Overview
Nvidia B100
|
Nvidia H20
|
|
|---|---|---|
| Summary | Blackwell data center GPU for next-generation AI training and inference. | Export-compliant Hopper inference GPU with lower compute than H100. |
| Architecture | Blackwell | Hopper |
| Memory | Up to 192GB HBM3e | 96GB HBM3 |
| Memory Bandwidth | 8,000 GB/s | 4,000 GB/s |
| Performance | -- | 148 TFLOPS (FP16) |
| Release Date | Q4 2024 | Early 2024 |
| Performance Segment | High Performance | High Performance |
| Datasheet | Link | -- |
Price comparison
B100 Pricing
The B100 is not currently offered by any cloud providers we track.
No active offerings found
H20 Pricing
The H20 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:
Nvidia B100
vs
Nvidia B200
Nvidia B100
vs
Nvidia H100
Nvidia H20
vs
Nvidia H100
Nvidia H20
vs
Nvidia L40S
Our data for Nvidia B100 was last updated on March 25, 2026, and for Nvidia H20 on March 25, 2026.