Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.nebius.com/llms.txt

Use this file to discover all available pages before exploring further.

Nebius AI Cloud offers AI services for containerized inference and training, experiment tracking, integrations with third-party software and orchestration with external tools. Different services can help you achieve different goals:
  • Use Serverless AI to run containerized AI workloads either as endpoints that return results immediately, or as background jobs that run in the background and stop when the task is complete.
  • Use Managed Service for MLflow to manage machine learning experiments by tracking runs, organizing models into versions, comparing metrics and deploying customized models.
  • Use Applications in Nebius AI Cloud to work with third-party software and services that help streamline AI and ML workflows.
  • Use third-party integrations to run and orchestrate AI workloads in the Nebius AI Cloud infrastructure.

Getting started with endpoints in Serverless AI

Deploy an interactive endpoint that listens for requests and returns results immediately

Getting started with jobs in Serverless AI

Run a non-interactive workload in the background and stop it automatically after completion

Getting started with MLflow

Create your first MLflow cluster and run an experiment in it

Deploying applications

Browse the marketplace and choose an application to deploy

Third-party integrations: SkyPilot

Run, manage and scale AI workloads on Nebius AI Cloud by using SkyPilot

Third-party integrations: Run:ai

Optimize your GPU resources for ML/AI workloads by using Run:ai and Managed Service for Kubernetes®

​
Tutorials and examples

Follow our Serverless AI tutorials for step-by-step instructions on setting up specific AI workloads, or explore tutorials that combine multiple Nebius AI Cloud services to build complete solutions.

Deploying a large language model and chatting with it

Create your first Serverless AI endpoint with an open-source large language model and send a chat request

Fine-tuning a large language model

Fine-tune a large language model by using a Serverless AI job with Axolotl

Speech synthesis

Convert text to speech by creating a model, running a fine-tuning job and deploying an endpoint

Building solutions with AI services

Explore tutorials that show you how to combine Nebius AI Cloud services to build complete solutions

Serverless cookbook

Explore runnable examples of Serverless AI workloads on GitHub