ComputeAssay

GPU price check · updated 2026-09-10

H100-SXM rental prices

H100-SXM rents for $1.99–$12.29 per GPU-hour on demand across 12 sellers this week. The spread is not noise — it is mostly the network fabric, which determines what the chip can actually do.

$1.99
lowest on-demand
$3.99
median
6
Grade A/B offerings
0
unverifiable listings

Fair-price band (verifiable training-class, Grade A/B): $3.85–$6.52 per GPU-hour on demand. Below that band, ask what the seller is not telling you about the fabric; above it, you are paying for something the rate card should be able to name.

Every listing we track

ProviderSKU$/GPU-hrTermsGradeFabric (as published)Src
Voltage ParkHGX H100 on-demand$1.99on-demandC3,200 Gbps NVIDIA Quantum-2 InfiniBand available (per homepage); Ethernet tier also offered; pricing page does not state which fabric the $1.99 rate applies to
NebiusNVIDIA HGX H100 — preemptible$2.15spotBInfiniBand, 400 Gbps per GPU / 3.2 Tbps per node (per docs.nebius.com)
CoreWeaveNVIDIA HGX H100 (gd-8xh100ib-i128) — Spot$2.46spotBInfiniBand (same instance as on-demand H100)
HyperstackNVIDIA H100 SXM (reserved)$2.72reservedCSR-IOV Ethernet up to 350 Gbps inter-VM (default up to 16 Gbps)
HyperstackNVIDIA H100 SXM$3.20on-demandCSR-IOV Ethernet up to 350 Gbps inter-VM (default up to 16 Gbps)
DataCrunch (rebranded to Verda, verda.com)H100 SXM5$3.25on-demandCnot published
ScalewayH100-SXM (H100-SXM-*-80G)$3.61on-demandCUp to 20 Gbps Ethernet
NebiusNVIDIA HGX H100 (gpu-h100-sxm platform)$3.85on-demandAInfiniBand, 400 Gbps per GPU / 3.2 Tbps per 8-GPU node, GPUDirect RDMA (per docs.nebius.com)
Crusoeh100-80gb-sxm-ib.8x$3.90on-demandBInfiniBand, 3200 Gbps per 8-GPU instance (per docs VM overview)
Lambda8x NVIDIA H100 SXM (on-demand instance)$3.99on-demandCnot published
CoreWeaveNVIDIA HGX H100 (gd-8xh100ib-i128)$6.16on-demandBInfiniBand (instance ID carries 'ib'; fabric generation per HPC interconnect docs is Quantum-2/Quantum-X800 non-blocking fat-tree, but per-node bandwidth for H100 not stated on fetched pages)
Lambda1-Click Cluster — NVIDIA H100$6.16reservedBNVIDIA Quantum-2 InfiniBand with SHARP acceleration (per-node bandwidth not stated on fetched pages)
AWSp5.48xlarge$6.88on-demandAEFA, 3200 Gbps per instance (AWS P5 page)
Oracle Cloud (OCI)BM.GPU.H100.8$10.00on-demand (PAY_AS_YOU_GO)BRDMA cluster network: '8x2x200 Gb/sec RDMA' = 3,200 Gbps per node (Oracle price-list shape table); 'up to 3,200 Gb/sec of RDMA cluster network bandwidth' (Oracle GPU page)
Google Clouda3-highgpu-8g$11.06on-demandCgVNIC with GPUDirect-TCPX, 1,000 Gbps max per VM, 5 NICs (Google accelerator-optimized machines doc table)
Google Clouda3-megagpu-8g$11.62on-demandCgVNIC with GPUDirect-TCPXO, 1,800 Gbps max per VM, 9 NICs — 'doubles the network bandwidth when compared to the A3 High' (Google doc)
Microsoft AzureStandard_ND96isr_H100_v5$12.29on-demandAInfiniBand NDR: 'dedicated, topology-agnostic 400 Gbps NVIDIA Quantum-2 CX7 InfiniBand connection' per GPU, 3.2 Tbps per VM, GPUDirect RDMA (Microsoft Learn)

Grades summarize deliverability from the seller's own published specs — methodology. A missing fabric field is recorded as unknown, never guessed. Weekly context in the Delivered Compute Report.

Prices move weekly. Get the report.

Headline prices, deliverability grades, and one teardown a week — free.

Subscribe free