◇ gpurentalsPREVIEW

GPU RENTALS / COMPUTE FOR YOUR NEXT IDEA

Big ideas.
Meet compute.

Your models. Your workflows. One focused workspace to plan GPU compute, explore templates and manage your next build.

Early access · Save configurations today. Live GPU deployment is not yet available.

PyTorchJupyterLabComfyUIOllama
GPU / COMPUTE
BUILT FOR POSSIBILITY
◇ FROM NOTEBOOK TO NEXT BREAKTHROUGH Concept illustration · not live capacity

CLEAR ECONOMICS

Plan your compute.
Know the cost.

Compare proposed hourly rates, choose a template and save your configuration in the customer console.

STANDARD MANAGED PODSTART HERE

Managed shared compute

Customer pays₹69/ pod hour
Contributor earns₹38/ hour

The remaining ₹31/hr covers infrastructure, platform operations and reserve.

Configure standard GPU ↗

Pilot planning rates · contributor payouts and real customer charges are not enabled yet. Allocation applies only to approved, allocated time; it is not a guaranteed income promise.

THE PILOT

From idle hardware
to a verified connection.

The real contributor path supports NVIDIA, AMD Radeon and Intel telemetry on supported Windows/Linux setups, plus Apple Silicon Macs through Metal.

01

Install & pair

The contributor detects its local NVIDIA, AMD Radeon, Intel or Apple Metal GPU and connects with a one-time administrator code.

02

Approve & control

The administrator verifies the device. The owner can pause, resume or disconnect the machine.

03

Verify safely

Run a platform-specific diagnostic. Arbitrary commands and customer workloads are blocked.

FOR GPU OWNERS

One download. No command line.

Download the app, open it, and enter the one-time pairing code from the administrator.

⊞ WINDOWS PILOT · V0.2.0

Windows contributor

A single Windows app that detects a real NVIDIA, AMD Radeon or Intel GPU, pairs it, and keeps the secure worker connected automatically.

Download Windows app ↓Requires the matching vendor telemetry tool. Early-access app; Windows may show an unsigned-publisher warning. Verify SHA-256.
>_ UBUNTU / DEBIAN · AMD64

Linux contributor

A normal .deb installer with the same pairing screen and secure background worker. No scripts or terminal commands.

Download Linux app ↓Supports NVIDIA, AMD Radeon and Intel where the required vendor telemetry tool is installed. Verify SHA-256.
⌘ APPLE GPU WORKER · METAL

macOS contributor

A native Metal worker detects Apple GPUs, pairs securely, sends heartbeats, and runs a verified Metal compute diagnostic.

The signed .app is built on a Mac with contributor/macos/build-app.sh before public download.