Market Trends
5,326 instances across 54 providers
$0.80
/hr
Cheapest H100
$0.47
/hr
Cheapest H200
$0.08
/hr
Cheapest A100
$6.39
2,154 spot
Avg Price
70B model needs ≥140GB VRAM. Ranked by $/1M tokens.
Verda
$0.80/hr
Nebius
$1.25/hr
GCP
$1.35/hr
Vast.ai
$1.47/hr
Shadeform
$1.66/hr
E2E Networks
$1.69/hr
RunPod
$1.75/hr
Cudo Compute
$1.87/hr
Hyperstack
$1.90/hr
AWS
$1.97/hr
Voltage Park
$1.99/hr
FluidStack
$2.09/hr
Hetzner
$2.49/hr
Sesterce
$2.49/hr
Thunder Compute
$2.49/hr
Novita
$2.49/hr
Together AI
$2.49/hr
TensorWave
$2.49/hr
Oblivus
$2.49/hr
Salad Cloud
$2.50/hr
Massed Compute
$2.69/hr
TensorDock
$2.69/hr
Jarvis Labs
$2.69/hr
Cerebrium
$2.95/hr
Civo
$2.99/hr
LeaderGPU
$2.99/hr
OVHcloud
$2.99/hr
Scaleway
$3.20/hr
Lambda Labs
$3.29/hr
Lightning.ai
$3.50/hr
Cirrascale
$3.50/hr
Fal.ai
$3.60/hr
Crusoe
$3.90/hr
Modal
$4.30/hr
Cherry Servers
$4.49/hr
CoreWeave
$49.24/hr
≤24 GB
$0.0133/hr
V100
Vast.ai · 3386 options
25–48 GB
$0.0800/hr
A100
Vast.ai · 680 options
49–80 GB
$0.452/hr
A100
Verda · 607 options
80 GB+
$0.152/hr
RTXPRO6000
GCP · 653 options