Memories.ai delivers continuous visual comprehension at a single-digit billion parameter scale, operating on the Qualcomm® Hexagon® NPU within Snapdragon® platforms.
By leveraging Snapdragon®'s AI capabilities, Memories.ai is helping unlock a new era of connected, personalized AI experiences that seamlessly enhance user's everyday experiences.”— Vinesh Sukumar, Vice President, Product Management, Qualcomm® TechnologiesSAN FRANSISCO, CA, UNITED STATES, September 24, 2026 /EINPresswire.com/ — SAN FRANCISCO, SEPTEMBER 24, 2026 – Memories.ai, the firm creating the visual memory layer for physical intelligence, disclosed at Snapdragon® Summit 2026 that its efficient video language model stack operates natively on Snapdragon® platforms, allowing continuous on-device perception across phones, PCs, and smart glasses. The first consumer manifestation of that stack is LUCI, a personal AI that indexes, retains, and retrieves a user's own context on their behalf, with the complete perception and retrieval cycle executing locally.
The Computational Challenge of Continuous Perception
Video comprehension has, until now, remained a cloud-based workload. Frontier vision-language models can reach tens or hundreds of billions of parameters and are typically built to reason over visual inputs that are explicitly handed to them—images or video clips sampled and fed into the model. That pattern works well for analyzing content on demand. However, those assumptions begin to falter when a device perceives continuously in real time. A wearable or PC observing a person's day generates hours of visual input daily, under a fixed thermal and power budget, using data that most individuals will not send to a server. Memories.ai's method restructures where computation occurs rather than shrinking a cloud model and hoping it fits. The stack separates perception into two stages with distinctly different cost profiles.
Architecture: Amortized Perception, Lightweight Retrieval
OmniCaptioner is an efficient video language model that runs at capture time. It transforms what a device sees and hears into a compact, structured, temporally grounded description of the moment. This is the sole step that processes raw pixels, and it executes once per moment.
OmniRetriever is a multimodal retrieval model that indexes those descriptions into a queryable personal memory, then resolves natural language queries against it at interactive latency.
The design consequence is that heavy visual understanding is amortized at capture rather than repaid at every query. Everything downstream—indexing, retrieval, and answer synthesis—operates on compact representations rather than on video. This keeps every model in the loop at a single-digit billion parameter scale, instead of the tens or hundreds of billions typical of larger models, making the entire pipeline NPU-resident.
Why This Fits Snapdragon®
Models at this scale, quantized and compiled for the Qualcomm® Hexagon® NPU, fit within the sustained power envelope of a phone, a laptop, and eventually a pair of glasses. That matters for three reasons.
–Continuity: Perception can remain active rather than being triggered on demand. Memory depends on what was observed, not on what the user remembered to capture.
–Privacy by architecture: The index never leaves the device. The cloud is used solely to connect context across a person's devices when they request it, not to store or process the underlying personal content.
–Cost: A continuously running cloud VLM per user is not economically viable. An on-device one is essentially free at the margin.
A full day of continuous memory compresses to a couple of gigabytes on device.
The Use Case: Personal AI
Personal AI is the primary application for on-device perception, and it has been structurally blocked until now. Most current assistants are stateless. Every session begins with zero context, lacking persistent memory of a person's prior activity, conversations, or environment, because the only memory available is what the user types into a chat window.
LUCI closes that gap by providing the model with a persistent, queryable memory built from what a person actually sees, hears, and does. LUCI first shipped on the desktop, indexing on-screen activity in real time. At Snapdragon® Summit, Memories.ai is extending the same architecture across every device a person carries, so a moment captured on one device becomes queryable memory on any of the others, with no centralized cloud index required.
According to Memories.ai founder and CEO Shawn Shen, "The interesting engineering result here is not that we made a model smaller. It is that we moved the expensive part of video understanding to capture time, once, so that everything after it fits on an NPU. The Qualcomm® Hexagon® NPU is what lets that pipeline run continuously and privately, on the device itself. Personal AI is the first place this pays off, but the same stack is what physical intelligence will need generally."
"At Qualcomm® Technologies, we believe the next generation of AI should be personalized, highly capable, and privacy-first. Memories.ai's Luci demonstrates how on-device AI can deliver meaningful experiences that understand context and provide intelligent assistance while keeping personal data under a user's control. By leveraging Snapdragon®'s AI capabilities across phones, PCs, and smart glasses, Memories.ai is helping unlock a new era of connected, personalized AI experiences that seamlessly enhance user's everyday experiences," said Vinesh Sukumar, Vice President, Product Management, Qualcomm® Technologies, Inc.
About Memories.ai
Memories.ai is constructing the visual memory layer for physical intelligence—technology that helps a model understand the physical world by giving it real, persistent memory rather than raw footage it must interpret after the fact. LUCI is Memories.ai's personal AI, built on that same memory engine, granting everyday devices the ability to privately see, remember, and act on a person's behalf.
Snapdragon® is a trademark or registered trademark of Qualcomm® Incorporated. Snapdragon® is a product of Qualcomm® Technologies, Inc. and/or its subsidiaries. Qualcomm® is a trademark or registered trademark of Qualcomm® Incorporated.
Theodora Lee
Memories.ai
+1 425-615-9692
email us here
Visit us on social media:
LinkedIn
Instagram
YouTube
X
