This past month, we explored the possibilities of K8s and JupyterLab in ML, announced our first live webinar featuring Anna Veronika Dorogush, founder and CEO of Recraft, and updated the documentation with handy tutorials on testing GPU clusters.
How to optimize K8s for ML?
Follow the steps of Boris, our Solutions Architect, as he sets up a GPU-enabled Kubernetes cluster with Terraform. This is perfect for DevOps engineers, data scientists, and anyone who’s into cloud computing or works with ML loads.
Live webinar: introduction to the platform
This will be our first live webinar, where Anna Veronika Dorogush, founder and CEO of Recraft, will share her company’s experience with heavy model training for geverative AI on Nebius infrastructure. Also, Andrew and Levon from Nebius will introduce the platform’s capabilities. Join us on February 29th.
Creating an InfiniBand GPU cluster
If you use multiple Compute Cloud VMs with GPUs, you can group them into an InfiniBand GPU cluster for faster model training. To create it, refer to this guide. A cluster may be created via console or CLI, either when creating the first VM in it or afterwards.
Which AI conferences to attend in 2024?
Recently, we had an insightful discussion with our machine learning experts at Nebius AI. From their wealth of hands-on experience in developing LLMs, generative networks, and other cutting-edge tech, they’ve curated this list of must-attend conferences.