Vast.aiIn stock
On-Demand from $0.37
From
$0.37 / GPU / hr
On-Demand · California, USVisit website →
Rental price index
NVIDIA A100 80GB
🏢The previous-generation workhorse, still widely rented and steadily cheaper.
Prices collected 2026-10-11 · 35 providers, 109 configurations
On-demandOn-demand (archive, Lambda only)ReservedSpot
Index median price per GPU per hour, by billing type. Nothing is interpolated.
At a glance
Median on-demand
$1.79/GPU/hr
Cheapest on-demand
$0.37Vast.ai
Vast.aiFloor, any billing
$0.37incl. spot
Per month at 720h
$1,289
NVIDIA A100 80GB pricing by provider
One card per provider, cheapest first. A configuration is one priced unit — an instance type, a SKU, a marketplace offer — and the per-GPU price is that unit's price divided by its GPU count. Open a card to see every configuration with its vCPU, RAM and region where the provider publishes them.
35 providers · 100 configurations
GPU.aiIn stock
AEUAE2 configs1xcollected 2026-10-11
On-Demand from $1.07
From
$1.07 / GPU / hr
On-DemandVisit website →
Thunder ComputeIn stock
USUSA3 configs1xcollected 2026-10-11
On-Demand from $1.09
From
$1.09 / GPU / hr
On-DemandVisit website →
SeewebIn stock
ITItaly1 config1xcollected 2026-10-11
On-Demand from $1.11
From
$1.11 / GPU / hr
On-DemandVisit website →
RunPod Community CloudIn stock
USUSA2 configs1xcollected 2026-10-11
On-Demand from $1.19
From
$1.19 / GPU / hr
On-DemandVisit website →
io.netIn stock
USUSA5 configs1x–8xcollected 2026-10-11
On-Demand from $1.29
From
$1.29 / GPU / hr
On-DemandVisit website →AkashIn stock
USUSA2 configs1xcollected 2026-10-11
On-Demand from $1.33
From
$1.33 / GPU / hr
On-DemandVisit website →Massed ComputeIn stock
USUSA2 configs1xcollected 2026-10-11
On-Demand from $1.35
From
$1.35 / GPU / hr
On-DemandVisit website →
HyperstackIn stock
GBUK3 configs1xcollected 2026-10-11
On-Demand from $1.35Spot from $1.08Reserved from $0.95
From
$1.35 / GPU / hr
On-DemandVisit website →
BeamIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $1.36
From
$1.36 / GPU / hr
On-DemandVisit website →
Packet.aiIn stock
1 config1xcollected 2026-10-11
On-Demand from $1.43
From
$1.43 / GPU / hr
On-DemandVisit website →
Denvr DataworksIn stock
CACanada1 config1xcollected 2026-10-11
On-Demand from $1.45
From
$1.45 / GPU / hr
On-DemandVisit website →
OblivusIn stock
GBUK7 configs1x–8xcollected 2026-10-11
On-Demand from $1.47
From
$1.47 / GPU / hr
On-DemandVisit website →SpheronIn stock
1 config1xcollected 2026-10-11
On-Demand from $1.48
From
$1.48 / GPU / hr
On-DemandVisit website →
JarvislabsIn stock
INIndia1 config1xcollected 2026-10-11
On-Demand from $1.49
From
$1.49 / GPU / hr
On-DemandVisit website →
KoyebIn stock
FRFrance8 configs1x–8xcollected 2026-10-11
On-Demand from $1.60
From
$1.60 / GPU / hr
On-DemandVisit website →
RuncrateIn stock
1 config1xcollected 2026-10-11
On-Demand from $1.76
From
$1.76 / GPU / hr
On-DemandVisit website →
NorthflankIn stock
GBUK1 config1xcollected 2026-10-11
On-Demand from $1.76
From
$1.76 / GPU / hr
On-DemandVisit website →
RunPod Secure CloudIn stock
USUSA2 configs1xcollected 2026-10-11
On-Demand from $1.79
From
$1.79 / GPU / hr
On-DemandVisit website →
CivoIn stock
GBUK4 configs1x–8xcollected 2026-10-11
On-Demand from $1.79
From
$1.79 / GPU / hr
On-DemandVisit website →
TensorDockIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $1.80
