L40S vs P4: Specs, VRAM & Price
Short answer: choose the L40S when 48 GB of VRAM is the gating requirement. Choose the P4 when 8 GB is enough for the workload and the lower hourly rate points that way.
The L40S has 48 GB of VRAM; the P4 has 8 GB.
You can rent both by the hour, no purchase required. L40S from $0.726/GPU/hr; P4 from $0.069/GPU/hr. The P4 is 90% cheaper at the entry rate.
L40S
Good for fine-tuning mid-size models and high-throughput inference
P4
Good for inference workloads on older-generation hardware
Last updated: 2026-10-10 22:00:31 UTCRefreshes hourly
Specs side by side
| Spec | L40S | P4 |
|---|---|---|
| VRAM | 48 GB | 8 GB |
| Memory bandwidth | 864 GB/s | – |
| Architecture | Ada Lovelace | Pascal |
| Compute capability | 8.9 | 6.1 |
| Precisions in hardware | FP32, FP16, BF16, FP8, INT4 | FP32, FP16 |
| Peak FP16/BF16 tensor TFLOPS | 362 TFLOPS | – |
| Peak FP8 tensor TFLOPS | 733 TFLOPS | – |
| Max board power (TDP) | 350W | – |
| Form factor | PCIe | PCIe |
| Regions | 5 | 1 |
On-demand price
| Rate | L40S | P4 |
|---|---|---|
| Lowest per GPU/hr | $1.07/GPU/hr | $0.069/GPU/hr (cheapest overall) |
Which is better for AI: L40S or P4?
Choose the L40S when 48 GB of VRAM is the gating requirement. Choose the P4 when 8 GB is enough for the workload and the lower hourly rate points that way.
L40S vs P4: common questions
Is the L40S or the P4 better for AI?
Short answer: choose the L40S when 48 GB of VRAM is the gating requirement. Choose the P4 when 8 GB is enough for the workload and the lower hourly rate points that way.
Is the L40S or the P4 cheaper to rent?
On Aquanode's live marketplace the L40S starts at $0.726/GPU/hr (median $1.20/GPU/hr) and the P4 starts at $0.069/GPU/hr (median $0.069/GPU/hr). The P4 is the cheaper of the two at the entry rate, by 90%.
What is the difference between the L40S and the P4?
The L40S has 48 GB of VRAM against the P4's 8 GB; the L40S is an Ada Lovelace part and the P4 is Pascal (compute capability 8.9 vs 6.1); only the L40S supports BF16, FP8, INT4 compute. On price, the L40S lists from $0.726/GPU/hr and the P4 from $0.069/GPU/hr.
How widely available are the L40S and the P4?
The L40S is listed in 5 regions and the P4 in 1 region right now.
How much does 1,000 GPU-hours cost on the L40S vs the P4?
At the lowest rates listed today, 1,000 GPU-hours costs $726 on the L40S and $69 on the P4, a difference of $657 for the same runtime. Rates are per GPU per hour and update hourly.
How much VRAM does the L40S have compared to the P4?
The L40S has 48 GB of VRAM; the P4 has 8 GB.
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 P4 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 P4 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 P4 comparisons
Related reading: L40S pricing and specs, and The best GPUs for AI, ranked.
