Overview

Nvidia logo Nvidia H20 Nvidia logo Nvidia P40
Summary Export-compliant Hopper inference GPU with lower compute than H100. Budget legacy GPU when VRAM matters more than speed.
Architecture Hopper Pascal
Memory 96GB HBM3 24GB GDDR5
Memory Bandwidth 4,000 GB/s 346GB/s
Performance 148 TFLOPS (FP16) --
Release Date Early 2024 Q3 2016
Performance Segment High Performance Budget
Datasheet -- Link

Price comparison

Nvidia H20 logo H20 Pricing

The H20 is not currently offered by any cloud providers we track.

No active offerings found

Nvidia P40 logo P40 Pricing

The P40 is not currently offered by any cloud providers we track.

Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.

Alternative GPUs

Here's how these GPUs compare against alternatives:

View all GPUs

Our data for Nvidia H20 was last updated on March 25, 2026, and for Nvidia P40 on March 25, 2026.