datacenter
Rent Nvidia L40S.
Ada Lovelace
48GB VRAM
350W
From $0.28/hr
Per hour
$0.28
Per day
$6.72
Per week
$47.04
Per month
$202
Provider spread
10 providers ·
up to 84% cheaper at the low end
Cheapest · $0.28/hr on Novita AI
Median $0.92/hr
Most expensive · $1.80/hr on AceCloud
Price history
Daily median across providers.
Loading...
Where to rent it
All providers carrying this GPU.
Buy vs rent
Should you rent or own?
Your usage
Rent on cloud
Per day
—
Per month
—
at $0.28/hr cheapest provider rate
Own on-prem
Electricity per day
—
350W TDP
Rent breakeven on $9,000 MSRP
— days
· — months
Workloads
Suitable workloads.
AI models that fit
See all 28 →
Run these on this GPU.
- DeepSeek R1 Distill Qwen 14B 1× · fp16
- FLUX.1 Schnell 1× · fp16
- Gemma 3 27B 1× · fp16
Cloud instances
See all 6 →
Hyperscaler bundles.
Pre-configured on 6 clouds — from $0.28/hr total
($0.28/hr per GPU).
FAQ
Frequently asked.
How is the Nvidia L40S price calculated?
We pull live listings from each provider's public API, take the median hourly rate across active offers, and refresh every hour. The rate shown is the median, so a single low-ball spot offer can't distort the headline.
Why does the Nvidia L40S cost different amounts on different providers?
P2P marketplaces like Vast.ai aggregate offers from individual hosts who set their own rates — bidding pushes prices down. First-party clouds (Lambda, hyperscalers) charge a managed-service premium for support, SLAs, and integrated networking. Decentralized networks (io.net, Akash) settle in tokens, which adds volatility but often the lowest base rate.
Can I really train an LLM on a single Nvidia L40S?
Depends on the model size. With 48GB of VRAM you can fine-tune 13B–34B models with LoRA, or run inference on 70B models at int4 quantization.
Spot vs on-demand on the Nvidia L40S — which should I rent?
On-demand keeps the same instance until you stop it; spot (or interruptible) is cheaper but the host can reclaim it when a higher-paying job lands. Use on-demand for training runs and anything stateful. Use spot for stateless inference, batch jobs, and experiments where a checkpoint every few minutes is enough to recover.
How is hourly billing measured for the Nvidia L40S?
Most providers bill per-second once the instance is running, with a small minimum (often 60 seconds). A handful of first-party clouds round up to the minute. Either way, headline $/hr is the right comparison unit.
Does the region of the host affect the Nvidia L40S price?
Yes — US and EU regions usually carry a premium over LATAM, India, and parts of APAC, especially on first-party clouds. P2P marketplaces hide this behind one global price because supply moves wherever bids exist.
Cloud instance options
Pre-configured instances on hyperscalers.
Whole-instance bundles (GPU + vCPU + RAM + disk) on the major clouds. Per-GPU rate often drops as the count rises. View = spec page · Launch = sign up (affiliate).
Cheapest bundle
$0.28/hr
Lowest $/hr per GPU
$0.28/hr
Providers
6
Instance shapes
6
| Provider | Instance | GPUs | vCPU | RAM | Disk | $/hr | $/hr per GPU | |
|---|---|---|---|---|---|---|---|---|
| novita-l40s-spot | 1× | — | — | — | $0.28/hr | $0.28/hr | ||
| rcrate-l40s | 1× | — | — | — | $1.10/hr | $1.10/hr | ||
| l40s-pcie | 1× | — | — | — | $1.30/hr | $1.30/hr | ||
| gpu-l40sx1 | 1× | — | — | — | $1.57/hr | $1.57/hr | ||
|
V
Vultr
|
vc2-gpu-l40s | 1× | — | — | — | $1.70/hr | $1.70/hr | |
| ace-l40s-on-demand | 1× | — | — | — | $1.80/hr | $1.80/hr |
Looking for the cheapest rate?
Hyperscaler bundles include managed networking + SLAs. Raw per-GPU rental on P2P marketplaces is typically 3–10× cheaper.
See raw rental rates on the Overview tab →
FAQ
Cloud instances — common questions.
How is the Nvidia L40S price calculated?
We pull live listings from each provider's public API, take the median hourly rate across active offers, and refresh every hour. The rate shown is the median, so a single low-ball spot offer can't distort the headline.
Why does the Nvidia L40S cost different amounts on different providers?
P2P marketplaces like Vast.ai aggregate offers from individual hosts who set their own rates — bidding pushes prices down. First-party clouds (Lambda, hyperscalers) charge a managed-service premium for support, SLAs, and integrated networking. Decentralized networks (io.net, Akash) settle in tokens, which adds volatility but often the lowest base rate.
Can I really train an LLM on a single Nvidia L40S?
Depends on the model size. With 48GB of VRAM you can fine-tune 13B–34B models with LoRA, or run inference on 70B models at int4 quantization.
Spot vs on-demand on the Nvidia L40S — which should I rent?
