, and provides standard interfaces, enabling agentic AI scenarios for the cabin. Additionally, it supports the smooth operation of multimodal large models with up to 13 billion parameters (13B), delivering powerful, efficient, and scalable core computing support for the next-generation intelligent cockpits.
Leveraging multiple advantages brought by its integrated hardware-software design, Auto AI Box accelerates the in-depth integration of edge-side multimodal large model scenarios into the cockpit:
Out-of-the-Box System Foundation
Built on the resource management and security framework of NVIDIA DriveOS, Auto AI Box is equipped with ArcherMind’s FusionOS 4.0 (AIOS) middleware, providing a one-stop, out-of-the-box hardware-software environment for AI scenario development, model deployment, and optimization.
Plug-in Design
The Auto AI Box allows OEMs to deploy state of the art Cockpit AI without modifying their EE architecture. The Auto AI-Box is an independent platform that can be flexibly deployed alongside any cockpit system. Since the Cockpit AI inference are fully decoupled from the infotainment SoC, the Auto AI-Box can be either designed in, or a drop-in addition to an existing infotainment platform.
Stable Hardware with Multiple Specifications
Auto AI Box provides multiple options for different customer requirements. Both aircooling and watercooling cooling form factors are supported. To support different Cockpit AI usecases, the platform can modularly scale to multiple tiers of compute power and memory bandwidth.
Rich Agent Reference Designs
The solution has been adapted to and pre-integrated with a number of mainstream domestic and international large models as well as supporting edge AI application scenarios. It also supports multiple languages and CP/SP (Content Provider/Service Provider), accelerating OEMs’ global promotion of AI across their vehicle models.
Henry Zou, Co-President of ArcherMind and Chairman of Arraymo, said:
ArcherMind/Arraymo latest AI SW middleware, FusionOS 4.0, integrates ArcherMind’s experience in mass-produced AIOS-equipped vehicles, featuring a fully optimized design tailored to edge AI requirements. It incorporates ArcherMind’s technical accumulation in model optimization and is paired with a self-developed Agent development toolchain, helping customers rapidly implement edge AI scenarios. With solutions developed on the NVIDIA DRIVE AGX Thor platform, ArcherMind has seamlessly integrated its software capabilities with Lenovo Vehicle Computing’s hardware resources. The two parties will jointly provide customers with the Auto AI BOX solution with optimal performance, and work together to drive a leap forward in computing power and scenario development for edge-side large models, delivering an enhanced intelligent cockpit experience to users.
Donny Tang, Vice President of Lenovo Group and Head of Lenovo Vehicle Computing, said:
“The joint launch of Auto AI Box with ArcherMind marks an important step in Lenovo Vehicle Computing’s platform and ecosystem strategy for in-vehicle AI computing. By working closely with ArcherMind Technology’s operating system and in-vehicle AI software expertise, we aim to accelerate the deployment of large models and agentic AI applications in the cabin, delivering a secure, efficient, and scalable computing foundation for OEMs.”
“The NVIDIA DRIVE AGX Thor platform is designed to provide a scalable, high-performance foundation for next-generation intelligent vehicles,” said Rishi Dhall, Vice President of Automotive at NVIDIA. “By building on DRIVE AGX Thor and NVIDIA DriveOS, ArcherMind and Lenovo are enabling a flexible, decoupled approach to deploying advanced, multimodal AI experiences in the cockpit, helping automakers accelerate development while maintaining safety, efficiency, and scalability.”
[About ArcherMind]
Founded in 2006 and headquartered in Nanjing, ArcherMind focuses on the R&D and innovation of intelligent connectivity and operating system technologies, and is committed to becoming a world-leading expert in intelligent operating systems. The company concentrates on technology R&D and solution services in fields such as intelligent automobiles, mobile terminals, artificial intelligence, mobile internet & internet of things (IoT), and information technology application innovation (ITA), and is a trusted partner of many Fortune 500 clients.
Arraymo is a subsidiary of ArcherMind, specializing in cutting-edge technology research in the intelligent automobile field and providing one-stop software solutions for a number of global clients.
[About Lenovo]
Lenovo (HKSE: 992) (ADR: LNVGY) is a US$62 billion revenue global technology powerhouse, ranked #217 in the Fortune Global 500, employing 77,000 people around the world, and serving millions of customers every day in 180 markets. Focused on a bold vision to deliver smarter technology for all, Lenovo has built on its success as the world’s largest PC company by further expanding into growth areas that fuel the advancement of ‘New IT’ technologies (client, edge, cloud, network, and intelligence) including server, storage, mobile, software, solutions, and services. This transformation together with Lenovo’s world-changing innovation is building a more inclusive, trustworthy, and smarter future for everyone, everywhere.