---
title: "A40 Cloud Pricing (Sep 2026): 1 Offer"
description: "Rent A40 cloud GPU: sourced hourly prices, 48 GB VRAM model-fit estimates, and inference cost inputs."
canonical: https://vpsledger.com/gpu/a40/
verified: 2026-10-23
format: text/markdown
html: https://vpsledger.com/gpu/a40/
---

# A40 Cloud Pricing (Sep 2026): 1 Offer

> Markdown alternate of <https://vpsledger.com/gpu/a40/>. 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/) / A40

# A40 cloud pricing

48 GB VRAM per GPU. Compare 1 sourced offers. Use measured tokens/second from your own model and setup to calculate per-million-token costs.

| GPU | Provider | VRAM | GPU count | USD/hour | $/GB/hr | Billing | Type | Verified | Source |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [A40](/gpu/a40/) | [runpod](/gpu/providers/runpod/) | 48 GB | 1 | $0.49 | $0.010 | per-second | on-demand | 2026-09-26 | [Official price ↗](https://www.runpod.io/pricing) |

## What fits in VRAM? (estimate)

4-bit weights with 20% headroom; real KV cache, context length and framework overhead vary. Multi-GPU memory is not automatically pooled.

- 7B: ~4.2 GB — fits by this estimate
- 13B: ~7.8 GB — fits by this estimate
- 30B: ~18.0 GB — fits by this estimate
- 70B: ~42.0 GB — fits by this estimate

[Adjust precision and headroom →](/tools/gpu-vram-calculator/)

## Inference throughput and cost per 1M tokens

No comparable public throughput benchmark has been verified for this card. Enter your own measured tokens/second; the site does not assume a speed.

[Calculate cost per million tokens →](/tools/llm-inference-cost-calculator/?gpu=a40)

## GPU comparisons

A40 vs RTX A6000

A10 vs A40

Prices can change. Each offer links its official source and verification date. Affiliate links, if added later, will be marked sponsored.