ComputeAssay

Provider record · updated 2026-09-10

AWS — GPU pricing & deliverability

4 tracked offerings · grade mix A:3 · B:1 · C:0 · D:0 · U:0. AWS P5 family: 3200 Gbps EFA per instance (p5=EFA, p5en=EFAv3 on Nitro v5), 900 GB/s NVSwitch intra-node; EC2 UltraClusters published claim of scaling up to 20,000 H100/H200 GPUs with up to 20 exaflops aggregate. P6: EFAv4 at 3.2 Tbps (B200) and 6.4 Tbps (B300) per instance; P6e-GB200 UltraServers publish 72 GPUs per NVLink domain (u-p6e-gb200x72) with 28,800 Gbps EFA per UltraServer and 'up to 130 TB/s' NVLink; UltraCluster scale for P6 stated only as 'tens of thousands of GPUs within a petabit-scale nonblocking network'.

Tracked offerings

GPUSKU$/GPU-hrTermsGradeFabric (as published)Src
H100-SXMp5.48xlarge$6.88on-demandAEFA, 3200 Gbps per instance (AWS P5 page)
H200-SXMp5en.48xlarge$7.91on-demandAEFAv3 (Nitro v5), 3200 Gbps per instance, 'up to 35% improvement in latency' vs earlier EFA (AWS P5 page)
B200p6-b200.48xlarge$14.24on-demandAEFAv4, 3.2 Tbps per instance (AWS P6 page)
B300p6-b300.48xlarge$17.80on-demandBEFAv4, 6.4 Tbps per instance (AWS P6 page)

What AWS does not publish

Unpublished fields are recorded as unknown and grade conservatively. Sellers improve their grades by disclosing — that is the point.

Weekly context on every provider

The Delivered Compute Report tracks how these prices and grades move.

Subscribe free