L40 vs RTX 4090: Specs, VRAM & Price
Short answer: choose the L40 when 48 GB of VRAM is the gating requirement. Choose the RTX 4090 when 24 GB is enough for the workload and the lower hourly rate points that way.
The L40 has 48 GB of VRAM; the RTX 4090 has 24 GB.
You can rent both by the hour, no purchase required. L40 from $0.759/GPU/hr; RTX 4090 from $0.374/GPU/hr. The RTX 4090 is 51% cheaper at the entry rate.
L40
Good for fine-tuning mid-size models and high-throughput inference
RTX 4090
Good for inference and light fine-tuning workloads
Last updated: 2026-10-10 22:01:07 UTCRefreshes hourly
Specs side by side
| Spec | L40 | RTX 4090 |
|---|---|---|
| VRAM | 48 GB | 24 GB |
| Memory bandwidth | 864 GB/s | 1,008 GB/s |
| Architecture | Ada Lovelace | Ada Lovelace |
| Compute capability | 8.9 | 8.9 |
| Precisions in hardware | FP32, FP16, BF16, FP8, INT4 | FP32, FP16, BF16, FP8, INT4 |
| Peak FP16/BF16 tensor TFLOPS | 181.05 TFLOPS | – |
| Peak FP8 tensor TFLOPS | 362 TFLOPS | – |
| Max board power (TDP) | 300W | 450W |
| Form factor | – | PCIe 4.0 x16 |
| Regions | 3 | 3 |
On-demand price
| Rate | L40 | RTX 4090 |
|---|---|---|
| Lowest per GPU/hr | $0.850/GPU/hr | $0.761/GPU/hr |
Which is better for AI: L40 or RTX 4090?
Choose the L40 when 48 GB of VRAM is the gating requirement. Choose the RTX 4090 when 24 GB is enough for the workload and the lower hourly rate points that way.
L40 vs RTX 4090: common questions
Is the L40 or the RTX 4090 better for AI?
Short answer: choose the L40 when 48 GB of VRAM is the gating requirement. Choose the RTX 4090 when 24 GB is enough for the workload and the lower hourly rate points that way.
Is the L40 or the RTX 4090 cheaper to rent?
On Aquanode's live marketplace the L40 starts at $0.759/GPU/hr (median $0.902/GPU/hr) and the RTX 4090 starts at $0.374/GPU/hr (median $0.766/GPU/hr). The RTX 4090 is the cheaper of the two at the entry rate, by 51%.
What is the difference between the L40 and the RTX 4090?
The L40 has 48 GB of VRAM against the RTX 4090's 24 GB; both are Ada Lovelace parts; both run BF16, FP8, INT4 workloads in hardware. On price, the L40 lists from $0.759/GPU/hr and the RTX 4090 from $0.374/GPU/hr.
How widely available are the L40 and the RTX 4090?
The L40 is listed in 3 regions and the RTX 4090 in 3 regions right now.
How much does 1,000 GPU-hours cost on the L40 vs the RTX 4090?
At the lowest rates listed today, 1,000 GPU-hours costs $759 on the L40 and $374 on the RTX 4090, a difference of $385 for the same runtime. Rates are per GPU per hour and update hourly.
How much VRAM does the L40 have compared to the RTX 4090?
The L40 has 48 GB of VRAM; the RTX 4090 has 24 GB.
Does the L40 or the RTX 4090 support NVLink?
Neither the L40 nor the RTX 4090 has a 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 L40 page and RTX 4090 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 L40 VRAM calculator and the RTX 4090 VRAM calculator, see vendor-published throughput in the GPU benchmarks and specs table, or browse every model by generation in the GPU index.
More L40 comparisons
More RTX 4090 comparisons
Related reading: L40 pricing and specs, The NVIDIA Inception program, explained, Google Colab alternatives for dedicated GPU access, Free GPU credits for students and researchers, and RunPod volume disk vs. network volume, compared.
