vs

GPUhub vs Vast.ai

Two providers, priced against each other on the 3 GPUs both of them rent. Published per-GPU on-demand rates as each provider last stated them, with what their own pages say about billing, commitment and egress beside the numbers.

GPUhub is cheaper on 2 of the 3 cards both of them rent, by a median of 7%. Vast.ai is cheaper on the other 1.

The 3 cards both of them rent, cheapest published per-GPU on-demand rate each. Dearest first.

Per GPU per hour, on-demand list prices as each provider last published them. Rates exclude tax and vary by region; a cheaper rate is not the same machine, the same network or the same support.

Neither site tells you whether your data trains their models, an uptime SLA, whether you get credits when it breaks or how long you have to dispute a bill. We looked on both and record the absence rather than filling it in — if you need any of it, ask before you commit.

• GPU.ai vs Vast.ai20 shared GPUs
• RunPod Secure Cloud vs Vast.ai16 shared GPUs
• Massed Compute vs Vast.ai12 shared GPUs
• Vast.ai vs Verda10 shared GPUs
• Spheron vs Vast.ai8 shared GPUs
• Lambda vs Vast.ai7 shared GPUs
• Hyperstack vs Vast.ai6 shared GPUs
• AceCloud vs Vast.ai5 shared GPUs