Complex Activity Classification
Beyond step-counting — gym exercises, swim strokes, PT movements.
The end of the cloud round-trip.
Variables shouldn't have to phone home to think. GPX runs the AI on the device.
Proven in production
The empirical advantage.
Real products, real numbers, with the AI running the whole time.
The Safety Win
90 Days
on a single charge
A pendant that watches over you, running fall detection and SOS voice recognition continuously, on-device, for 90 days on a single charge. No phone to carry, no cloud to depend on. Protection that never goes dark.
The Women’s-Health Win
>98%
fall & assault recognition
In DimensionNXG’s MAI wearable, GPX10 Pro recognizes falls and assault patterns on-device at >98% accuracy and still runs 14 days on a 200 mAh cell. Detection stays local and private; the radio fires only when there’s a real event to report.
The Industrial Win
2+ Years
on one cell
Predictive maintenance that lives on the machine, continuously analyzing vibration and sound to catch anomalies before it breaks, entirely at the edge. A retrofittable, battery-powered node that runs 2+ years on a single cell, sending alerts, not raw data.
AI MODEL ZOO
One chip. Every wearable AI feature.
The GPX architecture isn't an ASIC — it's a fully programmable AI engine. Here's what you run natively.
Continuous Biomarker Tracking
1D-CNNs for ECG and early anomaly detection without killing the coin cell.
Always-On Safety Monitoring
Fall detection and acoustic safety, always in the background.
Gesture Recognition
Micro-gestures for AR rings and glasses, zero cloud latency.
Bring Your Own Model
Proprietary model? Compile it to GPX with ModelForge.
The on-device advantage
Run any model. Keep all four advantages
CONTINUOUS AI
24/7
Always-on safety detection
Runs assault, anomaly, and motion-event detection continuously on-device, without waiting for a cloud round trip.
POWER EFFICIENCY
<1mW
Microwatt-level inference
Keeps AI models active in the background while consuming a fraction of the power required by conventional edge processing.
SMART TRANSMISSION
Only on event
BLE/LTE wakes only when needed
The device processes locally first, then activates communication only when a meaningful safety or health event is detected.
LOCAL AI
100%
No cloud dependency
Sensitive health and safety signals are processed locally, improving reliability, latency, and user privacy.
The host sleeps. The AI never does.
GPX runs continuous inference in its always-on island while the host processor stays in deep sleep — no wake events, no interrupts, no battery drain.
The Sleeping Host
The host processor remains in deep sleep while the Ambient AI engine handles continuous real-time processing. No wake events. No interrupts. No battery drain.
Continuous Sensing
ECG, motion sensors, and audio streams flow directly into the analog compute array. Raw data is processed at the source — no buffering, no transmission overhead.
Microwatt AI
Live AI processing indicators, telemetry, and µW power consumption prove what was thought impossible: hospital-grade intelligence running on coin-cell power.
From lab to product in months, not years.
Don't spend your first day writing sensor configuration code. The Cranium board comes ready to run out of the box, allowing you to instantly test physical AI models and validate performance on the metal with zero setup required.
01
VALIDATE
Test your TensorFlow/PyTorch models on the «NuraSense» Evaluation Kit and flash demos wirelessly with ApplicationForge.
02
COMPILE
Use «ModelForge» to quantize and compile your models for ultra-low-power A-Cube execution.
03
INTEGRATE
Drop our production «System-on-Modules (SOMs)» into your space-constrained carrier board.
Discuss Your Product Roadmap
Explore how Ambient AI can unlock new capabilities in your wearable product. Strategic planning session with our applications team.
Schedule Strategy Call
Talk to a Hardware Engineer
Be the first to access our upcoming Vision, Sound, and Industrial modules.
Schedule Technical Review