L40S vs V100: Specs, VRAM & Price
Short answer: choose the L40S when 48 GB of VRAM is the gating requirement. Choose the V100 when 32 GB is enough for the workload and the lower hourly rate points that way.
The L40S has 48 GB of VRAM; the V100 has 32 GB.
You can rent both by the hour, no purchase required. L40S from $0.844/GPU/hr; V100 from $0.110/GPU/hr. The V100 is 87% cheaper at the entry rate.
L40S
Good for fine-tuning mid-size models and high-throughput inference
V100
Good for inference workloads on older-generation hardware
Last updated: 2026-11-03 UTCRefreshes every 6 hours
Specs side by side
| Spec | L40S | V100 |
|---|---|---|
| VRAM | 48 GB | 32 GB |
| Memory bandwidth | 864 GB/s | 900 GB/s |
| Architecture | Ada Lovelace | Volta |
| Compute capability | 8.9 | 7.0 |
| Precisions in hardware | FP32, FP16, BF16, FP8, INT4 | FP32, FP16 |
| Peak FP16/BF16 tensor TFLOPS | 362 TFLOPS | 125 TFLOPS (SXM2), 112 TFLOPS (PCIe) |
| Peak FP8 tensor TFLOPS | 733 TFLOPS | – |
| NVLink / multi-GPU interconnect | – | NVLink, 300 GB/s (SXM2 only) |
| Max board power (TDP) | 350W | 300W (SXM2), 250W (PCIe) |
| Form factor | PCIe | PCIe 3.0 x16 |
| Regions | 4 | 2 |
On-demand price
| Rate | L40S | V100 |
|---|---|---|
| Lowest per GPU/hr | $1.07/GPU/hr | $0.110/GPU/hr (cheapest overall) |
Which is better for AI: L40S or V100?
Choose the L40S when 48 GB of VRAM is the gating requirement. Choose the V100 when 32 GB is enough for the workload and the lower hourly rate points that way.
L40S vs V100: common questions
Is the L40S or the V100 better for AI?
Short answer: choose the L40S when 48 GB of VRAM is the gating requirement. Choose the V100 when 32 GB is enough for the workload and the lower hourly rate points that way.
Is the L40S or the V100 cheaper to rent?
On Aquanode's live marketplace the L40S starts at $0.844/GPU/hr (median $2.36/GPU/hr) and the V100 starts at $0.110/GPU/hr (median $0.220/GPU/hr). The V100 is the cheaper of the two at the entry rate, by 87%.
What is the difference between the L40S and the V100?
The L40S has 48 GB of VRAM against the V100's 32 GB; the L40S is an Ada Lovelace part and the V100 is Volta (compute capability 8.9 vs 7.0); only the L40S supports BF16, FP8, INT4 compute. On price, the L40S lists from $0.844/GPU/hr and the V100 from $0.110/GPU/hr.
How widely available are the L40S and the V100?
The L40S is listed in 4 regions and the V100 in 2 regions right now.
How much does 1,000 GPU-hours cost on the L40S vs the V100?
At the lowest rates listed today, 1,000 GPU-hours costs $844 on the L40S and $110 on the V100, a difference of $734 for the same runtime. Rates are per GPU per hour and update hourly.
How much VRAM does the L40S have compared to the V100?
The L40S has 48 GB of VRAM; the V100 has 32 GB.
Does the L40S or the V100 support NVLink?
The V100 supports NVLink, 300 GB/s (SXM2 only); the L40S has no dedicated GPU-to-GPU interconnect.
How this comparison is calculated
Every price is normalized to a per-GPU hourly rate using the same pipeline as every other pricing surface on Aquanode, and only the cheapest qualifying offer per provider is shown. A dash means that provider does not currently list that GPU.
Architecture, compute capability and supported precisions come from each vendor's published datasheet for that generation, not from the marketplace feed. This page regenerates at most once per hour.
Memory bandwidth, tensor TFLOPS, NVLink/interconnect and max board power are the same verified-from-datasheet figures shown on each model's own L40S page and V100 page, which link the primary vendor source for that model. A dash here means Aquanode hasn't verified that figure from a primary source yet, never a guess.
Go further
Check what fits in each card's memory with the L40S VRAM calculator and the V100 VRAM calculator, see vendor-published throughput in the GPU benchmarks and specs table, or browse every model by generation in the GPU index.
More L40S comparisons
More V100 comparisons
Related reading: L40S pricing and specs, and The best GPUs for AI, ranked.