From
$1.80 / GPU / hr
On-DemandVisit website →
DataCrunch (Verda)In stock
FIFinland1 config1xcollected 2026-10-11
On-Demand from $1.93
From
$1.93 / GPU / hr
On-DemandVisit website →E2E NetworksIn stock
INIndia2 configs1xcollected 2026-10-11
On-Demand from $1.98
From
$1.98 / GPU / hr
On-DemandVisit website →
CrusoeIn stock
USUSA2 configs1xcollected 2026-10-11
On-Demand from $2.00
From
$2.00 / GPU / hr
On-DemandVisit website →
CerebriumIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $2.10
From
$2.10 / GPU / hr
On-DemandVisit website →
ModalIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $2.50
From
$2.50 / GPU / hr
On-DemandVisit website →
CoreWeaveIn stock
USUSA1 config8xcollected 2026-10-11
On-Demand from $2.70
From
$2.70 / GPU / hr
On-DemandVisit website →
Lightning AIIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $2.71
From
$2.71 / GPU / hr
On-DemandVisit website →
LambdaIn stock
USUSA1 config8xcollected 2026-10-11
On-Demand from $2.79
From
$2.79 / GPU / hr
On-DemandVisit website →
OVHcloudIn stock
FRFrance4 configs1x–4xcollected 2026-10-11
On-Demand from $3.07
From
$3.07 / GPU / hr
On-DemandVisit website →
Amazon Web ServicesIn stock
USUSA1 config8xcollected 2026-10-11
On-Demand from $3.43
From
$3.43 / GPU / hr
On-Demand · us-east-1Visit website →
Microsoft AzureIn stock
USUSA5 configs1x–8xcollected 2026-10-11
On-Demand from $3.67Spot from $0.68
From
$3.67 / GPU / hr
On-Demand · eastusVisit website →
Oracle CloudIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $4.00
From
$4.00 / GPU / hr
On-Demand · us-ashburn-1Visit website →
BasetenIn stock
USUSA1 config1xcollected 2026-10-11
On-Demand from $4.00
From
$4.00 / GPU / hr
On-DemandVisit website →
ReplicateIn stock
USUSA4 configs1x–8xcollected 2026-10-11
On-Demand from $5.04
From
$5.04 / GPU / hr
On-DemandVisit website →Recent price changes
Every time a provider's median for the NVIDIA A100 80GB moved between two consecutive snapshots. A listing appearing or vanishing is not a price change and is not shown.
| When | Provider | From | To | Change |
|---|---|---|---|---|
| 2026-10-11 09:14 | Vast.ai | $0.68 | $0.93 | +36.6% |
| 2026-10-11 09:14 | DataCrunch (Verda) | $1.89 | $1.93 | +2.1% |
| 2026-10-11 09:14 | Runcrate | $1.79 | $1.76 | -1.7% |
| 2026-10-10 08:53 | Vast.ai | $0.87 | $0.68 | -21.7% |
| 2026-10-10 08:53 | GPU.ai | $1.12 | $1.15 | +3.1% |
| 2026-10-09 09:31 | RunPod Secure Cloud | $1.59 | $1.79 | +12.6% |
| 2026-10-09 09:31 | Vast.ai | $0.92 | $0.87 | -5.3% |
| 2026-10-09 09:31 | DataCrunch (Verda) | $1.85 | $1.89 | +2.2% |
| 2026-10-09 09:31 | Runcrate | $1.76 | $1.79 | +1.7% |
| 2026-10-08 09:24 | GPU.ai | $1.02 | $1.12 | +9.3% |
Spot, reserved and where it is cheapest
Spot median
$0.8851% off · 2 providers
Reserved median
$0.9845% off · 1 providers
Italy providers
$1.11median of 1
UAE providers
$1.16median of 1
By provider country: Italy $1.11 (1) · UAE $1.16 (1) · Canada $1.45 (1) · India $1.77 (2) · UK $1.77 (4) · Finland $1.93 (1) · USA $2.12 (20) · France $2.47 (2). Full breakdown →
What fits on a NVIDIA A100 80GB
Weights + KV cache at 8K context + runtime overhead, against 72 GB usable of the 80 GB on the card. Capacity follows total parameters, even for mixture-of-experts models — any expert may be needed next. Where one card is not enough, the table says how many are, and prices the set at today's median.
