NVIDIA L4 (24GB GDDR6)
- 24GB GDDR6
- Memory
- 300GB/s
- Memory bandwidth
- Q1 2023
- Released
Price history
What it costs to rent an NVIDIA L4
The NVIDIA L4 is an Ada Lovelace-architecture inference NVIDIA GPU with 24GB of GDDR6 memory and 300 GB/s of memory bandwidth, released Q1 2023. It is built for inference, fine-tuning and mixed production workloads.
Renting one currently costs between $0.20 and $1.67 per GPU-hour across the 10 cloud providers we track, with a median of $0.80. The cheapest live rate is $0.20/hr from Vast.ai (Spot), and hosting one pays about $0.41/hr on RunPod while it is running. All rates below are normalised to a single GPU per hour so providers with multi-GPU nodes compare like-for-like.
Where to rent the L4
How we rank
The median is one provider, one vote: each provider contributes the median of its own on-demand listings, then the headline is the median across providers — so no single marketplace or hyperscaler sets the price alone. Below two providers we show the range instead.
Default order is relevance, weighed the way a buyer would: price per GPU first, lightly blended with total node cost; on-demand preferred over spot and reserved when prices are close; established clouds edged above marketplaces; in-stock, recently-checked offers above stale ones; and no single provider is allowed to crowd the top. Click a column to sort by the raw numbers. No paid placements — ranking is never sold.
Scroll horizontally to compare all columns →
| Provider | Tier | Config | Total VRAM | Billing | Availability | Go to provider | ||
|---|---|---|---|---|---|---|---|---|
 Vast.aiBest | GPU cloud | 1× L4 | 24GB | Spot | $0.20 | $0.20 | Rent | |
| Vast.ai | GPU cloud | 2× L4 | 48GB | Spot | $0.20 | $0.40 | Rent | |
 AWS | Hyperscaler | 4× L4 | 96GB | Spot | $0.25 | $1.01 | Rent | |
| Vast.ai | GPU cloud | 1× L4 | 24GB | On-demand | $0.27 | $0.27 | Rent | |
| Vast.ai | GPU cloud | 2× L4 | 48GB | On-demand | $0.32 | $0.64 | Rent | |
| Vast.ai | GPU cloud | 4× L4 | 96GB | On-demand | $0.32 | $1.28 | Rent | |
| AWS | Hyperscaler | 1× L4 | 24GB | Spot | $0.42 | $0.42 | Rent | |
| Vast.ai | GPU cloud | 4× L4 | 96GB | Spot | $0.29 | $1.17 | Rent | |
 RunPod | GPU cloud | 1× L4 | 24GB | On-demand | $0.59 | $0.59 | Rent | |
| AWS | Hyperscaler | 1× L4 | 24GB | Spot | $0.45 | $0.45 | Rent | |
| Vast.ai | GPU cloud | 5× L4 | 120GB | On-demand | $0.33 | $1.67 | Rent | |
 Google Cloud | Hyperscaler | 1× L4 | 24GB | On-demand | $0.70 | $0.70 | Rent | |
 Koyeb | GPU cloud | 1× L4 | 24GB | On-demand | $0.70 | $0.70 | Rent | |
| Vast.ai | GPU cloud | 6× L4 | 144GB | Spot | $0.29 | $1.76 | Rent | |
| Vast.ai | GPU cloud | 6× L4 | 144GB | On-demand | $0.33 | $2.01 | Rent | |
 Cerebrium | GPU cloud | 1× L4 | 24GB | On-demand | $0.80 | $0.80 | Rent | |
| AWS | Hyperscaler | 1× L4 | 24GB | Spot | $0.57 | $0.57 | Rent | |
| Vast.ai | GPU cloud | 5× L4 | 120GB | Spot | $0.32 | $1.60 | Rent | |
| AWS | Hyperscaler | 8× L4 | 192GB | Spot | $0.49 | $3.92 | Rent | |
 Sesterce | GPU cloud | 1× L4 | 24GB | On-demand | $1.05 | $1.05 | Rent | |
 OVH | GPU cloud | 1× L4 | 24GB | On-demand | $1.00 | $1.00 | Rent | |
| Sesterce | GPU cloud | 2× L4 | 48GB | On-demand | $1.05 | $2.09 | Rent | |
| OVH | GPU cloud | 2× L4 | 48GB | On-demand | $1.00 | $2.00 | Rent | |
| AWS | Hyperscaler | 4× L4 | 96GB | Spot | $0.60 | $2.40 | Rent | |
 Scaleway | Enterprise | 1× L4 | 24GB | On-demand | $0.88 | $0.88 | Rent | |
 Runcrate | Marketplace | from L4 | — | On-demand | $1.23 | — | Rent | |
| OVH | GPU cloud | 4× L4 | 96GB | On-demand | $1.00 | $4.00 | Rent | |
| Scaleway | Enterprise | 2× L4 | 48GB | On-demand | $0.88 | $1.77 | Rent | |
| AWS | Hyperscaler | 1× L4 | 24GB | Spot | $0.82 | $0.82 | Rent | |
| Scaleway | Enterprise | 4× L4 | 96GB | On-demand | $0.88 | $3.53 | Rent | |
| Scaleway | Enterprise | 8× L4 | 192GB | On-demand | $0.88 | $7.06 | Rent | |
| AWS | Hyperscaler | 4× L4 | 96GB | On-demand | $1.15 | $4.60 | Rent | |
| AWS | Hyperscaler | 1× L4 | 24GB | Spot | $1.14 | $1.14 | Rent | |
| AWS | Hyperscaler | 4× L4 | 96GB | On-demand | $1.67 | $6.68 | Rent | |
| AWS | Hyperscaler | 8× L4 | 192GB | On-demand | $1.67 | $13.35 | Rent |
Earn by hosting the L4
How this is worked out
“Renters pay” is the median live rate for this GPU on that marketplace — the median, not the floor, because the cheapest listing is one host undercutting everyone. “You earn” is that figure less the platform’s cut.
“Per day” is a ceiling: 24 hours, every hour rented, which no host achieves — scale it by the utilisation you expect. A card sitting idle earns nothing.Fees marked as published come from the platform’s terms; the rest are our estimates, labelled on each row.
Technical specifications
| FP64 | 0.5 TFLOPS |
|---|---|
| FP32 | 30 TFLOPS |
| TF32 | 60 TFLOPS |
| FP16/BF16 | 121 TFLOPS |
- Memory
- 24 GB
- Memory bandwidth
- 300 GB/s
- TDP
- 72 W
- INT8
- 242 TOPS
- Process
- 5 nm
- Die size
- 294 mm²
- Transistors
- 35.8 B
- Foundry
- TSMC
- Released
- 21 Mar 2023
| Architecture | Ada Lovelace |
|---|---|
| Memory | 24 GB GDDR6 |
| Bandwidth | 300 GB/s |
| Form factor | low-profile PCIe |
| Cooling | passive |
| Power | 72 W |
- Class
- Inference
- Cloud providers
- 10 tracked
- Price range
- $0.20 – $1.67 / GPU / hr
Other inference cards
The Compute
NVIDIA A10SNAPSHOT 2026-10-12T06:41:40.609517+00:00 · BACK TO THE FULL INDEX