Start with the GPU memory your workload needs, then compare provider rates and minimum instance sizes. These starting prices are on-demand rates from the current catalog, not guaranteed capacity.
Compare H100 cloud rental rates across providers, including the reported memory configuration and minimum instance size. An eight-GPU rate per GPU is useful for comparison, but the full eight-GPU instance is what you pay for.
From $3.19/GPU·hr on-demand
1-GPU minimum · $3.19/instance·hr
Compare A100 40GB and 80GB rentals separately. The lower hourly price is only useful if your model, activations and batch size fit the memory available on each GPU.
From $1.00/GPU·hr on-demand
1-GPU minimum · $1.00/instance·hr
Compare listed H200 rental prices for memory-intensive AI workloads. Check the minimum GPU count alongside the headline per-GPU rate before planning a budget.
From $4.50/GPU·hr on-demand
1-GPU minimum · $4.50/instance·hr
Compare B200 GPU rentals with whole-instance costs, region and pricing type visible together. Newer accelerators should be evaluated on completed-work cost as well as hourly price.
From $6.69/GPU·hr on-demand
8-GPU minimum · $53.52/instance·hr
Compare L40S cloud prices for inference, rendering and workloads that fit its available memory. Provider host resources can differ even when the GPU model is the same.
From $1.09/GPU·hr on-demand
1-GPU minimum · $1.09/instance·hr
Compare listed RTX 4090 rentals for development and workloads that fit within the reported GPU memory. The price table covers our tracked providers, not every marketplace listing.
From $0.390/GPU·hr on-demand
1-GPU minimum · $0.390/instance·hr