From $0.01 usage-based · Free trial🇺🇸
US inference platform for open, fine-tuned and custom models on autoscaling GPUs, with Model APIs, training and self-hosting.
Modal AI features and pricing (October 2026)
Serverless Python platform for AI inference, training, batch jobs and sandboxes with per-second GPU billing.
Usage-basedFree plan
from $250 per month
Pricing details
Starter with no base fee and $30 of compute per month, Team at $250 per month with $100 of compute, Enterprise by quote; GPUs billed per second from $0.000164 per second (NVIDIA T4), as of September 2026.
What is Modal?
Modal is a serverless compute platform where developers run AI inference, training, batch jobs and sandboxes from Python code, with GPUs billed per second and a free Starter plan.
Modal built its own file system, container runtime and scheduler to start containers in under a second and scale GPU workloads from zero. Developers define functions and their hardware in Python, and Modal handles containers, scaling and logs. The company is headquartered in New York. Plans range from Starter with three seats to Team and Enterprise, where SSO and a HIPAA business associate agreement are available. Region selection is available in all plans at 1.15 to 1.75 times the base price.
Modal AI features and use cases
- Deploys models as autoscaling endpointsLanguage, audio and image models run as functions that scale with requests and stop when idle. Small teams avoid paying for GPUs that sit unused between requests.
- Runs evaluations and dataset jobsBatch processing spreads evaluations or synthetic data generation across many containers at once. Jobs that take hours on one machine finish faster and cost only while they run.
- Executes code from AI agents safelySandboxes run untrusted or AI-generated code in isolated, short-lived environments. Teams building coding agents keep that code away from their own systems.
What sets Modal apart
- Infrastructure is defined in Python code instead of YAML or container configuration files.
- Free Starter plan with $30 of monthly compute; SSO and HIPAA only on Enterprise.
Where Modal is strongest
United States Available internationally
Modal is headquartered in New York, with further offices in San Francisco, Stockholm and London.
Questions to ask before you buy Modal
Select the region for functions and storage explicitly and budget for the regional surcharge. Inputs and outputs are kept for up to seven days, and logs for one day on Starter, 30 days on Team and a configurable period on Enterprise. A data processing agreement and subprocessor list are not linked publicly, so request them before processing personal data.
Modal reviews
No reviews yet. Be the first to share how your team uses it.
Every review shows its verification level. We read each one before it goes live. How we check reviews
Questions about Modal
Is there a free plan?
Yes. Starter has no base fee, includes $30 of compute per month and allows three seats and ten concurrent GPUs.
How are GPUs billed?
Per second of use. The cheapest listed GPU in September 2026 was an NVIDIA T4 at $0.000164 per second.
Which certifications does Modal have?
Modal has completed a SOC 2 Type II audit. HIPAA use requires a business associate agreement on the Enterprise plan.
Modal update historyno updates yet
No changes recorded for Modal yet.
Changes the vendor made and confirmed user reports, each checked against the vendor's own pages. We review every tool monthly.
Alternatives to Modal
Tools in the same category (Serverless inference) with the most overlapping markets.
From $0.02 usage-based🇺🇸
Inference API for more than 100 open and commercial models billed per token, plus GPU instances and dedicated clusters.
Usage-based🇺🇸
GPU cloud with low-cost hourly Pods, serverless endpoints that scale to zero and clusters for developers and small teams.
From $1.50 usage-based🇺🇸
US AI cloud on its own data centers, with NVIDIA and AMD GPUs, managed Kubernetes and Slurm, and managed inference.