1× L424 GB VRAM
Germany · 8 vCPU · 32 GB RAM · 15 Gbit/s
Details
Stock and price checked 6 min ago.
You start it yourself once there is stock.
pi5e.2xlarge.4
Full price
€0.75
per hour, whole instance, 1 GPU
NVIDIA · Ada Lovelace
Rent L4
NVIDIA L4 Tensor Core GPU, 24 GB.inference smallVirtual desktopsRendering
Full price from
—
in stock, per GPU-hour
Spot from
—
in stock, per GPU-hour
Offers in stock
0 of 27
full price or spot
Providers in stock
0 of 3
full price or spot
Offers (28)
View offer
1× L424 GB VRAM
Germany · 8 vCPU · 32 GB RAM · 15 Gbit/s
Details
Stock and price checked 6 min ago.
You start it yourself once there is stock.
pi5e.2xlarge.4
Full price
€0.75
per hour, whole instance, 1 GPU
4× L424 GB each, 96 GB total
France · 90 vCPU · 360 GB RAM · 25 Gbit/s · 400GB NVMe
Details
Stock and price checked 6 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.
l4-360
Full price
€3.00
per hour, whole instance, 4 GPUs
€0.75/h per GPU
1× L424 GB VRAM
France · 22 vCPU · 90 GB RAM · 8 Gbit/s · 400GB NVMe
Details
Stock and price checked 6 min ago.
You start it yourself once there is stock.
l4-90
Full price
€0.75
per hour, whole instance, 1 GPU
2× L424 GB each, 48 GB total
France · 45 vCPU · 180 GB RAM · 16 Gbit/s · 400GB NVMe
Details
Stock and price checked 6 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.
l4-180
Full price
€1.50
per hour, whole instance, 2 GPUs
€0.75/h per GPU
4× L424 GB each, 96 GB total
France · 32 vCPU · 192 GB RAM · 10 Gbit/s · Block
Details
Stock and price checked 6 min ago.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
L4-4-24G
Full price
€3.15
per hour, whole instance, 4 GPUs
€0.79/h per GPU
4× L424 GB each, 96 GB total
Poland · 32 vCPU · 192 GB RAM · 10 Gbit/s · Block
Details
Stock and price checked 6 min ago.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
L4-4-24G
Full price
€3.15
per hour, whole instance, 4 GPUs
€0.79/h per GPU
2× L424 GB each, 48 GB total
France · 16 vCPU · 96 GB RAM · 5 Gbit/s · Block
Details
Stock and price checked 6 min ago.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
L4-2-24G
Full price
€1.58
per hour, whole instance, 2 GPUs
€0.79/h per GPU
Full price
from €0.75
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 24 GB. That suits running a separate job on each card. Splitting one model across them works, but slowly.
Specifications
1-slot low-profile; no NVLink, host link PCIe Gen4 x16 64 GB/s. Datasheet gives no CUDA core count. memoryDataRateGbps is derived (300 GB/s x 8 / 192 bits, matching the 6251 MHz memory clock x 2).
Memory
- VRAM
- 24 GB
- Memory type
- GDDR6
- Memory bandwidthHow fast the GPU reaches its memory. Often limits LLM speed
- 300 GB/s
- Bus width
- 192 bit
- ECC memoryDetects and corrects memory errors
- Yes
Compute
- FP32Speed at 32-bit precision
- 30.3 TFLOPS
- FP16Speed at 16-bit precision
- 242 TFLOPS
- BF1616-bit AI format. Standard for training and full-quality inference
- 242 TFLOPS
- FP88-bit AI format. About half the memory of BF16, slightly lower quality
- 485 TFLOPS
- INT88-bit integer speed, used by quantized models
- 485 TOPS
- FP8 support
- Yes
- FP4 support
- No
- Transformer EngineHardware that speeds up FP8 and FP4
- Yes
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
- 2023
- Process node
- TSMC 4N
- Power (TDP)Maximum power draw
- 72 W
- Cooling
- passive
Video
- NVENC engines
- 2
- NVDEC engines
- 4
- AV1 encode
- Yes
- AV1 decode
- Yes
Sources (39)
- https://www.nvidia.com/en-us/data-center/l4/
- https://dam-cdn.nvd.orangelogic.com/AssetLink/8f4cjsr458mk07kxxt608b1ct0t570x7.pdf
- https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/l4/PB-11316-001_v01.pdf
- https://images.nvidia.com/data-center/technical-brief-gpu-positioning-virtualized-compute-and-graphics-workloads.pdf
- https://nvidianews.nvidia.com/news/nvidia-launches-inference-platforms-for-large-language-models-and-generative-ai-workloads
- 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