L40S vs A100
Ada workstation silicon against Ampere datacenter silicon. The L40S has FP8 and the newer architecture; the A100 has 80 GB of HBM and the bandwidth to feed it. Which one a job needs, and what each costs to rent today.
- The L40S 48GB holds 48 GB to the A100 SXM4 80GB's 80 GB — 0.60× the capacity, which is the difference between a model fitting on one card and needing two.
- It rents for 1.05× the price cheapest-to-cheapest, across the 8 providers listing both.
- We have not measured either card on any workload. Everything above is the vendors' own peak figures beside today's published rates.
Violet marks the stronger figure. Throughput numbers are the vendors' own published dense peaks, read from L40S 48GB's product page and A100 SXM4 80GB's product page — not speeds we measured. Process node and transistor count are absent because they are not on the pages we read; architecture comes from our hand-compiled hardware registry, which has no fetch behind it.
On-demand, per GPU, USD, from providers publishing a comparable rate for both cards. 16 more list one of the two but not both. Spot and committed rates are excluded — they are not comparable to on-demand, and mixing them would turn a different product into what looks like a discount.
FP8 is present on one side and absent on the other
- The L40S 48GB publishes 733 dense TFLOPS at FP8. The A100 SXM4 80GB publishes no FP8 figure at all.
- That is an absent capability rather than a smaller number — the blank cell in the spec table is the silicon, not a gap in our data.
Where you rent moves more money than which card you pick
- The same L40S 48GB is published between $0.79 and $7.58 — a 9.6× spread across 18 providers for identical hardware.
- The gap between the two cards is 1.05×. Choosing the provider is worth more than choosing the card, and it is the part a spec sheet cannot tell you.
More pairings need both cards' specs fetched first.