Edge Runtime

Runs on-site. Safety works even when the internet doesn't.

Xolver runs directly at your facility. AI decisions, safety checks, and logs all happen on-site — so a network drop doesn't stop your machines or compromise safety.

Edge_Live_Feed
E1-Core Status: Ready
Hardware: Under DevelopmentLocal dev
Xolver Edge Preview
console_v0.9.4.dev
System: Active

Joint DiagnosticsArm: Under Development

J0112°
Torque4.2 Nm
Load15%
Temp38°C
J02-45°
Torque8.1 Nm
Load22%
Temp41°C
J0388°
Torque12.4 Nm
Load45%
Temp44°C
J04-10°
Torque2.1 Nm
Load8%
Temp36°C
J055°
Torque1.5 Nm
Load5%
Temp35°C
J0645°
Torque3.8 Nm
Load12%
Temp37°C

End-Effector / Haptics

Finger Position

65%

Grip Force

15.0Newton

Tactile State

Stable

Haptic FB

Active

Xolver Edge Telemetry

NPU Load

82%

Latency

11.2 ms

Active Models

  • - TRN-VLM-Edge_v2
  • - KIN-DET-Layer_v0
  • - SAF-ENF-Module
GeofenceActive
Auth: 0x82f...a1 // Data_Throughput: 1.2 GB/sXolver_Edge_Ref: 2026_PROD_BETA

// Engineer's View: Joint health, contact sensing, and AI telemetry — aggregated and processed entirely on-site.

See inside your machine — in real time.

The Console shows you what's happening inside the robot right now — joint health, safety status, AI activity, and what it can see. Everything runs on-site.

Live 3D workspace view

See a real-time 3D model of the robot arm in its environment. Includes live feedback on how the arm is gripping objects — pressure, contact area, and grip quality — so you can spot problems as they happen.

Joint health

See the internal health of every joint — load, temperature, and power draw — computed from signals on your own hardware. No data leaves the site.

Grip precision

Finger position, gripping force, and slip detection are shown in real time. Catch grip failures before they cause a dropped part or a stopped line.

AI activity

See which AI model is active, how fast it's responding, and how much of the on-site processor it's using — all running locally, no cloud round trip.

Safety state

Obstacle detection, safety boundary checks, and a continuous safety heartbeat keep the machine within its approved zone — even as the environment around it changes.

All signals in one view

Rather than checking five separate displays, Xolver aggregates everything — charts, live numbers, and graphical overlays — into a single on-site console for operators and engineers.

Bring Xolver to your facility.

Robotics arm and hardware platform currently under active development. Early access available for research partners.

FAQ

Why does Xolver run on-site instead of in the cloud?

Machines can't wait for a cloud round trip before every move. Running AI decisions, safety checks, and monitoring on-site keeps response times under 5ms and means a network outage doesn't affect machine safety.

What happens if the internet goes down?

Your machines keep operating safely. Depending on what was approved for the deployment, the system can finish the current move safely, hold position, or stop cleanly — without waiting for a network connection.

Does Xolver replace our existing operator displays?

No. The Edge Console adds a diagnostic and review view alongside your existing plant interfaces. You don't need to replace anything — it sits next to what you already have.

Does camera footage or sensor data leave the site?

No. Raw sensor streams stay on-site. The only information that can go to a cloud service is summary records, audit references, and health signals — never raw video or machine data.

What do the on-site diagnostics show?

Joint health, AI activity, safety state, and environment awareness — all computed from signals on your own hardware, without sending anything off-site.

Can I approve machine actions from the Edge Console?

The console shows you readiness and run records. Physical actions still require approval through the standard deployment flow — the console itself doesn't override safety gates.