Liquid AI and Qualcomm announce Liquid Context, an on-device memory layer for Snapdragon

Liquid AI and Qualcomm announced the Liquid Context layer that offers on-device memory for Snapdragon processors. Details are in our news!

With the new partnership announced at Snapdragon Summit 2026, Liquid AI announced that it has optimized the on-device context layer Liquid Context technology for Snapdragon processors. Leveraging Qualcomm's Hexagon NPU hardware power, the technology processes user data directly on the device without sending it to the cloud, paving the way for artificial intelligence agents to work much more personally, proactively and privacy-oriented.

Artificial Intelligence Creating Personal Memory Without Going to the Cloud

Whether artificial intelligence assistants can truly help the user depends on how well they know the user's habits, priorities and daily routines. This is exactly where Liquid Context comes into play. The system, which constantly analyzes in-device signals (calendar, notifications, location or sensor data) in line with the permissions given by the user, builds a local memory layer.

This layer serves as a secure memory that all artificial intelligence models can benefit from, whether independent agents running on the device or third-party cloud services. The biggest advantage is that this analysis takes place entirely on the device; In other words, no data is transferred to the cloud for each data update, thus reducing latency and protecting user privacy.

What is Changing in Daily Life?

Emergency Management: For example, when you receive a message from school that your child needs to be picked up early, the system scans your calendar to identify meetings that can be postponed and drafts rescheduling emails with your approval.

Event and End of Day Summaries: After a busy conference or work day, the system filters the context of the day and your photo archive and creates a LinkedIn or social media sharing draft that suits your style.

Seamless Context Between Devices: The data of a run you go out with your smart watch without your phone is transferred to the in-car system when you get in your car; While the vehicle adjusts the air conditioning according to your personal preference, it suggests a route for a drink break on the road.

Ultimate Efficiency with Hexagon NPU and LFM2.5-2.6B Model

Optimized hardware architecture is critical to ensure that an artificial intelligence model constantly running in the background does not consume battery life. Liquid AI specifically optimized the LFM2.5-2.6B model for Qualcomm's Hexagon NPU, which powers both the Liquid Context memory layer and its own embedded AI agent, the Liquid Agent platform. This structure, which combines scalar, vector and tensor calculations with hardware accelerators, provides continuous learning with minimum power consumption.

Thanks to this infrastructure, hardware manufacturers (OEMs) will be able to integrate the Liquid Agent model into their own devices or connect their own artificial intelligence assistants directly to the Liquid Context layer. This development stands out as an important step in transforming artificial intelligence from a passive chat bot into a proactive assistant architecture that can anticipate user needs.