Liquid AI and Qualcomm optimized the on-device context layer Liquid Context for Snapdragon processors
Liquid AI and Qualcomm Technologies announced at Snapdragon Summit 2026 that the on-device context layer Liquid Context has been optimized for Hexagon NPU in Snapdragon processors — it is intended to enable device manufacturers to build personalized AI agents with local context processing.
Liquid AI announced at Snapdragon Summit 2026 that its on-device context layer Liquid Context is now optimized for Snapdragon processors, specifically for Qualcomm Hexagon NPU. Development took place in collaboration with Qualcomm Technologies. According to the company, Liquid Context builds and maintains a persistent understanding of the user — their routines, preferences and needs — based on signals from the device, locally and with the user’s permission. It then makes the resulting context available to AI agents, whether they run on the device, in the cloud or in a hybrid combination.
Alongside Liquid Context, the company is also introducing Liquid Agent, its own efficient embedded agent built on the LFM2.5-2.6B model, also optimized to run on Hexagon NPU. According to Liquid AI, the new optimization enables ongoing context updates to be processed in the background without requiring a cloud model to process every update. This gives device manufacturers (OEMs) a foundation for offering personalized and proactive AI experiences, while allowing them to choose their own combination of agents and services.
The companies cite illustrative scenarios such as rescheduling meetings when a family event comes up, automatically putting together a LinkedIn post after a conference, or transferring context from a run recorded on a smartwatch to a car’s onboard system. According to statements from representatives of both companies (Liquid AI CEO Ramin Hasani and Qualcomm Technologies EVP Durga Malladi), the companies are also exploring further opportunities to collaborate on improving the performance of Liquid Context on Snapdragon platforms so that OEMs can offer it as a built-in device feature.
The source text does not specify an availability date for end-user devices, a price, or precise performance or power consumption figures. Details can be found in the source article.
Why it matters
For device manufacturers (OEMs), this is a concrete technology offering — a combination of Hexagon NPU and the Liquid Context software layer (potentially also Liquid Agent) — for building agentic features that process personal context locally instead of sending data to the cloud, which may matter for both privacy and responsiveness. For end users, the impact is indirect for now: this is a platform that manufacturers still need to integrate into specific products, and the source does not name any specific device or availability date.
Two audiences, two different impacts
What this means
For individuals
According to the companies, future devices with Snapdragon processors could offer proactive AI assistants that build an understanding of the user locally on the device and share it with selected agents with permission — for now, this is a technology platform for manufacturers, not a finished feature on a user’s phone.
More practical updates →For a business
Device manufacturers (OEMs) get a ready-made combination of hardware and software — Hexagon NPU and Liquid Context, potentially also the Liquid Agent built on the LFM2.5-2.6B model — for building personalized agentic features without having to send context to the cloud for processing.
DevelopmentCheck the original
Event sources
only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.