Skip to main content

Weights & Biases Documentation home page

Log in
Sign Up
Sign Up

W&B Launch

Tutorial: W&B Launch basics

Launch terms and concepts

Set up Launch

Set up Launch
Configure launch queue
Set up launch agent

Create and deploy jobs

Create and deploy jobs
Create a launch job
Add job to queue
Manage job inputs
View launch jobs

Advanced

Create sweeps with W&B Launch
Monitor launch queue
Launch integration guides

Launch FAQ

Launch FAQ
Are there best practices for using Launch effectively?
I do not want W&B to build a container for me, can I still use Launch?
I do not like clicking- can I use Launch without going through the UI?
How do I control who can push to a queue?
When multiple jobs in a Docker queue download the same artifact, is any caching used, or is it re-downloaded every run?
Can I specify a Dockerfile and let W&B build a Docker image for me?
Can Launch automatically provision (and spin down) compute resources for me in the target environment?
How does W&B Launch build images?
Is `wandb launch -d` or `wandb job create image` uploading a whole docker artifact and not pulling from a registry?
Does Launch support parallelization? How can I limit the resources consumed by a job?
How do I make W&B Launch work with Tensorflow on GPU?
How do I fix a "permission denied" error in Launch?
What permissions does the agent require in Kubernetes?
What requirements does the accelerator base image have?
How can admins restrict which users have modify access?
Can you specify secrets for jobs/automations? For instance, an API key which you do not wish to be directly visible to users?

Launch Library API Reference

launch api
launch_add
LaunchAgent

W&B Launch

Scale and manage ML workloads with W&B Launch by configuring jobs, queues, and agents on your compute infrastructure.

Was this page helpful?

Suggest edits Raise issue

Tutorial: W&B Launch basics

⌘I

github discord x youtube linkedin

Assistant

Responses are generated using AI and may contain mistakes.