Pricing

The most cost-effective way to build and run AI applications, at scale, and in minutes with a cloud native platform.

Basic

$0
+ usage fee
/ month

Perfect for individuals and small teams to get started.

GET STARTED

Standard

$30
+ usage fee
/ month

Designed for collaborative teams and growing businesses.

SIGN IN TO UPGRADE

Enterprise

Custom

For organizations requiring high SLAs, performance, and compliance.

CONTACT US

Compare Plans

Compare the three plans to find the best fit for your needs.

Compute Usage

Pay exclusively for your actual compute usage, billed by the minute.

CPU
$0.000825 / min
NVIDIA-A10
$0.0202 / min
NVIDIA-A100
$0.0535 / min
NVIDIA-A6000
$0.0275 / min
NVIDIA-H100
$0.05 / min

Built-in Storage

Saving your models, data and logs just right in the platform, and pay only for what you use.

$0.153 / GB / month

Serverless Endpoints

Enhance your application with premium open-source models while leveraging Lepton's exceptional runtime performance and reliability.

ModelPrice
Llama3.2 1b$0.01 / million tokens
Llama3.2 3b$0.03 / million tokens
Llama3 8b$0.07 / million tokens
Llama3.1 8b$0.07 / million tokens
Llama2 13b$0.18 / million tokens
Llama3 70b$0.8 / million tokens
Llama3.1 70b$0.8 / million tokens
Llama3.3 70b$0.8 / million tokens

Frequently Ask Questions

Find answers to the most common questions about Lepton AI.

How are compute usages billed?
Can I cancel my subscription at any time?
What kind of support does Lepton offer?
Is there a limit for workspace members number?
Do I need to pay for serverless endpoints usage fee if I upgrade to the standard plan?
Start building with Lepton AI