AMD MI250
vs
Nvidia B100
The B100 is newer, with 50% more memory and 2.5x the bandwidth.
At a glance
Memory
50% more on B100
- MI250
- 128 GB
- B100
- 192 GB
Memory bandwidth
2.5x higher on B100
- MI250
- 3.20 TB/s
- B100
- 8.00 TB/s
Cheapest on-demand
- MI250
- --
- B100
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Price history
MI250 vs B100 price history
Weekly median on-demand price per GPU per hour
Aggregating historical prices...
Specs comparison
| Spec | MI250 | B100 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 128 GB HBM2e | 192 GB HBM3e | 50% more on B100 |
| Bandwidth | 3,200 GB/s | 8,000 GB/s | 2.5x higher on B100 |
| Bus width | 8,192-bit | ||
| Compute · dense | |||
| FP4 | Not supported | 7,000 TFLOPS | B100 only |
| FP8 | Not supported | 3,500 TFLOPS | B100 only |
| FP16 / BF16 | 362.1 TFLOPS | 1,750 TFLOPS | 4.8x higher on B100 |
| INT8 | 362.1 TOPS | 3,500 TOPS | 9.7x higher on B100 |
| FP32 | 90.5 TFLOPS | 60 TFLOPS | 51% higher on MI250 |
| FP64 | 45.3 TFLOPS | 30 TFLOPS | 51% higher on MI250 |
| Platform | |||
| Interconnect | Infinity Fabric 400 GB/s | NVLink 1,800 GB/s | 4.5x higher on B100 |
| Released | Q4 2021 | Q4 2024 | 3 years newer on B100 |
| Precision | |||
| Supported formats | FP16BF16FP32FP64INT4INT8 | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI250 datasheet and B100 datasheet.
Similar GPUs
Other comparisons in the same class.
Our data for AMD MI250 was last updated on Sept. 14, 2026, and for Nvidia B100 on Sept. 14, 2026.