Baseten
HomeChangelogStatusSupport
Search…
Welcome to Baseten!
Build your first app
Local development setup
Public beta
MODELS
Getting started: Models
Deploying models
Managing models
Interacting with models
Tracking model metadata
APPLICATIONS
Getting started: Applications
Views
Worklets
Files
Draft environment
Publishing your application
DATA
Getting started: Data
Baseten Postgres
Data connections
Query builder
Database access
SETTINGS
Account settings
API keys
Managing members
Self-hosting Baseten
APPENDIX
Glossary
Python client reference
CLI reference
API reference
Keyboard shortcuts
Python packages
Powered By GitBook
Getting started: Models
Deploy, manage, and invoke your models on robust production infrastructure.
A well-trained ML model feels like magic to use, but what good is that magic if it stays shut in your spellbook ... er, Jupyter notebook? Baseten isn't just for building full-stack applications, it's for building ML-powered full-stack applications, which begins with deploying powerful models on robust infrastructure and accessing them via an API.
Deploying and managing models
Models are trained using heterogeneous technologies and methods, so there are many ways to deploy a model to Baseten, including pre-trained models, models you've trained locally, and models from SageMaker. Once you have deployed your model, you can monitor its health, resources, and logs.
Previous
Public beta
Next - MODELS
Deploying models
Last modified 1mo ago
Copy link