With dat1, you only pay for the time your model actively processes requests—never for startup time, idle periods, or timeouts.
$0.0014/second
contact us for other GPU models
Estimate your monthly cost based on usage.
Typical image generation with SDXL takes about 5 seconds on an NVIDIA H100 GPU.
Price guaranteed. What you see in this calculator is what you'll pay. No hidden fees.
