---
title: "L40 vs L40S for Inference (Sep 2026)"
description: "Compare L40 and L40S cloud pricing, VRAM and estimated model fit using sourced offers."
canonical: https://vpsledger.com/gpu/l40-vs-l40s/
verified: 2026-10-23
format: text/markdown
html: https://vpsledger.com/gpu/l40-vs-l40s/
---

# L40 vs L40S for Inference (Sep 2026)

> Markdown alternate of <https://vpsledger.com/gpu/l40-vs-l40s/>. Generated at build time from the
> same dataset as the HTML page; prices verified 2026-10-23. Data is a build-time
> snapshot — no live API, no auth. Machine entry point: <https://vpsledger.com/ai/>.

[GPU cloud](/gpu/) / L40 vs L40S

# L40 vs L40S for inference

VRAM: [L40](/gpu/l40/) 48 GB; [L40S](/gpu/l40s/) 48 GB. 70B 4-bit estimated memory with 20% headroom: 42.0 GB. Throughput depends on model and workload; enter a measured rate in the [cost calculator](/tools/llm-inference-cost-calculator/).

| GPU | Provider | VRAM | GPU count | USD/hour | $/GB/hr | Billing | Type | Verified | Source |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [L40](/gpu/l40/) | [runpod](/gpu/providers/runpod/) | 48 GB | 1 | $0.82 | $0.017 | per-second | on-demand | 2026-09-26 | [Official price ↗](https://www.runpod.io/pricing) |
| [L40](/gpu/l40/) | [hyperstack](/gpu/providers/hyperstack/) | 48 GB | 1 | $1.00 | $0.021 | per-minute | on-demand | 2026-09-26 | [Official price ↗](https://www.hyperstack.cloud/gpu-pricing) |
| [L40S](/gpu/l40s/) | [runpod](/gpu/providers/runpod/) | 48 GB | 1 | $1.09 | $0.023 | per-second | on-demand | 2026-09-26 | [Official price ↗](https://www.runpod.io/pricing) |
| [L40S](/gpu/l40s/) | [modal](/gpu/providers/modal/) | 48 GB | 1 | $1.95GPU only | $0.041 | per-second | on-demand | 2026-09-26 | [Official price ↗](https://modal.com/pricing) |

Each price has an official source and verification date. Rates and availability change; affiliate links, if added later, will be marked sponsored.