Neocloud

Packet.ai

The hosted.ai team's GPU cloud: no-contract A100 capacity in US and EU regions with a 99% SLA.

Built by the team behind hosted.ai, Packet.ai sells on-demand A100 80GB capacity with no contracts, US and EU regions, and straightforward monthly pricing, with H100 capacity on the roadmap. A clean mid-market option for teams that want committed-grade reliability without committed-grade paperwork.

Visit Packet.ai ↗

GPU pricing on Packet.ai

On-demand list per GPU-hour, last reviewed 2026-09. Methodology.

GPU $/GPU-hr vs cheapest Notes
RTX 4090 $0.39 cheapest Dedicated
RTX PRO 6000 $0.66 1.7× Dynamic (shared) tier
L40S $0.92 2.4× Dedicated
A100 80GB $1.43 3.7× No-contract on-demand; monthly from $940
B200 $3.75 9.6× Dynamic (shared multi-tenant) tier; dedicated $6.99

Strengths

  • No contracts, per-hour or monthly billing
  • US and EU regions with a stated SLA

Watch for

  • Catalog currently centers on the A100
  • H100 still waitlisted

Alternatives

CoreWeave

The flagship neocloud: Kubernetes-native GPU fleets built for training clusters, now public company.

Lambda

The ML researcher's default: one-click clusters, sane pricing, no quota theater.

RunPod

Community and secure cloud tiers, per-second billing, and serverless GPU endpoints.