NVIDIA L40S
The L40S is optimized for AI inference and generative AI with 48GB GDDR6 and Ada Lovelace architecture. Combines strong AI performance with graphics capabilities for diverse workloads.
VRAM
48 GB
Memory
GDDR6
Bandwidth
864 GB/s
FP16
362 TF
TDP
350 W
Memory and compute
Architecture and power
Notes
AI inference and graphics
Models that may fit on NVIDIA L40S
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
More GPUs by NVIDIA
The manufacturer's other cards in the catalog, largest memory first, with bandwidth and power where published.
Frequently asked questions about NVIDIA L40S
Answered from the entry's own figures: memory, which models fit, bandwidth, power and the card's class.
NVIDIA L40S has 48 GB of GDDR6 at 864 GB/s. At INT8 a model's weights take about one byte per parameter, so the card holds roughly a 38B-parameter model with room for context; larger models need several cards or a lower precision.
From the catalog's estimates at INT8, Gemma 4 31B IT, Qwen 2.5 32B Instruct, DeepSeek R1 Distill Qwen 32B and Aya Vision 32B fit on a single NVIDIA L40S with their default context. The list on this page shows the estimated memory each takes; the capacity planner sizes any model against this card for your own context length, batch size and request rate.
864 GB/s. Bandwidth bounds how fast a model generates tokens, because every token reads the whole set of weights and the KV cache from memory; at this rate a 38B model at INT8 could be read on the order of 22 times a second on one card, before batching and framework efficiency.
NVIDIA L40S has a TDP of 350 W, with a maximum of 350 W, in the PCIe form factor. Budget the TDP per card plus host overhead when sizing a server's power and cooling.
NVIDIA L40S is a datacenter accelerator, meant for servers and multi-GPU serving, built on the Ada Lovelace architecture and released in 2023. It has no multi-GPU interconnect, so a model must fit on one card or be split over the slower PCIe bus.
Specifications from the manufacturer's published material; a figure that is not published is shown as such, and a derived one is explained in the notes.
Added Jan 25, 2026 · Last updated Jan 25, 2026
Explore more
Companion tools that draw on the same catalog and routing engine.
Put intelligence in the middle.
One gateway in front of every model, with your policies applied and every decision on record. Start on the free plan with 5,000 requests a month, no card required.