datacenter
Nvidia L40 mieten.
Ada Lovelace
48GB VRAM
300W
From $0.64/hr
Per hour
$0.64
Per day
$15.40
Per week
$107.81
Per month
$462
Provider spread
2 providers ·
up to 36% cheaper at the low end
Cheapest · $0.64/hr on RunPod
Median $0.82/hr
Most expensive · $1.00/hr on Hyperstack
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 $0.64/hr cheapest provider rate
Own on-prem
Electricity per day
—
300W TDP
Rent breakeven on $7,900 MSRP
— days
· — months
AI models that fit
See all 6 →
Run these on this GPU.
- Qwen 2.5 72B 1× · int4
- Qwen2.5 72B Instruct 1× · int4
- Qwen2 72B Instruct 1× · int4
Cloud instances
See all 1 →
Hyperscaler bundles.
Pre-configured on 1 clouds — from $1.00/hr total
($1.00/hr per GPU).
FAQ
Frequently asked.
How is the Nvidia L40 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 L40 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 L40?
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 L40 — 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 L40?
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 L40 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
$1.00/hr
Lowest $/hr per GPU
$1.00/hr
Providers
1
Instance shapes
1
| Provider | Instance | GPUs | vCPU | RAM | Disk | $/hr | $/hr per GPU | |
|---|---|---|---|---|---|---|---|---|
| l40-on-demand | 1× | — | — | — | $1.00/hr | $1.00/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 L40 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 L40 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 L40?
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 L40 — 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 L40?
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 L40 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
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 L40.
Qwen 2.5 72B
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
73B
Context
128K tokens
Qwen2.5 72B Instruct
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
72B
Context
33K tokens
Qwen2 72B Instruct
by Togethercomputer
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
72B
Context
33K tokens
Qwen2.5 72B Instruct Turbo
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
72B
Context
131K tokens
Qwen2-VL (72B) Instruct
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
72B
Context
33K tokens
Qwen: Qwen2.5 VL 72B Instruct
by Alibaba (Qwen Team)
Required GPUs
1× Nvidia L40
Total VRAM
48 GB
Parameters
72B
Context
131K 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 L40 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 L40 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 L40?
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 L40 — 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 L40?
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 L40 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.