Two Nvidia data center GPUs, one Ampere and one Ada Lovelace: 80 GB HBM2e against 24 GB GDDR6, 6.5x the memory bandwidth on the A100, and NVLink 600 GB/s on the A100 where the L4 is PCIe only.
At a glance
Memory
- A100
- 80 GB
- L4
- 24 GB
3.3x more on A100
Memory bandwidth
- A100
- 1,935 GB/s
- L4
- 300 GB/s
6.5x higher on A100
Cheapest on-demand
- A100
- $0.40 /GPU/hr
- L4
- $0.32 /GPU/hr
20% cheaper on L4
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | A100 | L4 | Difference |
|---|---|---|---|
| On-demand | $1.74 /GPU/hr from $0.40 | $0.90 /GPU/hr from $0.32 | 48% cheaper on L4 |
| Reserved | $1.63 /GPU/hr from $0.48 (1 mo) | $0.52 /GPU/hr from $0.32 (36 mo) | 3.1x more expensive on A100 |
| Spot | $0.97 /GPU/hr from $0.29 | $0.53 /GPU/hr from $0.14 | 45% cheaper on L4 |
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. Verify before provisioning. More on how we price.
Price history
A100 vs L4 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A100 | L4 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 80 GB HBM2e 40 GB on SXM, PCIe | 24 GB GDDR6 | 3.3x more on A100 |
| Bandwidth | 1,935 GB/s 2,039 GB/s on SXM 1,555 GB/s on SXM, PCIe | 300 GB/s | 6.5x higher on A100 |
| Bus width | 5,120-bit | 192-bit | 27x wider on A100 |
| Compute · dense | |||
| FP8 | Not supported | 242 TFLOPS | L4 only |
| FP16 / BF16 | 312 TFLOPS | 121 TFLOPS | 2.6x higher on A100 |
| INT8 | 624 TOPS | 242 TOPS | 2.6x higher on A100 |
| FP32 | 19.5 TFLOPS | 30.3 TFLOPS | 1.6x higher on L4 |
| FP64 | 9.7 TFLOPS | Not supported | A100 only |
| Platform | |||
| Interconnect | NVLink 600 GB/s | PCIe only | NVLink on A100 only |
| Board power | 300 W | 72 W | 4.2x higher on A100 |
| Released | Q2 2020 | Q1 2023 | 3 years newer on L4 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32FP64INT4INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A100 datasheet and L4 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A100 /GPU/hr | L4 /GPU/hr |
|---|---|---|
Vast.ai
|
On-Demand $0.40 | On-Demand $0.32 |
GPU.ai
|
On-Demand $0.67 | On-Demand $0.32 |
Runpod
|
On-Demand $1.00 | On-Demand $0.49 |
Thunder Compute
|
On-Demand $1.09 | -- |
Lium
|
On-Demand $1.23 | -- |
Massed Compute
|
On-Demand $1.35 | -- |
Packet·ai
|
On-Demand $1.43 | -- |
Oblivus
|
On-Demand $1.47 | -- |
UpCloud
|
-- | On-Demand $0.69 |
Scaleway
|
-- | On-Demand $0.91 |
Sesterce
|
-- | On-Demand $1.04 |
Runcrate
|
-- | On-Demand $1.05 |
Hexabyte
|
-- | On-Demand $0.42 |
44 providers list the A100 and 19 the L4. Every listing is on the A100 page and the L4 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
From $1.41 /GPU/hr Compare -
Nvidia RTX 5090
From $0.35 /GPU/hr Compare -
Nvidia RTX 4090
From $0.23 /GPU/hr Compare -
Nvidia A10
From $0.20 /GPU/hr Compare
Our data for Nvidia A100 was last updated on Sept. 14, 2026, and for Nvidia L4 on Sept. 14, 2026.