On-demand keeps the same instance until you stop it; spot (or interruptible) is cheaper but the host can reclaim it when a higher-paying job lands. Use on-demand for training runs and anything stateful. Use spot for stateless inference, batch jobs, and experiments where a checkpoint every few minutes is enough to recover.
How is hourly billing measured for the Nvidia L40S?
Most providers bill per-second once the instance is running, with a small minimum (often 60 seconds). A handful of first-party clouds round up to the minute. Either way, headline $/hr is the right comparison unit.
Does the region of the host affect the Nvidia L40S price?
Yes — US and EU regions usually carry a premium over LATAM, India, and parts of APAC, especially on first-party clouds. P2P marketplaces hide this behind one global price because supply moves wherever bids exist.
AI models
FP16
FP16
FP16
FP8
FP8
FP8
FP8
FP8
FP8
FP8
FP8
FP8
FP8
FP8
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
INT4
Models that run on this GPU.
GPU-count + quantization recommendations covering fine-tuning, inference, and run-it-yourself scenarios on the Nvidia L40S.
DeepSeek R1 Distill Qwen 14B
by DeepSeek
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
15B
Context
128K tokens
FLUX.1 Schnell
by Black Forest Labs
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
12B
Gemma 3 27B
by Google DeepMind
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
27B (27B active)
Context
128K tokens
Qwen 2.5 Coder 32B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
33B
Context
128K tokens
DeepSeek R1 Distill Qwen 32B
by DeepSeek
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
33B
Context
128K tokens
Qwen 2.5 32B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
33B
Context
128K tokens
Qwen 3 32B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
33B
Context
128K tokens
LLaVA 34B
by LLaVA Project
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
34B
Context
4K tokens
Code Llama 34B
by Meta AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
34B
Context
16K tokens
DeepSeek Coder 33B
by DeepSeek
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
33B
Context
16K tokens
Yi-34B
by 01.AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
34B
Context
32K tokens
Qwen: Qwen3.6 35B A3B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
35B
Context
262K tokens
Qwen: Qwen3.5-35B-A3B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
35B
Context
262K tokens
TheDrummer: Skyfall 36B V2
by Thedrummer
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
36B
Context
33K tokens
Llama 3.3 70B
by Meta AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
128K tokens
Qwen 2.5 72B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
73B
Context
128K tokens
DeepSeek R1 Distill Llama 70B
by DeepSeek
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
128K tokens
Llama 3.1 70B
by Meta AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
128K tokens
Code Llama 70B
by Meta AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
16K tokens
Hermes 3 70B
by Nous Research
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
128K tokens
Nous: Hermes 4 70B
by Nous Research
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
131K tokens
Qwen: Qwen2.5 VL 72B Instruct
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
72B
Context
131K tokens
Sao10K: Llama 3.1 70B Hanami x1
by Sao10k
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
16K tokens
Sao10K: Llama 3.3 Euryale 70B
by Sao10k
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
131K tokens
Magnum v4 72B
by Anthracite Org
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
72B
Context
33K tokens
Sao10K: Llama 3.1 Euryale 70B v2.2
by Sao10k
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
131K tokens
Sao10k: Llama 3 Euryale 70B v2.1
by Sao10k
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
8K tokens
Meta: Llama 3 70B Instruct
by Meta AI
Required GPUs
1× Nvidia L40S
Total VRAM
48 GB
Parameters
70B
Context
8K tokens
Renting for inference?
Pair these models with the cheapest provider on the rental table.
See rental rates on the Overview tab →
FAQ
AI models on this GPU.
How is the Nvidia L40S price calculated?
We pull live listings from each provider's public API, take the median hourly rate across active offers, and refresh every hour. The rate shown is the median, so a single low-ball spot offer can't distort the headline.
Why does the Nvidia L40S cost different amounts on different providers?
P2P marketplaces like Vast.ai aggregate offers from individual hosts who set their own rates — bidding pushes prices down. First-party clouds (Lambda, hyperscalers) charge a managed-service premium for support, SLAs, and integrated networking. Decentralized networks (io.net, Akash) settle in tokens, which adds volatility but often the lowest base rate.
Can I really train an LLM on a single Nvidia L40S?
Depends on the model size. With 48GB of VRAM you can fine-tune 13B–34B models with LoRA, or run inference on 70B models at int4 quantization.
Spot vs on-demand on the Nvidia L40S — which should I rent?
On-demand keeps the same instance until you stop it; spot (or interruptible) is cheaper but the host can reclaim it when a higher-paying job lands. Use on-demand for training runs and anything stateful. Use spot for stateless inference, batch jobs, and experiments where a checkpoint every few minutes is enough to recover.
How is hourly billing measured for the Nvidia L40S?
Most providers bill per-second once the instance is running, with a small minimum (often 60 seconds). A handful of first-party clouds round up to the minute. Either way, headline $/hr is the right comparison unit.
Does the region of the host affect the Nvidia L40S price?
Yes — US and EU regions usually carry a premium over LATAM, India, and parts of APAC, especially on first-party clouds. P2P marketplaces hide this behind one global price because supply moves wherever bids exist.