Two Nvidia data center GPUs, one Hopper and one Ada Lovelace: 141 GB HBM3e against 48 GB GDDR6, 5.6x the memory bandwidth on the H200, and NVLink 900 GB/s on the H200 where the L40 is PCIe only.
At a glance
Memory
- H200
- 141 GB
- L40
- 48 GB
2.9x more on H200
Memory bandwidth
- H200
- 4,800 GB/s
- L40
- 864 GB/s
5.6x higher on H200
Cheapest on-demand
- H200
- $2.09 /GPU/hr
- L40
- $0.33 /GPU/hr
6.3x more expensive on H200
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | H200 | L40 | Difference |
|---|---|---|---|
| On-demand | $4.40 /GPU/hr from $2.09 | $1.00 /GPU/hr from $0.33 | 4.4x more expensive on H200 |
| Reserved | $3.84 /GPU/hr from $2.29 (1 mo) | -- /GPU/hr from $0.70 | |
| Spot | $2.99 /GPU/hr from $1.32 | $0.79 /GPU/hr from $0.77 | 3.8x more expensive on H200 |
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. No median for reserved on the L40: we only show one when at least 3 providers list the GPU on that billing type. Verify before provisioning. More on how we price.
Price history
H200 vs L40 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | H200 | L40 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 141 GB HBM3e | 48 GB GDDR6 | 2.9x more on H200 |
| Bandwidth | 4,800 GB/s | 864 GB/s | 5.6x higher on H200 |
| Bus width | 384-bit | ||
| Compute · dense | |||
| FP8 | 1,979 TFLOPS | 362 TFLOPS | 5.5x higher on H200 |
| FP16 / BF16 | 989.5 TFLOPS | 181.1 TFLOPS | 5.5x higher on H200 |
| INT8 | 1,979 TOPS | 362 TOPS | 5.5x higher on H200 |
| FP32 | 67 TFLOPS | 90.5 TFLOPS | 1.4x higher on L40 |
| FP64 | 34 TFLOPS | Not supported | H200 only |
| Platform | |||
| Interconnect | NVLink 900 GB/s | PCIe only | NVLink on H200 only |
| Board power | 700 W | 300 W | 2.3x higher on H200 |
| Released | Q4 2023 | Q4 2022 | 13 months newer on H200 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: H200 datasheet and L40 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | H200 /GPU/hr | L40 /GPU/hr |
|---|---|---|
Lium
|
On-Demand $2.85 | On-Demand $0.33 |
GPU.ai
|
On-Demand $3.29 | On-Demand $0.78 |
Runpod
|
On-Demand $3.59 | On-Demand $0.82 |
Massed Compute
|
On-Demand $3.62 | On-Demand $0.86 |
Koyeb
|
On-Demand $3.00 | -- |
Vast.ai
|
On-Demand $3.60 | -- |
Verda
|
On-Demand $4.00 | -- |
Lyceum
|
On-Demand $4.29 | -- |
EmpirioLabs AI
|
-- | On-Demand $0.70 |
Thunder Compute
|
-- | On-Demand $0.79 |
Sesterce
|
-- | On-Demand $0.97 |
Runcrate
|
-- | On-Demand $0.97 |
46 providers list the H200 and 13 the L40. Every listing is on the H200 page and the L40 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
From $1.10 /GPU/hr Compare -
Nvidia B200
From $3.75 /GPU/hr Compare -
Nvidia RTX PRO 6000
From $0.66 /GPU/hr Compare -
Nvidia A100
From $0.80 /GPU/hr Compare -
Nvidia L40S
From $0.38 /GPU/hr Compare
Our data for Nvidia H200 was last updated on Sept. 7, 2026, and for Nvidia L40 on Sept. 7, 2026.