Modal
AI infrastructure that developers love.
New York City, United States 11 GPU modelsAt a glance
prices checked 3d ago- Cheapest GPU
- $0.59/hr
- NVIDIA T4 16GB
- Fastest GPU
- $6.25/hr
- NVIDIA B200 SXM 180GB
- GPU models
- 11
- 11 configurations
- Certifications
- 2
- SOC 2 Type II, HIPAA
What's good about Modal
- Model Serverless
- Billing Per second
- GPUs offered H100, A100, A10G, B200
- Funding raised $466M+
- Founders Erik Bernhardsson, Akshat Bubna
- Named customers Runway, Suno, Quora, Physical Intelligence
Available GPUs
Cheapest first, per GPU per hour.
| GPU | Configuration | Memory | Price | Region | |
|---|---|---|---|---|---|
| NVIDIA T4 16GB Turing | 1 GPU | 16 GB | $0.59/hr | - | Rent |
| NVIDIA L4 24GB Ada Lovelace | 1 GPU | 24 GB | $0.80/hr | - | Rent |
| NVIDIA A10 24GB Ampere | 1 GPU | 24 GB | $1.10/hr | - | Rent |
| NVIDIA L40S 48GB Ada Lovelace | 1 GPU | 48 GB | $1.95/hr | - | Rent |
| NVIDIA A100 SXM4 40GB Ampere | 1 GPU | 40 GB | $2.10/hr | - | Rent |
| NVIDIA A100 SXM4 80GB Ampere | 1 GPU | 80 GB | $2.50/hr | - | Rent |
| NVIDIA RTX PRO 6000 Blackwell Workstation Edition Blackwell | 1 GPU | 96 GB | $3.03/hr | - | Rent |
| NVIDIA H100 SXM5 80GB Hopper | 1 GPU | 80 GB | $3.95/hr | - | Rent |
| NVIDIA H200 SXM 141GB Hopper | 1 GPU | 141 GB | $4.54/hr | - | Rent |
| NVIDIA B200 SXM 180GB Blackwell | 1 GPU | 180 GB | $6.25/hr | - | Rent |
Modal head to head
Compare any two
About Modal
Modal sells serverless GPU compute: developers write Python against its SDK and deploy to the cloud without provisioning machines, running inference, batch jobs, training and sandboxed environments on H100s, A100s, A10Gs and B200s across globally distributed regions. The distinction from an inference platform is that customers run arbitrary code with custom images and dependencies rather than calling fixed model endpoints. Billing is per second with no reserved capacity or commitments. Founded by Erik Bernhardsson and Akshat Bubna, the company reports over $466 million raised from investors including General Catalyst, Redpoint Ventures, Lux Capital, Amplify Partners and Creandum, and works from offices in New York, Stockholm and San Francisco. Named customers include Runway, Suno, Quora, Physical Intelligence, Chai Discovery and Reducto.
Pricing options
Services
- Serverless Model
- Deploy Python through the SDK; no machines to provision and no reserved capacity.
Compliance and certifications
Independently verified standards governing security, privacy and operations.
Ready to deploy on Modal?
Visit Modal