The B100 is newer, with 2.4x the memory and 4x the bandwidth.
At a glance
Memory
2.4x more on B100
- B100
- 192 GB
- H800
- 80 GB
Memory bandwidth
4x higher on B100
- B100
- 8.00 TB/s
- H800
- 2.00 TB/s
Cheapest on-demand
- B100
- --
- H800
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Specs comparison
| Spec | B100 | H800 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 192 GB HBM3e | 80 GB HBM2e | 2.4x more on B100 |
| Bandwidth | 8,000 GB/s | 2,000 GB/s | 4x higher on B100 |
| Bus width | 5,120-bit | ||
| Compute · dense | |||
| FP4 | 7,000 TFLOPS | Not supported | B100 only |
| FP8 | 3,500 TFLOPS | 1,513 TFLOPS | 2.3x higher on B100 |
| FP16 / BF16 | 1,750 TFLOPS | 756.5 TFLOPS | 2.3x higher on B100 |
| INT8 | 3,500 TOPS | 1,513 TOPS | 2.3x higher on B100 |
| FP32 | 60 TFLOPS | 51 TFLOPS | 18% higher on B100 |
| FP64 | 30 TFLOPS | ||
| Platform | |||
| Interconnect | NVLink 1,800 GB/s | NVLink 400 GB/s | 4.5x higher on B100 |
| Released | Q4 2024 | Q1 2023 | 2 years newer on B100 |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: B100 datasheet .
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
80 GB HBM3 · 58% less bandwidth than the B100
From $1.30 /GPU/hr Compare -
Nvidia B200
180 GB HBM3e · released Q1 2024
From $3.75 /GPU/hr Compare
-
Nvidia H200
141 GB HBM3e · 40% less bandwidth than the B100
From $2.09 /GPU/hr Compare
Our data for Nvidia B100 was last updated on Sept. 14, 2026, and for Nvidia H800 on Sept. 14, 2026.