L40 rental prices: cheapest $/hr today
The cheapest on-demand L40 rental today is $0.79/hr on Thunder Compute, updated 2026-10-12. 48 GB VRAM, NVIDIA.
Community cloud from $0.69 (RunPod), can be interrupted
Marketplace hosts from $0.72 (Lium, Montreal, Canada), one independent owner's server
Pricing at a glance
All current L40 offers
More filters: provider, memory, node size, stock, sort, price unit
| GPU / provider | Billing | $/GPU-hour | Node · total per hour | Rent |
|---|---|---|---|---|
| L40 · Thunder Compute48 GB | on-demand | $0.79 | – | Rent → |
| L40 · RunPod48 GB | secure cloud | $0.82 | – | Rent → |
| L40 · Massed Compute48 GB | on-demand | $0.86 | 1× $0.86 | Rent → |
| L40 · TensorDock48 GB | on-demand | $0.95 | – | Rent → |
| L40 · Hyperstack48 GB | on-demand | $1.00 | – | Rent →2026-10-12Details
|
| L40 · CoreWeave48 GB | on-demand | $1.25 | 8× $10.00 | Rent → |
| L40 · RunPod48 GB | community cloud | $0.69 | – | Rent → |
| L40 · Hyperstack48 GB | reserved / committedTerm and minimum volume not published in this table; confirm reservation quoteStarting rate | $0.70 | – | Rent → |
| L40 · Lium45 GB · 1-GPU host | marketplace host | $0.72 | 1× $0.72 | In stockRent →2026-10-12Details
|
Try another provider or reduce the minimum memory.
Tables rank by price, lowest first. Commercial relationships never affect ordering. How we collect and rank prices.
Same GPU, provider, billing and price on several hosts or listings: one row, with the count. – means the provider does not state it.
Monthly view assumes 730 hours of continuous use. Daily view assumes 24 hours.
On-demand and secure-cloud rates are capacity the provider runs, with no reservation. Spot and community capacity can be interrupted. A marketplace host is one independent owner's server listed on a marketplace. Prices are per GPU per hour, before storage and data transfer. Every row shows the date it was checked.
L40 hardware specs
| Architecture | Ada Lovelace |
|---|---|
| VRAM | 48 GB |
| FP16 TFLOPS (dense) | 181 |
| Memory bandwidth | 864 GB/s |
| TDP | 300 W |
| Interconnect | PCIe Gen4 |
| Launched | 2022 |
Datasheet spec for the L40 variant. Source: www.nvidia.com.
Compare L40 configurations
L40 purchase prices · Rent or buy calculator · Compare two GPUs
- RunPod L40 pricing
- Hyperstack L40 pricing
- Lium L40 pricing
- Thunder Compute L40 pricing
- Massed Compute L40 pricing
- TensorDock L40 pricing
- CoreWeave L40 pricing
LLM memory estimates for this GPU
These models have an estimated inference footprint within 48 GB at one or more precisions. The scenario uses 4,096 tokens, batch 1, a 16-bit KV cache and 20% runtime headroom. A memory fit does not establish speed or software compatibility.
| Model | 16-bit GB | 8-bit GB | 4-bit GB |
|---|---|---|---|
| Qwen2.5 72B | 176.1 | 88.9 | 45.2 ✓ |
| DeepSeek R1 Distill Llama 70B | 170.9 | 86.3 | 43.9 ✓ |
| Mixtral 8x7B v0.1 | 112.7 | 56.7 | 28.7 ✓ |
| Qwen2.5 32B | 79.9 | 40.6 ✓ | 20.9 ✓ |
| Qwen2.5 Coder 32B | 79.9 | 40.6 ✓ | 20.9 ✓ |
| DeepSeek R1 Distill Qwen 32B | 79.9 | 40.6 ✓ | 20.9 ✓ |
| Qwen3 32B | 79.9 | 40.6 ✓ | 20.9 ✓ |
| Qwen3 30B A3B | 73.8 | 37.1 ✓ | 18.8 ✓ |
| Mistral Small 24B 2501 | 57.4 | 29.1 ✓ | 14.9 ✓ |
| Qwen2.5 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| Qwen2.5 Coder 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| DeepSeek R1 Distill Qwen 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| Qwen3 14B | 36.2 ✓ | 18.5 ✓ | 9.7 ✓ |
| phi 4 | 36.2 ✓ | 18.6 ✓ | 9.8 ✓ |
| Mistral Nemo 2407 | 30.2 ✓ | 15.5 ✓ | 8.2 ✓ |
| Qwen3 8B | 20.4 ✓ | 10.6 ✓ | 5.6 ✓ |
| DeepSeek R1 Distill Llama 8B | 19.9 ✓ | 10.3 ✓ | 5.5 ✓ |
| Qwen2.5 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| Qwen2.5 Coder 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| DeepSeek R1 Distill Qwen 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| Mistral 7B v0.3 | 18.0 ✓ | 9.3 ✓ | 5.0 ✓ |
| Qwen3 4B | 10.4 ✓ | 5.6 ✓ | 3.1 ✓ |
| Qwen2.5 3B | 7.6 ✓ | 3.9 ✓ | 2.0 ✓ |
| Qwen2.5 Coder 3B | 7.6 ✓ | 3.9 ✓ | 2.0 ✓ |
| Qwen3 1.7B | 5.4 ✓ | 3.0 ✓ | 1.8 ✓ |
| DeepSeek R1 Distill Qwen 1.5B | 4.4 ✓ | 2.3 ✓ | 1.2 ✓ |
| Qwen2.5 1.5B | 3.8 ✓ | 2.0 ✓ | 1.1 ✓ |
| Qwen2.5 Coder 1.5B | 3.8 ✓ | 2.0 ✓ | 1.1 ✓ |
| Qwen3 0.6B | 2.4 ✓ | 1.5 ✓ | 1.0 ✓ |
| Qwen2.5 0.5B | 1.2 ✓ | 0.7 ✓ | 0.4 ✓ |
| Qwen2.5 Coder 0.5B | 1.2 ✓ | 0.7 ✓ | 0.4 ✓ |
Compare model memory requirements or change context and memory assumptions on a model's page.
When to choose L40
Use the model memory estimates above to check whether your weights, context and batch size fit within 48 GB. Then compare the exact variant, node size and software support. Capacity alone does not establish inference speed or training throughput.
If your workload fits within 32 GB, compare RTX PRO 4500 rental prices from $0.72/GPU-hour. The smaller memory budget is the trade-off; price alone does not make the cards interchangeable.
L40 price changes
- 30 days: 0.0% across 6 matching firm-price configurations since 2026-09-12.
- 90 days: 0.0% across 4 matching firm-price configurations since 2026-07-14.
Changes use the geometric mean of price ratios for the same provider, variant, tier, GPU count, instance and region. New listings are not counted as price changes.
Price history
on-demand rate per day · median on-demand rate per day (dashed), since 2026-07-05: $0.79 low · $0.82 high. Full ledger on the history hub.
FAQ
The cheapest L40 rental we have is $0.79 per GPU-hour on Thunder Compute (on-demand), updated 2026-10-12.
48 GB.
Thunder Compute has the cheapest verified on-demand rate today at $0.79 per GPU-hour; interruptible spot/community capacity starts at $0.69. Verified 2026-10-12.
Related
All 48 GB+ GPUs · GPUs under $1/hr · Price movers this week · Thunder Compute review · A10 vs L40 · A30 vs L40 · A40 vs L40
Should you rent or buy?
Compare L40 purchase prices and quote availability, then enter your usage and electricity costs in the rent-versus-buy calculator. Purchase listings have their own variants, stock status and update dates.