| Model | Params | 4-bit | 8-bit | 16-bit |
|---|---|---|---|---|
| 2.8T104B active | 22× GPU$39.38/hr | 42× GPU$75.18/hr | 78× GPU$139.62/hr | |
| 552B16B active | 5× GPU$8.95/hr | 9× GPU$16.11/hr | 16× GPU$28.64/hr | |
 Llama 4 Maverick | 400B17B active | 4× GPU$7.16/hr | 6× GPU$10.74/hr | 12× GPU$21.48/hr |
 gpt-oss-120b | 117B5.1B active | ✓ 67.9 GB~569 tok/s · $0.87/M tok | 2× GPU$3.58/hr | 4× GPU$7.16/hr |
| Llama 3.1 70B | 70B | ✓ 44.5 GB~41 tok/s · $12.13/M tok | 2× GPU$3.58/hr | 3× GPU$5.37/hr |
 Gemma 4 31B | 31B | ✓ 21.2 GB~94 tok/s · $5.29/M tok | ✓ 36.7 GB~50 tok/s · $9.94/M tok | ✓ 65.7 GB~26 tok/s · $19.12/M tok |
 Qwen3 Coder 30B-A3B | 30.5B3.3B active | ✓ 19 GB~879 tok/s · $0.57/M tok | ✓ 34.3 GB~465 tok/s · $1.07/M tok | ✓ 62.9 GB~247 tok/s · $2.01/M tok |
| Qwen3.8 27B | 27B | ✓ 18.7 GB~107 tok/s · $4.65/M tok | ✓ 32.2 GB~57 tok/s · $8.72/M tok | ✓ 57.6 GB~30 tok/s · $16.57/M tok |
| gpt-oss-20b | 21B3.6B active | ✓ 13.7 GB~806 tok/s · $0.62/M tok | ✓ 24.2 GB~426 tok/s · $1.17/M tok | ✓ 43.9 GB~227 tok/s · $2.19/M tok |
| Llama 3.1 8B | 8B | ✓ 6.9 GB~362 tok/s · $1.37/M tok | ✓ 10.9 GB~192 tok/s · $2.59/M tok | ✓ 18.4 GB~102 tok/s · $4.87/M tok |
tok/s is decode throughput from memory bandwidth, derated 20%, single stream — an estimate, not a benchmark. $/M tok is the hourly median divided by that rate: the one-user floor for a million generated tokens. Set cost uses today's median × GPUs required.
Buy or rent a A100 80GB?
Approx. list price
$15,000not tracked
Rent, per month
$1,289at median, 720 h
Break-even
8,380 h≈ 11.6 months
Rent for fewer than about 11.6 months of continuous use and renting wins on the card alone; past that, owning does — before the server, power, cooling and the person who keeps it running, which for a datacenter card are not small. The list price is a typical quoted figure and is not tracked by the index.
Specifications
Memory80 GB HBM2e
Memory bandwidth2,039 GB/s
ArchitectureAmpere
VendorNVIDIA
ClassDatacenter
Released2020-11
Bandwidth is listed because it governs generation speed: each token streams the active weights once, so tokens per second is bandwidth divided by bytes read. TFLOPs decide prompt processing and training, not how fast text appears.
Alternatives and comparisons
A100 80GB vs H100 SXMPrice, memory, what fits, cost per tokenA100 80GB vs A100 40GBPrice, memory, what fits, cost per tokenA100 80GB vs L40SPrice, memory, what fits, cost per tokenA100 80GB vs RTX PRO 6000 BlackwellPrice, memory, what fits, cost per tokenA100 80GB vs RTX 4090Price, memory, what fits, cost per tokenA100 80GB vs RTX 5090Price, memory, what fits, cost per token
NVIDIA H100 SXM80 GB · $3.50/hr median · 43 providersNVIDIA H100 PCIe80 GB · $2.29/hr median · 8 providersNVIDIA H100 NVL94 GB · $2.80/hr median · 6 providers
Questions
How much does it cost to rent a NVIDIA A100 80GB?
As of 2026-10-11, the median on-demand price across 35 providers is $1.79 per GPU-hour, about $1,289 a month at 720 hours. The cheapest we saw was $0.37 at Vast.ai.
Why do prices for the same GPU vary so much?
Hyperscalers bundle CPU, RAM and network into an instance price and charge for the ecosystem around it. Neoclouds sell the GPU more directly. Marketplaces are independent hosts competing on price, with the trade-offs of shared, third-party hardware. The per-GPU figure is the instance price divided by GPU count, which is the only way to compare an 8-GPU node with a single-card listing.
Is the median the price I will pay?
No — it is the middle of the market. Each provider's own median is taken first so a provider with many instance sizes counts once, then the median across providers. Spot and reserved rates are excluded from it; they set the floor shown separately. Use it to judge whether a quote is high or low, then verify with the provider.
Which models can a NVIDIA A100 80GB run?
With 80 GB per GPU, the table above shows which reference models fit on one card at 4, 8 and 16-bit, using the same memory arithmetic as the rest of Runyard: weights plus KV cache plus runtime overhead, with capacity following total parameters even for mixture-of-experts models. Where a model needs more than one GPU it says how many and what that set costs at today's median.
Own the hardware instead? Work out what your card holds with the same arithmetic.
Check what fits →