Liquid AI has announced that Liquid Context is now optimized for Snapdragon processors running Qualcomm Hexagon NPU hardware. The software builds a persistent personal context profile directly on the device to help virtual assistants anticipate daily user needs. Device manufacturers now have an efficient foundation to deploy proactive agent software across phones and connected hardware.
Most digital assistants today rely on sending private user data to remote cloud servers for every calculation. Liquid AI takes a different route by processing personal context locally on hardware. Using permitted device signals, the system tracks routines, calendar changes, and user preferences without requiring cloud models for constant background updates. Connected software agents then pull from this local layer to suggest actions or draft messages with direct permission from the user.
Personal AI starts with understanding how you live and what you need, when you need it. Liquid Context builds that understanding on your device so the agents you choose can offer more relevant help and anticipate your needs. The Hexagon NPU makes this continuous, local intelligence practical. Together, we are bringing personal context to the devices people rely on every day.
Ramin Hasani, the CEO and co founder of Liquid AI, outlined this design during the announcement at the Snapdragon Summit. By keeping the context engine running inside the Hexagon NPU, hardware brands can run continuous background learning without draining battery reserves.
The system is built to manage complex real world scenarios across different types of hardware. If a parent receives an urgent message about an early school pickup, the context engine understands family priorities. A connected assistant can immediately scan the calendar, protect critical commitments, and draft reschedule emails for approval.
The software also transfers contextual data between separate hardware products. A user running with a smartwatch can record a personal fitness record during a workout. When they return to their vehicle, that data syncs locally with the car system. The automotive assistant can recognize the completed run, adjust cabin climate settings to cool the driver down, and suggest a stop for a recovery drink.
Alongside the context engine, Liquid AI introduced its companion Liquid Agent software. This tool runs on the compact LFM2.5 2.6B model optimized specifically for the processing architecture of the Hexagon NPU. Device manufacturers can tailor this agent to match their own brand identity, interface style, and local hardware services.
Qualcomm Technologies is at the forefront of advancing AI across devices, edge solutions and datacenters, driving the innovations needed to deliver increasingly intelligent, personalized experiences for consumers around the world. By combining our industry leading Snapdragon platforms with innovative AI technologies from companies like Liquid AI, we are helping accelerate the next generation of Agentic AI experiences.
Durga Malladi, executive vice president at Qualcomm Technologies, confirmed that both companies plan to deepen their collaboration. This partnership provides electronics brands with a standardized path to build private context into future consumer devices.
