L40S vs H100
L40S is 62% cheaperComparing the L40S and H100 for cloud rental? The L40S is available from 5 providers starting at $0.43/hr, while the H100 starts at $1.14/hr across 8 providers. Below is a live breakdown of pricing, VRAM, compute throughput, and where to rent each GPU.
L40S vs H100 at a glance
Spec profile
Each axis is scaled to the stronger GPU — the outer edge is better.
Value for money
Specs are manufacturer peak FP16 Tensor figures; real-world throughput varies by workload, precision, and software stack.
L40S vs H100 specs & pricing
| Specification | L40S | H100 |
|---|---|---|
| Cheapest price | $0.43/hr | $1.14/hr |
| VRAM | 48 GB | 80 GB |
| FP16 TFLOPS | 366 | 495 |
| Memory bandwidth | 864 GB/s | 3350 GB/s |
| Release year | 2023 | 2023 |
| Cloud providers | 5 | 8 |
Which should you choose?
- •L40S is the cheaper option to rent — about 62% less per GPU-hour ($0.43/hr vs $1.14/hr).
- •H100 has more VRAM (80 GB), so it fits larger models and bigger batch sizes without sharding.
- •H100 offers higher FP16 throughput, training and serving faster on compute-bound workloads.
Provider availability
Frequently asked questions
Is the L40S or H100 cheaper to rent?
L40S is currently cheaper, from $0.43/hr versus $1.14/hr for the H100 — roughly 62% less per GPU-hour.
How much VRAM do the L40S and H100 have?
The L40S has 48 GB of VRAM and the H100 has 80 GB. More VRAM lets you serve larger models and longer context windows on a single GPU.
Which is faster, the L40S or H100?
On paper the H100 is faster, with 495 FP16 TFLOPS versus 366. Real-world speed also depends on memory bandwidth and software stack.
Which cloud providers offer both the L40S and H100?
Both GPUs are available from AWS, Nebius, OCI, Verda. PlanetGPU tracks live L40S and H100 prices across every major cloud.
Live prices come from public provider catalogs and can change quickly. How we collect prices.
Related GPU comparisons
Explore related GPU pages
Crawlable paths from this comparison into prices, model fit, and neighboring comparisons.