Compare GPU cloud pricing across major providers. Estimate hourly, daily, monthly, and annual compute costs for B200, H200, H100, A100, L40S, and RTX 4090 GPUs with on-demand and reserved pricing.
Hourly Cost
$10.60
$10.60/GPU
Daily Cost
$84.80
8 hrs
Monthly Cost
$1,865.60
22 days
Annual Cost
$22,387.20
12 months
Google Cloud
269
GPU-hours
Lambda
264
GPU-hours
Jarvislabs
263
GPU-hours
CoreWeave
257
GPU-hours
AWS
94
GPU-hours
Azure
73
GPU-hours
GPU cloud pricing varies dramatically between providers and GPU models. Major cloud providers like AWS and Azure offer enterprise reliability and ecosystem integration, while specialized providers like Lambda, CoreWeave, and RunPod focus on GPU workloads with competitive pricing. The best choice depends on your scale, reliability requirements, and budget.
GPU cloud pricing changes frequently. Prices shown are approximate on-demand rates as of April 2026. Actual costs may vary based on region, availability, and specific instance configurations.
For frontier-scale training, the Blackwell B200 is now the top tier, with the H200 and H100 remaining the workhorses for most transformer workloads. The A100 is still a solid choice for smaller models and fine-tuning. For inference and smaller workloads, the L40S and RTX 4090 offer excellent price-to-performance ratios.
On-demand pricing lets you pay by the hour with no commitment — ideal for experimentation and burst workloads. Reserved instances (1-year or 3-year) offer 30-50% discounts but require upfront commitment. Choose reserved pricing when you have predictable, sustained compute needs.
Spot instances can save 60-90% over on-demand but can be interrupted with little notice. They work well for fault-tolerant workloads like hyperparameter sweeps, batch inference, and training with checkpointing. This calculator focuses on on-demand and reserved pricing for predictable budgeting.
LLM API Cost Calculator
Calculate and compare the cost of using large language model APIs from OpenAI, Anthropic, Google, Mistral, xAI, and DeepSeek. Estimate daily, monthly, and annual spend based on token usage.
AI Scaling Cost Estimator
Project how AI generation costs scale as your volume grows. Estimate monthly costs at different volume levels and plan your budget for scaling AI content production.