L40S vs L4: specifications compared
Comparison of L40S and L4 based on data from the manufacturer or provider.
According to the manufacturer, L40S and L4 differ mainly in power (TDP) (L40S: 350 W, L4: 72 W), memory bandwidth (L40S: 864 GB/s, L4: 300 GB/s) and memory (L40S: 48 GB, L4: 24 GB). The form factor also differs: PCIe, dual-slot versus PCIe, single-slot low-profile. In the Clouwino points, which are calculated from the specifications, L40S gets 19 and L4 gets 16. Checked on 19 October 2026, no prices.
How are the points calculated?
Clouwino points are an arithmetic summary of the specifications the manufacturer states, on a scale of 0 to 100. It is not a benchmark and not a judgement on quality or suitability: a product with fewer points may suit your situation better. Products are compared with other products of the same type. For each property we count how far a product sits between the lowest and highest value in the group:
- FP32 compute (for: consumer), carries heavy weight
- Memory (for: consumer), carries heavy weight
- Memory bandwidth (for: consumer), carries heavy weight
- Compute units (for: consumer), carries light weight
- Boost clock (for: consumer), carries light weight
- Memory (for: professional and data centre), carries medium weight
- Memory bandwidth (for: professional and data centre), carries medium weight
- FP16/BF16 (for: professional and data centre), carries heavy weight
- FP8 (for: professional and data centre), carries medium weight
- FP64 (for: professional and data centre), carries medium weight
If too much data is missing, we do not show points. Points are relative: when a new product is added, they can shift. Read more at how we compare.
L40S and L4 side by side
| Specificatie | NVIDIA L40S | NVIDIA L4 |
|---|---|---|
| Overview | ||
| Clouwino points | 19 punten | 16 punten |
| General | ||
| Architecture | Ada Lovelace | Ada Lovelace |
| Manufacturing process | Not stated by the manufacturer | Not stated by the manufacturer |
| Form | PCIe, dual-slot | PCIe, single-slot low-profile |
| Launch | Not stated by the manufacturer | Not stated by the manufacturer |
| Compute | ||
| FP32 compute | 91,6 TFLOPS | 30,3 TFLOPS |
| FP64 | Not stated by the manufacturer | Not stated by the manufacturer |
| FP16/BF16 | 362,05 TFLOPS | 242 TFLOPS |
| FP8 | 733 TFLOPS | 485 TFLOPS |
| FP4 | Not stated by the manufacturer | Not stated by the manufacturer |
| Performance notes | NVIDIA states dense first, then with sparsity (*). FP16/BF16: 362.05 dense (733 with sparsity). FP8: 733 dense (1,466 with sparsity). TF32 183 dense / 366 sparse. FP32 91.6 TFLOPS without a label. FP64 and FP4 not stated. | FP16/BF16 242 and FP8 485 TFLOPS are values WITH SPARSITY (NVIDIA footnote: 'Shown with sparsity. Specifications are one-half lower without sparsity'); dense values are not explicitly on the page and have not been calculated. Also TF32 120 TFLOPS and INT8 485 TOPS with sparsity. FP32 30.3 TFLOPS without a label. FP64 and FP4 not stated. |
| Memory | ||
| Memory | 48 GB | 24 GB |
| Memory type | GDDR6 (ECC) | Not stated by the manufacturer |
| Memory bandwidth | 0,86 TB/s | 0,3 TB/s |
| Power and connections | ||
| Power (board power or TDP) | 350 W | 72 W |
| GPU interconnect | PCIe Gen4 x16: 64 GB/s bidirectional; no NVLink | PCIe Gen4 x16 64 GB/s |
| Multiple instances | Not stated by the manufacturer | Not stated by the manufacturer |
Note: Figures per GPU. Max. power consumption 350 W. MIG: not supported; NVLink: not supported. Peak values based on boost clock. Process node and launch date not on the page. The datasheet link was not opened (redirect).
Note: Figures per GPU. Memory type, process node, MIG support and launch date are not on the page (null). Architecture taken from the introductory text on the page.
Main differences between L40S and L4
These are the differences the manufacturer lists in the specifications. The full table is further down.
- Power (TDP): L40S 350 W, L4 72 W. L40S lists about 4.9 times as much.
- Memory bandwidth: L40S 864 GB/s, L4 300 GB/s. L40S lists about 2.9 times as much.
- Memory: L40S 48 GB, L4 24 GB. L40S lists about 2 times as much.
- Form factor: L40S PCIe, dual-slot, L4 PCIe, single-slot low-profile.
What should you look at when choosing?
A higher number is not better for everyone. It depends on what you want to use it for.
- Power (TDP). A higher power rating calls for a suitable power supply and cooling, and can use more electricity.
- Memory bandwidth. Higher bandwidth means data can move to and from memory faster.
- Memory. Memory determines how much data the GPU can hold at once, for example large models, datasets or textures.
Also look at the rest of your system, such as power supply, cooling and software. Real-world performance depends on more than the specifications alone.
What do the Clouwino points say?
L40S gets 19 points and L4 gets 16 points. The points are calculated from the manufacturer’s specifications, within the same kind of products. They are not benchmark scores and say nothing about price or availability. Read how we calculate them at How we compare.
Frequently asked questions
What is the difference between L40S and L4?
According to the manufacturer, L40S and L4 differ mainly in power (TDP) (L40S: 350 W, L4: 72 W), memory bandwidth (L40S: 864 GB/s, L4: 300 GB/s) and memory (L40S: 48 GB, L4: 24 GB). The form factor also differs: PCIe, dual-slot versus PCIe, single-slot low-profile. In the Clouwino points, which are calculated from the specifications, L40S gets 19 and L4 gets 16.
Which one has the highest power (TDP)?
According to the manufacturer, L40S lists the highest power (TDP): 350 W versus 72 W.
What do L40S and L4 cost?
Clouwino does not show prices. You can see the current price and availability with the partner, via the button on the product page.
Where does the data come from?
The specifications come from the manufacturers' official pages and were checked on 19 October 2026. The sources are at the bottom of this page.
Last updated on
