Effortlessly deploy machine learning models and applications
Export your models or applications in a Docker image, and AI Deploy will take care of the rest.
High-Availability, auto scaling with CPU and GPU, authentication, monitoring and more!
What is AI Deploy ?
Deploying machine learning models or AI Applications can be tough. You need constantly to think about compute resources, securization, availability for your customers and more. It can be a real challenge.
AI Deploy is here to facilitate this. Deploy your Docker images effortlessly and without Kubernetes expertise using AI Deploy. Carry out requests via API for your models and via the web interface for your production applications, while we manage the infrastructure and security of the environments.
If you need some ideas, we also provide a portfolio of Docker images to start with some use cases !
AI Deploy, part of our AI Solutions
When building AI models or applications, you will need more than one tools. From storage to deployment, OVHcloud provides the full toolbox for you !
Each AI product is managed, meaning that you can focus on you code, and we take care of the infrastructure and securization.
How does it work ?
AI Deploy just require your Docker Image. Put your AI model or application in a Dockerfile, Build your Docker image and push it in a repository.
Once done, simply deploy your app with our control panel, CLI or API.
Try it, it's 100% free!
Our goal is to build good products, fitting your needs. We provide here a free beta to gather as much feedback as we can.
All you need is an OVHcloud account then create a Public Cloud project.
In few minutes, you can deploy your apps with free CPU and GPU (NVIDIA V100s).
Demo : deploy your apps easily
AI Deploy can be used through OVHcloud control panel web interface, but we also provide API and CLI to automate your deployments.
If will fit perfectly with orchestration tools such as Apache Airflow, Prefect or Dagster!
Read our official documentation for more details about AI Deploy or watch our 2 minutes video:
YouTube conditions the playback of its videos on the deposit of tracers in order to offer you targeted advertising based on your browsing.
In order to watch the video, you need to accept the Sharing cookies on third-party platforms privacy category in our Privacy Center. You have the option of withdrawing your consent at any time.
For more information,visit the YouTube cookies policy and the OVHcloud cookies policy .
How do I start ?
All you need is an OVHcloud account. Create one (top menu) for free, then connect to our control panel, and go into Public Cloud.
Public Cloud works with projects. If you are new to OVHcloud, you can enter the voucher FREETRIAL to benefit from $200.
To avoid frauders, a payment methode may be required, even if AI Deploy is 100% free during Beta phase
What is the cost of AI Deploy ? Is it free even for GPU ?
AI Deploy is 100% free during BETA. You will pay nothing for CPU and GPU (NVIDIA V100s). If you need to use other products such as Object Storage or Docker registries, regular prices will apply.
Once the Beta period will be over, we will communicate widely to avoid any payment issue. Expected prices are the same as AI Notebook or AI Training.
How many apps can I deploy ?
It will depends of you personal quotas and the available stock (limited stock for this BETA).
Usually, a new user with standard quotas will be able to deploys 1 app with GPU and a dozen apps with CPU.
What can I deploy exactly?
You can deploy pretty much anything, as long as is can be run inside a Docker Image.
Main limitations are :
- your need to create a user and a folder called "workspace" inside your Dockerfile (read our official documentation about it)
- Dockerfile have to be build on linux/AMD64 architecture
- Docker compose is not compatible
- we only provide HTTP endpoints (so, not compatible for streaming)
Where can I find help and documentation ? How to share feedback ?
You can find our official documentation here : https://docs.ovh.com/gb/en/publiccloud/ai/
And don't hesitate to join our official Discord Community on https://discord.gg/ovhcloud