H100 vs L40S
Explore a head to head comparison of specifications, performance, and pricing.
H100
The NVIDIA H100 is a Hopper-based GPU that provides exceptional performance, scalability, and economics for AI, deep learning, and HPC workloads.
L40S
The NVIDIA L40S delivers high-performance computing capabilities for AI, machine learning, and data science applications.
H100 vs L40S: Which Should You Choose?
The H100 offers 80 GB of VRAM — 1.7× the 48 GB on the L40S — making it better suited for large model workloads that require holding more parameters in GPU memory. On FP16 throughput, the H100 delivers 267.6 TFLOPS versus 91.61 TFLOPS on the L40S — 3× faster for mixed-precision training and inference. Memory bandwidth favors the L40S at 0.86 TB/s compared to 0.00 TB/s on the H100, which directly impacts inference latency for memory-bandwidth-bound models. Architecturally, the H100 is built on Hopper while the L40S uses Ada Lovelace, reflecting different generational capabilities and optimizations. On Shadeform, the L40S starts from $0.74/hr versus $1.66/hr for the H100 — 124% more expensive — reflecting the performance premium. The H100 is available across 13 cloud providers on Shadeform compared to 7 for the L40S, giving more options for region and pricing flexibility.
H100 — Best Use Cases
- •Training large language models (7B–405B parameters)
- •High-throughput LLM inference
- •Mixture-of-experts and transformer workloads
- •Distributed multi-GPU training runs
Choose H100 when:
- ✓You need 80 GB+ VRAM for large models or long context windows
- ✓Maximum performance justifies the higher cost
- ✓You are training large models or running high-throughput inference
- ✓You need flexibility across multiple cloud providers or regions
L40S — Best Use Cases
- •LLM inference and model serving
- •Image generation and diffusion models
- •Smaller fine-tuning runs
- •Cost-efficient GPU compute
Choose L40S when:
- ✓48 GB VRAM is sufficient for your workload
- ✓Cost efficiency is your primary concern
- ✓Your workload does not require peak FP16 throughput
- ✓Your preferred provider already has availability
See how the H100 & L40S compare
Compare detailed hardware specifications and average pricing for the H100 and L40S.
Compare Hardware Specifications
| H100 | L40S | |
|---|---|---|
| GPU Type | H100 | L40S |
| VRAM per GPU | 80 GB | 48 GB |
| Manufacturer | NVIDIA | NVIDIA |
| Architecture | Hopper | Ada Lovelace |
| Interconnect | PCIe Gen5 or SXM5 | PCIe Gen4 |
| Memory Bandwidth | 3.35 TB/s | 864 GB/s |
| FP16 TFLOPS | 267.6 TFLOPS (4:1) | 91.61 TFLOPS (1:1) |
| CUDA Cores | 16896 | 18176 |
| Tensor Cores | 528 (4th Gen) | 568 (4th Gen) |
| RT Cores | N/A | 142 (3rd Gen) |
| Base Clock | 1365 MHz | 1110 MHz |
| Boost Clock | 1785 MHz | 2520 MHz |
| TDP | 350-700W | 350W |
| Process Node | TSMC 4N | TSMC 4N |
| Data Formats | FP8, INT8, BF16, FP16, TF32, FP32, FP64 | INT4, INT8, FP8, BF16, FP16, TF32, FP32 |
Compare Average On-Demand Pricing
| H100 | L40S | |
|---|---|---|
| 1 GPU | $2.85 /hr | $1.27 /hr |
| 2 GPUs | $5.19 /hr | $2.60 /hr |
| 4 GPUs | $9.79 /hr | $5.18 /hr |
| 8 GPUs | $19.35 /hr | $8.70 /hr |
Frequently Asked Questions: H100 vs L40S
The main differences are VRAM (80 GB vs 48 GB), FP16 throughput (267.6 vs 91.61 TFLOPS), architecture (Hopper vs Ada Lovelace). The H100 uses the Hopper architecture while the L40S is based on Ada Lovelace, giving each GPU different generational capabilities.
The H100 is generally better for large language model training due to its higher throughput and 80 GB of VRAM, which allows fitting larger models or larger batch sizes in a single pass. For smaller models or fine-tuning tasks where cost matters more, both GPUs can be effective.
On Shadeform, the L40S is available from $0.74/hr. The H100 starts from $1.66/hr. Prices vary by provider, region, and contract length. Reserved commitments can reduce hourly costs significantly compared to on-demand pricing.
The H100 has more VRAM at 80 GB, compared to 48 GB on the L40S. Higher VRAM allows you to run larger models without quantization, use longer context windows, and process larger batch sizes — all of which improve throughput and reduce latency for memory-bound workloads.
Based on TFLOPS per dollar, the H100 offers better raw compute value at current Shadeform on-demand rates. However, the best choice depends on your specific workload — if you need the extra VRAM or throughput of the L40S, paying the premium may be justified by faster job completion and lower total cost.
The H100 is currently available across 13 cloud providers on Shadeform's network, compared to 7 for the L40S. Shadeform lets you deploy either GPU across all available providers from a single platform, so you can always find available capacity without manually checking each cloud.
Mixing different GPU types in a single training cluster is generally not recommended, as it creates performance bottlenecks where faster GPUs wait for slower ones. For best results, use a homogeneous cluster of either H100 or L40S. Shadeform supports on-demand clusters of up to 64 GPUs of the same type with no commitment required.
Explore H100 & L40S Instances
Browse available instances with H100 and L40S GPUs. Filter by provider, availability, and more to find the perfect instance for your needs.
Explore more GPU comparisons
Select any two GPUs to compare their specifications and explore pricing across providers.
