Overview
Nvidia H20
|
Nvidia H800
|
|
|---|---|---|
| Summary | Export-compliant Hopper inference GPU with lower compute than H100. | H100 variant with reduced interconnect for the Chinese market. |
| Architecture | Hopper | Hopper |
| Memory | 96GB HBM3 | 80GB HBM2e |
| Memory Bandwidth | 4,000 GB/s | 2,000 GB/s |
| Performance | 148 TFLOPS (FP16) | 1,513 TFLOPS (FP16) |
| Release Date | Early 2024 | March 2023 |
| Performance Segment | High Performance | High Performance |
| Datasheet | -- | -- |
Price comparison
H20 Pricing
The H20 is not currently offered by any cloud providers we track.
No active offerings found
H800 Pricing
The H800 is not currently offered by any cloud providers we track.
Heads up: We do our best to keep these 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 with the provider before provisioning.
Alternative GPUs
Here's how these GPUs compare against alternatives:
Nvidia H20
vs
Nvidia H100
Nvidia H20
vs
Nvidia L40S
Nvidia H800
vs
Nvidia H100
Nvidia H800
vs
Nvidia A800
Our data for Nvidia H20 was last updated on March 25, 2026, and for Nvidia H800 on March 25, 2026.