1× L4048 GB VRAM
Canada · 28 vCPU · 58 GB RAM · 100GB root + 725GB ephemeral
Details
Stock and price checked 7 min ago.
You start it yourself once there is stock.
n3-L40x1
Full price
€0.89
per hour, whole instance, 1 GPU
Spot €0.71/h · no stock data
NVIDIA · Ada Lovelace
Rent L40
NVIDIA L40 GPU, 48 GB.RenderingVirtual desktopsgeneral ml
Full price from
—
in stock, per GPU-hour
Spot from
—
in stock, per GPU-hour
Offers in stock
0 of 4
full price or spot
Providers in stock
0 of 1
full price or spot
Offers (5)
View offer
2× L4048 GB each, 96 GB total
Canada · 60 vCPU · 116 GB RAM · 100GB root + 1550GB ephemeral
Details
Stock and price checked 7 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-L40x2
Full price
€1.78
per hour, whole instance, 2 GPUs
€0.89/h per GPU
Spot €1.43/h instance · €0.71/h per GPU · no stock data
4× L4048 GB each, 192 GB total
Canada · 124 vCPU · 232 GB RAM · 100GB root + 3200GB ephemeral
Details
Stock and price checked 7 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-L40x4
Full price
€3.57
per hour, whole instance, 4 GPUs
€0.89/h per GPU
Spot €2.85/h instance · €0.71/h per GPU · no stock data
8× L4048 GB each, 384 GB total
Canada · 252 vCPU · 464 GB RAM · 100GB root + 6500GB ephemeral
Details
Stock and price checked 7 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-L40x8
Full price
€7.13
per hour, whole instance, 8 GPUs
€0.89/h per GPU
Spot €5.71/h instance · €0.71/h per GPU · no stock data
Full price
from €0.89
per GPU, per hour
Price and availability history
Price per GPU-hour
Excluding VAT.
Providers with stock
Each provider counts once. Full price and spot are separate.
Not enough history in this range yet. We record every 10 minutes since 9 November 2026.
Share of offers in stock
Full price. Offers we can't say are really sold there are left out until they've been in stock once.
Not enough history in this range yet. We record every 10 minutes since 9 November 2026.
Units free
No provider reports a number for this GPU.
Not enough history in this range yet. We record every 10 minutes since 9 November 2026.
Stock by provider
In stock, yes or no, for every provider selling this GPU. A filled band means in stock: green at full price, orange as spot.
Not enough history in this range yet. We record every 10 minutes since 9 November 2026.
Using several together
Can't be linked. This GPU has no NVLink, so several in one server only talk over PCIe at up to 64 GB/s. Each card keeps its own 48 GB. That suits running a separate job on each card. Splitting one model across them works, but slowly.
Specifications
No NVLink; host link PCIe Gen4 x16 64 GB/s. Datasheet is marked preliminary. memoryDataRateGbps is derived (864 GB/s x 8 / 384 bits). fp64Tflops is derived: NVIDIA states AD102 FP64 is 1/64th of FP32 (90.5 / 64).
Memory
- VRAM
- 48 GB
- Memory type
- GDDR6
- Memory bandwidthHow fast the GPU reaches its memory. Often limits LLM speed
- 864 GB/s
- Bus width
- 384 bit
- ECC memoryDetects and corrects memory errors
- Yes
Compute
- FP64Speed at 64-bit precision, used for scientific simulation
- 1.41 TFLOPS
- FP32Speed at 32-bit precision
- 90.5 TFLOPS
- FP16Speed at 16-bit precision
- 362.1 TFLOPS
- BF1616-bit AI format. Standard for training and full-quality inference
- 362.1 TFLOPS
- FP88-bit AI format. About half the memory of BF16, slightly lower quality
- 724 TFLOPS
- INT88-bit integer speed, used by quantized models
- 724 TOPS
- FP8 support
- Yes
- FP4 support
- No
- Transformer EngineHardware that speeds up FP8 and FP4
- Yes
- Compute cores
- 18,176
Multi-GPU and connectivity
- NVSwitchLets every GPU in a server talk to every other at full speed
- No
- PCIe generation
- 4
- PCIe lanes
- 16
- PCIe bandwidth
- 64 GB/s
Sharing and security
- MIG partitioningSplits one GPU into several isolated smaller GPUs
- No
- Max MIG instances
- 0
- vGPU
- Yes
- Confidential computing
- No
Hardware
- Architecture
- Ada Lovelace
- Launch year
- 2022
- Process node
- TSMC 4N
- Transistors
- 76.3 billion
- Die size
- 608.5 mm²
- Power (TDP)Maximum power draw
- 300 W
- Cooling
- passive
Video
- NVENC engines
- 3
- NVDEC engines
- 3
- AV1 encode
- Yes
- AV1 decode
- Yes
Sources (43)
- https://www.nvidia.com/content/dam/en-zz/Solutions/design-visualization/support-guide/NVIDIA-L40-Datasheet-January-2023.pdf
- https://nvidianews.nvidia.com/news/nvidia-announces-ovx-computing-systems-the-graphics-and-simulation-foundation-for-the-metaverse-powered-by-ada-lovelace-gpu
- https://www.nvidia.com/en-us/data-center/solutions/confidential-computing/
- https://docs.nvidia.com/590trd1-trusted-computing-solutions-release-notes.pdf
- https://developer.nvidia.com/blog/inside-nvidia-blackwell-ultra-the-chip-powering-the-ai-factory-era/
- https://images.nvidia.com/aem-dam/Solutions/geforce/ada/nvidia-ada-gpu-architecture.pdf
- https://developer.nvidia.com/video-encode-decode-support-matrix
- https://docs.nvidia.com/vgpu/gpus-supported-by-vgpu.html
- https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/datasheets/L-40/product-brief-L40.pdf