A16 vs L40S
The A16 and the L40S are both data centre GPUs we supply tested from Hong Kong, and the figures below are the same ones published on each product page. They share the same supply and warranty terms and separate on memory, power, interface, so the decision is less about "which is better" and more about which of those differences your platform or workload can actually use.
Nothing on this page is a price: used and refurbished hardware is repriced weekly, so both are quoted on request against confirmed stock with the date the quote holds until. If you are replacing a part, send the platform and we will confirm the population before you commit.
Side by side
A16 vs L40S specifications
| Specification | A16 | L40S |
|---|---|---|
| Memory | 4x 16GB GDDR6 with ECC (64GB total) | 48GB GDDR6 with ECC, 864 GB/s |
| Power | 250W TDP, 8-pin auxiliary power connector | 350W TDP, 16-pin (CEM5) power connector |
| Interface | PCIe Gen4 x16, full-height full-length dual-slot, passive | PCIe Gen4 x16, dual-slot full-height, passive |
| Architecture | Ampere, 4x 1,280 CUDA cores (5,120 total), 4x 40 third-gen Tensor cores, 4x 10 second-gen RT cores | Ada Lovelace, 18,176 CUDA cores, 568 fourth-gen Tensor cores, 142 third-gen RT cores |
| Compute | not stated | FP8 733 TFLOPS / FP16-BF16 362 TFLOPS / TF32 183 TFLOPS / FP32 91.6 TFLOPS |
| NVLink / MIG | not stated | No NVLink; MIG not supported on this board |
| Virtualization | vGPU support: NVIDIA vPC, vApps, RTX vWS and vCS | vGPU supported; NVIDIA AI Enterprise ready |
| Model | NVIDIA A16 64GB PCIe (quad-GPU board; Dell and other OEM-branded boards are the same GPU) | NVIDIA L40S 48GB PCIe; OEM-branded Dell / Lenovo / HP boards are the same GPU |
| Media | 4x NVENC / 8x NVDEC — includes AV1 decode | 3x NVENC / 3x NVDEC — AV1 encode and decode |
| User Density | Up to 64 concurrent VDI users per board | not stated |
| Compliance | Secure and measured boot with hardware root of trust; NEBS Level 3 | not stated |
Rows that differ are highlighted. Figures are the same ones published on each product page; where a figure is community-sourced rather than vendor-published it is stated on that page. Confirm every part against your build sheet when ordering.
Where they differ
- ·Memory — A16: 4x 16GB GDDR6 with ECC (64GB total); L40S: 48GB GDDR6 with ECC, 864 GB/s.
- ·Power — A16: 250W TDP, 8-pin auxiliary power connector; L40S: 350W TDP, 16-pin (CEM5) power connector.
- ·Interface — A16: PCIe Gen4 x16, full-height full-length dual-slot, passive; L40S: PCIe Gen4 x16, dual-slot full-height, passive.
- ·Architecture — A16: Ampere, 4x 1,280 CUDA cores (5,120 total), 4x 40 third-gen Tensor cores, 4x 10 second-gen RT cores; L40S: Ada Lovelace, 18,176 CUDA cores, 568 fourth-gen Tensor cores, 142 third-gen RT cores.
- ·Virtualization — A16: vGPU support: NVIDIA vPC, vApps, RTX vWS and vCS; L40S: vGPU supported; NVIDIA AI Enterprise ready.
- ·Media — A16: 4x NVENC / 8x NVDEC — includes AV1 decode; L40S: 3x NVENC / 3x NVDEC — AV1 encode and decode.
Decision
Which one to buy
Data centre GPUs are chosen by memory footprint per job and by the deployment you already have. More memory per card means bigger models, larger batches and fewer cards for the same workload; less memory per card is often the better buy when you scale out across many nodes. Memory bandwidth and compute format matter for the workload — FP8 and BF16 throughput for inference, FP64 for simulation — and the power envelope decides how many cards fit per chassis. Check the form factor and cooling as well: a passive server card and an actively cooled workstation card are not interchangeable.
Choose the A16 if
- ·Memory: 4x 16GB GDDR6 with ECC (64GB total)
- ·Power: 250W TDP, 8-pin auxiliary power connector
- ·Interface: PCIe Gen4 x16, full-height full-length dual-slot, passive
Choose the L40S if
- ·Memory: 48GB GDDR6 with ECC, 864 GB/s
- ·Power: 350W TDP, 16-pin (CEM5) power connector
- ·Interface: PCIe Gen4 x16, dual-slot full-height, passive
Frequently asked questions
How much do the A16 and the L40S cost?
We do not publish prices for either: secondary-market hardware is repriced weekly. Send the quantity, the platform and the condition you want and we quote both against confirmed stock, with the date the figure holds until.
What is the main difference between the A16 and the L40S?
Memory (A16: 4x 16GB GDDR6 with ECC (64GB total); L40S: 48GB GDDR6 with ECC, 864 GB/s); Power (A16: 250W TDP, 8-pin auxiliary power connector; L40S: 350W TDP, 16-pin (CEM5) power connector); Interface (A16: PCIe Gen4 x16, full-height full-length dual-slot, passive; L40S: PCIe Gen4 x16, dual-slot full-height, passive).
Can I use the A16 and the L40S in the same system?
They can share a server if the power and slot budget allows it, but they target different jobs: A16 and L40S differ mainly on Memory and Power. Mixed GPU models in one host is a driver/CUDA compatibility decision, not a hardware one.
Which one should I buy?
Buy the one whose strong points match the job: see the "Choose" section above. If you are replacing an existing part, send the platform or the old part number and we will confirm compatibility before you order.
What condition are they in, and do you ship worldwide?
Both are supplied tested with a 1-year written warranty. Condition is stated line by line before you order — new, decommissioned pull or OEM tray — and we ship worldwide from the Hong Kong warehouse with full export documentation, typically 3–7 days by air.
Get a quote for either — or both
Tell us the quantity, the platform and the condition you need. We confirm stock and put a firm figure in writing with the date it holds until.
Request a Quote