Flopper.io Flopper.io
vs

Google TPU v5e 16GB vs NVIDIA L40S 48GB

The L40S delivers 84% more BF16 throughput than the TPU v5e (362 vs 197 TFLOPS dense). The L40S also carries 32 GB more memory (48 GB vs 16 GB).

TPU v5e: TPU, 2023 L40S: Ada Lovelace, 2023
BF16 dense lead
+84%
L40S: 362 vs 197 TFLOPS
Memory
16 vs 48 GB
32 GB more for the L40S
Bandwidth
800 GB/s vs 864 GB/s
L40S moves data faster
See rental pricing from $0.60/hr Download comparison sheet

Share this page

Google TPU v5e 16GB vs NVIDIA L40S 48GB

A post with this link shows the card on the Image tab as its preview.

Post on X Share on LinkedIn
 Printable version →

Google TPU v5e 16GB

Full specs →

What is the Google TPU v5e?

The TPU v5e is a TPU from Google on the TPU architecture, an accelerator built for AI training and inference. TPU v5e full specifications →

Can I rent the TPU v5e in the cloud?

Flopper is not tracking any live rental listings for the TPU v5e yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the TPU v5e compare with other accelerators?

This page sets the TPU v5e against the L40S. Both parts launched in the same year. This is a cross-vendor matchup, Google against NVIDIA. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the TPU v5e figures come from?

Every TPU v5e figure on this page traces to Google's own datasheet or specification table. How Flopper verifies specifications →

NVIDIA L40S 48GB

Leads on BF16 Full specs →

What is the NVIDIA L40S?

The L40S is a GPU from NVIDIA on the Ada Lovelace architecture, an accelerator for AI training, inference and high-performance computing. It is a PCIe add-in card, so it fits standard servers and workstations. L40S full specifications →

Can I rent the L40S in the cloud?

Yes. Cloud providers tracked by Flopper rent the L40S by the hour, and the pricing page compares their current rates and the price history side by side. L40S rental prices →

How does the L40S compare with other accelerators?

This page sets the L40S against the TPU v5e. Both parts launched in the same year. This is a cross-vendor matchup, NVIDIA against Google. Flopper also compares it head to head with the ATOM-Max, across every published precision, memory and power. L40S vs ATOM-Max →

Where do the L40S figures come from?

Every L40S figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

Specifications

TPU v5e
L40S
Architecture
TPUAda Lovelace
Launch Year
20232023
Form Factor
—PCIe
Memory
16 GB48 GB
Memory Bandwidth
800 GB/s864 GB/s
TDP
—350 W
Process Node
—4nm

Performance (TFLOPS)

TPU v5e
L40S
No verified data available1.4 TFLOPS
No verified data available92 TFLOPS
No verified data available183 TFLOPS
366 TFLOPS sparse
197 TFLOPS 362 TFLOPS
733 TFLOPS sparse
No verified data available362 TFLOPS
733 TFLOPS sparse
No verified data available733 TFLOPS
1,466 TFLOPS sparse
FP6
No verified data available No verified data available
No verified data available No verified data available
393 TOPS 733 TOPS
1,466 TOPS sparse

Figures are dense. Where a vendor publishes a figure it sits under the dense one and never decides a winner. Hover any precision for its definition, or read the GPU glossary.

FLOPS by Precision

What actually differs

Both launched in 2023, the TPU v5e on TPU and the L40S on Ada Lovelace.

The L40S carries 32 GB more memory (48 GB against 16 GB), so larger models fit on one GPU before you shard. Bandwidth is 800 GB/s against 864 GB/s, which sets the ceiling on decode speed.

Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP64, FP32, TF32, BF16, FP16, FP8, INT8, Sparsity and TFLOPS. For how the generations relate, see NVIDIA architectures explained.

Frequently asked questions

Is the TPU v5e faster than the L40S?

At BF16 precision the L40S reaches 362 TFLOPS dense against 197 TFLOPS for the TPU v5e. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the TPU v5e or the L40S?

Choose the L40S for training and high-throughput inference: it has 84% more BF16 throughput than the TPU v5e. It also carries 32 GB more memory, so larger models fit on one GPU before you shard.

Which has more memory, the TPU v5e or the L40S?

The L40S carries 48 GB of memory versus 16 GB for the TPU v5e. Memory bandwidth is 800 GB/s for the TPU v5e and 864 GB/s for the L40S.

How much do the TPU v5e and the L40S cost to rent?

Live cloud listings tracked by Flopper price no live listings yet for the TPU v5e, and the L40S from $0.60 per GPU-hour across 23 providers. The rental section on this page links to every provider, rate and the price history for each GPU.

Where to Rent

Google TPU v5e 16GB

No live rental listings tracked for the TPU v5e yet.

Browse GPU cloud providers

NVIDIA L40S 48GB

Rent from

$0.60 / GPU-hr

Lowest of 40 live listings across 23 providers · median $1.46

See all L40S rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.

Open in Full Compare Tool → TPU v5e full specs L40S full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI