AMD MI250
vs
Nvidia L40S
Two data center GPUs, one CDNA2 and one Ada Lovelace: 128 GB HBM2e against 48 GB GDDR6, 3.7x the memory bandwidth on the MI250, and Infinity Fabric 400 GB/s on the MI250 where the L40S is PCIe only.
At a glance
Updated 6 minutes ago- Memory
- 128 GB
- HBM2e
- Memory bandwidth
- 3.20 TB/s
- From
- --
- Not sold on demand · 1 provider
- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 864 GB/s
- From
- $0.55 /GPU/hr
- 34 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | MI250 | L40S | Difference |
|---|---|---|---|
| On-demand | Not sold on demand | $0.55 | |
| Spot | No spot listings | $0.74 | |
| Reserved | $1.30 | $0.60 | 2.2x more expensive on MI250 |
| Median | $1.50 | $1.60 | 6% cheaper on MI250 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
MI250 vs L40S price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI250 | L40S | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 128 GB HBM2e | 48 GB GDDR6 | 2.7x more on MI250 |
| Bandwidth | 3,200 GB/s | 864 GB/s | 3.7x higher on MI250 |
| Bus width | 8,192-bit | 384-bit | 21x wider on MI250 |
| Compute · dense | |||
| FP8 | Not supported | 733 TFLOPS | L40S only |
| FP16 / BF16 | 362.1 TFLOPS | 362.1 TFLOPS | Within 5% |
| INT8 | 362.1 TOPS | 733 TOPS | 2x higher on L40S |
| FP32 | 90.5 TFLOPS | 91.6 TFLOPS | Within 5% |
| FP64 | 45.3 TFLOPS | Not supported | MI250 only |
| Platform | |||
| Interconnect | Infinity Fabric 400 GB/s | PCIe only | Infinity Fabric on MI250 only |
| Board power | 500 W | 350 W | 1.4x higher on MI250 |
| Released | Q4 2021 | Q3 2023 | 2 years newer on L40S |
| Precision | |||
| Supported formats | FP16BF16FP32FP64INT4INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI250 datasheet and L40S datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | MI250 /GPU/hr | L40S /GPU/hr |
|---|---|---|
Cirrascale
|
Reserved (12mo) $1.30 | -- |
Novita
|
-- | On-Demand $0.55 |
GPU.ai
|
-- | On-Demand $0.80 |
Packet·ai
|
-- | On-Demand $0.92 |
Sesterce
|
-- | On-Demand $0.97 |
Massed Compute
|
-- | On-Demand $0.97 |
Runcrate
|
-- | On-Demand $0.97 |
UpCloud
|
-- | On-Demand $1.19 |
Koyeb
|
-- | On-Demand $1.20 |
1 provider list the MI250 and 34 the L40S. Every listing is on the MI250 page and the L40S page.
Similar GPUs
Other comparisons in the same class.
Our data for AMD MI250 was last updated on Sept. 7, 2026, and for Nvidia L40S on Sept. 7, 2026.