Skip to main content

Serverless Inference Done Right

Running big custom generative AI models is hard and expensive. We offer a better way. Dat1 is a machine learning model hosting that saves you time, money and respects your privacy. 

Want to talk to us? Book a meeting

 

Join our Beta Now

Our beta is live, and we're actively onboarding new customers! Sign up now to get started immediately.

Coder

Easy Scale & Lower Costs

With dat1, you don't need to manage hardware and think about scaling. We manage everything in the background and reduce operational and hardware costs by efficiently sharing the GPUs between our customers.

You don't need to worry about high fixed costs typically associated with GPU inference, we charge per second ($0.005 per second for an NVIDIA A100and we only charge for the time your model runs inference—no costs for idle time or timeouts.

Launch

Low Cold Start

We worked hard to reduce cold starts which is an important point in all serverless platforms. With dat1, you can expect a 15-20-second cold start for a 70Gb model. We will also work closely with you to optimize it.

 
Launch

Low Cold Start

We worked hard to reduce cold starts which is an important point in all serverless platforms. With dat1, you can expect a 15-20-second cold start for a 70Gb model. We will also work closely with you to optimize it.

 
To-do-list

Privacy & Security

We value our customers' privacy. We are fully GDPR and CCPA-compliant and follow top cybersecurity best practices.

20181218-Nvidia-Inception

Dat1.co is part of the Nvidia Inception program