Cloud GPU pricing

Live snapshot across 11 providers, sorted by VRAM tier and cheapest hourly. Refreshed periodically.

16 GB VRAM

Tiny inference, T4 / L4 / RTX A4000

Cheapest: $0.59/hr
ProviderGPU$/hrNotesOpen
ModalT4$0.59Serverless · Scale to zeroRent →
PaperspaceRTX A4000$0.76Gradient notebooksRent →

24 GB VRAM

RTX 3090 / 4090 — sweet spot for 8–13 B models

Cheapest: $0.10/hr
ProviderGPU$/hrNotesOpen
SaladRTX 3090$0.10Distributed · Best $/hrRent →
Vast.aiRTX 3090$0.15Cheapest 24 GB · Spot pricingRent →
SaladRTX 4090$0.20Distributed computeRent →
RunPodRTX 3090$0.22Community cloud · Cheapest 24 GBRent →
Vast.aiRTX 4090$0.25Variable availabilityRent →
CudoRTX 4090$0.30Renewable-only DCsRent →
TensorDockRTX 4090$0.34Marketplace · Per-minute billingRent →
RunPodRTX 4090$0.44On-demand · Serverless availableRent →
ModalL4$0.80Serverless · Per-second billingRent →

40 GB VRAM

A100 40 GB — fits 30 B at Q4

Cheapest: $1.10/hr
ProviderGPU$/hrNotesOpen
LambdaA100 40GB$1.101-Click Cluster · Pre-installed CUDARent →

48 GB VRAM

A6000 / L40S — the 70 B Q4 sweet spot

Cheapest: $0.31/hr
ProviderGPU$/hrNotesOpen
Massed ComputeRTX A6000$0.31LLM templates · Cheap A6000Rent →
HyperstackRTX A6000$0.50EU regions · Reserved discountsRent →
RunPodRTX A6000$0.7648GB VRAM · Great for 70BRent →
RunPodL40S$0.99Ada Lovelace · Inference-tunedRent →
CrusoeL40S$1.45Flared-gas poweredRent →

80 GB VRAM

A100 80 GB / H100 — production / fine-tuning

Cheapest: $0.90/hr
ProviderGPU$/hrNotesOpen
Vast.aiA100 80GB$0.90Cheapest A100Rent →
TensorDockA100 80GB$1.20Marketplace pricingRent →
HyperstackA100 80GB$1.35EU bare-metalRent →
RunPodA100 80GB$1.6480GB · Top performanceRent →
LambdaA100 80GB$1.79No queue (reserved)Rent →
Massed ComputeH100$1.95Reserved discountsRent →
Vast.aiH100 SXM$2.10Cheapest H100Rent →
HyperstackH100$2.40EU H100Rent →
CudoH100 PCIe$2.45Renewable energyRent →
CrusoeH100$2.65Reserved capacityRent →
LambdaH100 SXM$2.99NVLinkRent →
PaperspaceA100 80GB$3.18Persistent storageRent →
RunPodH100 SXM$3.29HBM3 · Fastest inferenceRent →
ModalA100 80GB$4.10Per-second · Auto-scalingRent →
ModalH100$7.50Per-secondRent →

141 GB VRAM

H200 — biggest single-card

Cheapest: $4.49/hr
ProviderGPU$/hrNotesOpen
RunPodH200$4.49141 GB HBM3eRent →

Pay-per-token alternative

Skip the rental entirely — these providers run open-source models for you and bill by token.

Don’t need cloud? Check if your local hardware can run it →