Liquid AI Joins SOAFEE to Scale In-Vehicle AI

Liquid AI is proud to announce its membership in the SOAFEE Special Interest Group. We are joining a global community of industry leaders to accelerate the development of the Software-Defined Vehicle (SDV) through extremely efficient, powerful and scalable AI.

At Liquid AI, we believe in the power of open source and collaborative ecosystems. Our goal in joining SOAFEE is to contribute our expertise to the community, helping Tier 1s and OEMs bridge the gap between cloud-native development and embedded hardware.

Having worked alongside leading global automakers, we understand that the next generation of mobility requires AI that can deliver powerful performance in constrained environments. 

Efficiency is at the core of everything we do at Liquid AI. Our proprietary architecture achieves unmatched efficiency, enabling multimodal models that deliver exceptionally powerful performance in a footprint small enough to be deployed on even the most constrained devices. This combination unlocks markets where hardware constraints, privacy, security and latency constraints are critical barriers to effective AI transformation in the automotive industry. 

By providing the tools for scalable AI, we are committed to helping the industry drive innovation and bring safer, more intelligent vehicles to the road faster.

"Joining SOAFEE is a natural step for Liquid AI. We look forward to working with the community to build the next generation of software-defined vehicles with efficient, reliable, and high-performance AI." — Mathias Lechner, CTO Liquid AI

Our Liquid Foundation Models (LFMs) and developer platform LEAP for building, specializing and deploying on-device AI, are specifically engineered for the edge to address these needs:

• Edge-first AI: Designed and optimized for CPUs, NPUs, and GPUs, delivering sub-second responses for infotainment, navigation, and safety prompts on existing SoCs.
• Real-Time Performance: Fast, reliable, and optimized for the sub-millisecond latency requirements of automotive systems.
• Privacy & Security at the Edge: By processing data locally on edge silicon, we eliminate the latency and security risks of cloud-dependency, ensuring driver data remains private and systems remain responsive even without a cellular connection.

Liquid AI builds efficient, general-purpose AI at every scale. Our ultra-efficient multimodal models enable planet-scale, sustainable deployment of generative AI, dramatically reducing energy use and cost while powering privacy-, latency-, and security-critical applications both within and beyond the cloud.

Using a hardware-in-the-loop approach, Liquid AI co-designs models with modern silicon from day one, optimizing performance across CPUs, GPUs, and NPUs. Through deep partnerships with leading chipmakers, we deliver tightly integrated solutions to OEMs and enterprises, enabling real-time AI on devices and cost-efficient, low-latency AI in data centers. 

Through our developer platform, LEAP, we democratize access to enterprise-grade AI, making it easy to deploy customized, powerful, and efficient AI—from developers just getting started to teams building at global scale. Turning the promise of an AI-powered world into reality.