Modal

Serverless GPUs defined entirely in Python, no Dockerfiles or YAML. Per-second billing, scale-to-zero and instant fan-out make it ideal for batch jobs, inference APIs and pipelines.

Open Modal pricing ↗ ← All providers

On-demand GPUs

GPUVRAM$/hrType
H10080 GB$3.95Serverless
A100 80GB80 GB$2.50Serverless
A100 40GB40 GB$2.10Serverless
L40S48 GB$1.95Serverless
A10G24 GB$1.10Serverless

On-demand prices are indicative snapshots for orientation only. Spot/community and committed-use rates are often 30 to 70% lower. Always confirm on the provider's pricing page before committing.