How it works

A GPU is not
one indivisible thing.

Most clouds sell a GPU as "one job, one card", so it sits idle whenever that job waits. Our scheduler treats each card as memory, compute and bandwidth that different workloads stress at different times, and packs compatible work onto it safely.

The scheduler

Most clouds sell you a whole card that sits idle.
We pool it.

Six GPUs running a mixed fleet. Switch to "Reserved GPU" to see the gaps a one-job-per-card cloud leaves empty. Our scheduler fills them with other tenants' work, so the hourly price is up to 47% lower.

packet.ai scheduler
94%utilization
Reserved: you pay for the whole card, gaps included.
Dynamic: gaps filled by other jobs, so the hourly price drops.
The premise

Workloads don't use
all of a GPU at once.

Inference is bound by memory bandwidth, training by compute, a notebook mostly by your typing speed. We schedule on what each one actually uses.

Measured, not reserved

The scheduler tracks how each workload uses memory, compute and bandwidth over time, not just whether a card is occupied.

Isolated in software

GPU calls are mediated per tenant. Memory limits are enforced, and every tenant gets its own container, storage and network.

Rebalanced continuously

As workloads start, burst and stop, placement adjusts so busy tenants aren't stacked on the same card.

Compared

Sharing a GPU,
three ways.

Other ways to share a card trade flexibility for isolation, or the reverse.

NVIDIA MIGNVIDIA MPSpacket.ai Dynamic
How the card is splitFixed hardware slicesConcurrent processesScheduled pool, whole card
Resize without a resetNoYesYes
Per-tenant memory limitsYesNoYes
Uses idle capacityNo, slices sit idleYesYes
Works on any recent NVIDIA cardData-centre cards onlyYesYes
h·ai
Built on hosted.ai

packet.ai runs on hosted.ai's GPU cloud platform, the software GPU providers use to run, share and bill their own fleets. packet.ai is that platform, run by its own team, sold by the minute.

About hosted.ai

Your next run
starts at $0.36.

Sign up, top up, launch. You'll be on a GPU in minutes.

No contractHourly or flat monthlyTalk to a human any time