Comparison · GPU vs GPU · read 00:32 UTCL40 vs A40
Specifications, today's prices from 28 providers, price per unit of memory and compute, and who rents each cheapest. Better values are highlighted; lower is better for prices.
Side by sideSpecs, prices, value
Highlighted cells are the better of the two: more memory and compute, lower prices, lower cost per unit.
| L40 | A40 | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 48 GB | 48 GB | 0% |
| Memory bandwidth | 0.86 TB/s | 0.70 TB/s | +24% |
| FP16 tensor | 181 TFLOPS | 150 TFLOPS | +21% |
| FP32 | 90.5 TFLOPS | 37.4 TFLOPS | +142% |
| Power | 300 W | 300 W | 0% |
| Prices today | |||
| Cheapest on-demand | $0.27 | $0.35 | -23% |
| Median on-demand | $0.88 | $0.90 | -2% |
| Highest on-demand | $2.58 | $1.71 | +51% |
| Cheapest spot | $0.33 | $0.19 | +79% |
| Cheapest reserved | $0.70 | $0.36 | +93% |
| Current offers (n) | 44 | 13 | +238% |
| Providers with a fresh offer | 18 | 10 | +80% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.006 | $0.007 | -23% |
| Per GB of VRAM · hour (median) | $0.018 | $0.019 | -2% |
| Per PFLOP FP16 · hour (median) | $4.882 | $5.991 | -19% |
| Per TB/s · hour (median) | $1.023 | $1.291 | -21% |
| Trend | |||
| 30-day median change | -2.3% | — | — |
| 90-day median change | -2.3% | +33.1% | — |
90 days · median on-demand price · rebased to 100How the two have moved
Both medians indexed to 100 ninety days ago, so the slopes compare even though the price levels differ.
Cheapest providers today · on-demand · current offersWho rents each one cheapest
- 1
Vast.aiTrust 782× $0.27 - 2
io.netTrust 424× $0.66 - 3
RunpodTrust 951× $0.69
- 1
GPURento1× $0.35 - 2Vast.aiTrust 782× $0.47
- 3
LeaderGPUTrust 598× $0.58
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the L40 if…
- the hourly price matters most: median $0.88 against $0.90, 2% less.
- you serve LLMs: memory bandwidth per dollar is better ($1.023 per TB/s-hour against $1.291).
- you want choice: 18 providers have a fresh offer today against 10.
Choose the A40 if…
- No metric above favours it today.
Questions about L40 vs A40
Which is cheaper to rent, the L40 or the A40?
Today the cheapest current on-demand L40 is $0.27 per GPU-hour and the cheapest A40 is $0.35; medians are $0.88 and $0.90. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the L40 and A40 have?
48 GB of GDDR6 on the L40, 48 GB of GDDR6 on the A40, at 0.86 and 0.70 TB/s.
Where do these numbers come from?
From 57 current offers read on the providers' own pages and APIs over the last 45 days (reference 00:32 UTC), or in our October 2026 market study for the few providers our collector does not read. Specifications come from manufacturer datasheets. Every price on the GPU pages links to its receipt and shows its read date.